diff --git a/cmake/taosadapter_CMakeLists.txt.in b/cmake/taosadapter_CMakeLists.txt.in index ed8216be9127b8c5164c977420d3ccd8f6521d31..f182beed33c76200649f93d96b68c153ec452b9a 100644 --- a/cmake/taosadapter_CMakeLists.txt.in +++ b/cmake/taosadapter_CMakeLists.txt.in @@ -2,7 +2,7 @@ # taosadapter ExternalProject_Add(taosadapter GIT_REPOSITORY https://github.com/taosdata/taosadapter.git - GIT_TAG 3d21433 + GIT_TAG abed566 SOURCE_DIR "${TD_SOURCE_DIR}/tools/taosadapter" BINARY_DIR "" #BUILD_IN_SOURCE TRUE diff --git a/contrib/CMakeLists.txt b/contrib/CMakeLists.txt index 494d1577fc7a170c4671e0dea423253d68fd06e1..2dc7622f4601ad8b1fb5c578c6aa8b5f0df02daf 100644 --- a/contrib/CMakeLists.txt +++ b/contrib/CMakeLists.txt @@ -331,9 +331,11 @@ endif(${BUILD_WITH_TRAFT}) # LIBUV if(${BUILD_WITH_UV}) - if (NOT ${CMAKE_SYSTEM_NAME} MATCHES "Windows") - MESSAGE("Windows need set no-sign-compare") - add_compile_options(-Wno-sign-compare) + if (TD_WINDOWS) + # There is no GetHostNameW function on win7. + file(READ "libuv/src/win/util.c" LIBUV_WIN_UTIL_CONTENT) + string(REPLACE "if (GetHostNameW(buf, UV_MAXHOSTNAMESIZE" "DWORD nSize = UV_MAXHOSTNAMESIZE;\n if (GetComputerNameW(buf, &nSize" LIBUV_WIN_UTIL_CONTENT "${LIBUV_WIN_UTIL_CONTENT}") + file(WRITE "libuv/src/win/util.c" "${LIBUV_WIN_UTIL_CONTENT}") endif () add_subdirectory(libuv EXCLUDE_FROM_ALL) endif(${BUILD_WITH_UV}) diff --git a/docs/en/01-index.md b/docs/en/01-index.md index f5b7f3e0f61507efbb09506b48548c12317e700b..4c2f9b02b90b5bedce21e27026f4fde683881837 100644 --- a/docs/en/01-index.md +++ b/docs/en/01-index.md @@ -4,7 +4,8 @@ sidebar_label: Documentation Home slug: / --- -TDengine is a [high-performance](https://tdengine.com/fast), [scalable](https://tdengine.com/scalable) time series database with [SQL support](https://tdengine.com/sql-support). This document is the TDengine user manual. It introduces the basic, as well as novel concepts, in TDengine, and also talks in detail about installation, features, SQL, APIs, operation, maintenance, kernel design and other topics. It’s written mainly for architects, developers and system administrators. + +TDengine is an open source, cloud native time-series database optimized for Internet of Things (IoT), Connected Cars, and Industrial IoT. It enables efficient, real-time data ingestion, processing, and monitoring of TB and even PB scale data per day, generated by billions of sensors and data collectors. This document is the TDengine user manual. It introduces the basic, as well as novel concepts, in TDengine, and also talks in detail about installation, features, SQL, APIs, operation, maintenance, kernel design and other topics. It’s written mainly for architects, developers and system administrators. To get an overview of TDengine, such as a feature list, benchmarks, and competitive advantages, please browse through the [Introduction](./intro) section. diff --git a/docs/en/02-intro/index.md b/docs/en/02-intro/index.md index f6766f910f4d7560b782bf02ffa97922523e6167..23a79aa2299b092dbeb797696ef678c02517b264 100644 --- a/docs/en/02-intro/index.md +++ b/docs/en/02-intro/index.md @@ -3,7 +3,7 @@ title: Introduction toc_max_heading_level: 2 --- -TDengine is a high-performance, scalable time-series database with SQL support. Its code, including its cluster feature is open source under GNU AGPL v3.0. Besides the database engine, it provides [caching](/develop/cache), [stream processing](/develop/continuous-query), [data subscription](/develop/subscribe) and other functionalities to reduce the complexity and cost of development and operation. +TDengine is an open source, high-performance, cloud native time-series database optimized for Internet of Things (IoT), Connected Cars, and Industrial IoT. Its code, including its cluster feature is open source under GNU AGPL v3.0. Besides the database engine, it provides [caching](/develop/cache), [stream processing](/develop/continuous-query), [data subscription](/develop/subscribe) and other functionalities to reduce the system complexity and cost of development and operation. This section introduces the major features, competitive advantages, typical use-cases and benchmarks to help you get a high level overview of TDengine. @@ -31,25 +31,21 @@ For more details on features, please read through the entire documentation. ## Competitive Advantages -Time-series data is structured, not transactional, and is rarely deleted or updated. TDengine makes full use of [these characteristics of time series data](https://tdengine.com/2019/07/09/86.html) to build its own innovative storage engine and computing engine to differentiate itself from other time series databases, with the following advantages. +By making full use of [characteristics of time series data](https://tdengine.com/tsdb/characteristics-of-time-series-data/), TDengine differentiates itself from other time series databases, with the following advantages. -- **[High Performance](https://tdengine.com/fast)**: With an innovatively designed and purpose-built storage engine, TDengine outperforms other time series databases in data ingestion and querying while significantly reducing storage costs and compute costs. +- **High-Performance**: TDengine is the only time-series database to solve the high cardinality issue to support billions of data collection points while out performing other time-series databases for data ingestion, querying and data compression. -- **[Scalable](https://tdengine.com/scalable)**: TDengine provides out-of-box scalability and high-availability through its native distributed design. Nodes can be added through simple configuration to achieve greater data processing power. In addition, this feature is open source. +- **Simplified Solution**: Through built-in caching, stream processing and data subscription features, TDengine provides a simplified solution for time-series data processing. It reduces system design complexity and operation costs significantly. -- **[SQL Support](https://tdengine.com/sql-support)**: TDengine uses SQL as the query language, thereby reducing learning and migration costs, while adding SQL extensions to better handle time-series. Keeping NoSQL developers in mind, TDengine also supports convenient and flexible, schemaless data ingestion. +- **Cloud Native**: Through native distributed design, sharding and partitioning, separation of compute and storage, RAFT, support for kubernetes deployment and full observability, TDengine is a cloud native Time-Series Database and can be deployed on public, private or hybrid clouds. -- **All in One**: TDengine has built-in caching, stream processing and data subscription functions. It is no longer necessary to integrate Kafka/Redis/HBase/Spark or other software in some scenarios. It makes the system architecture much simpler, cost-effective and easier to maintain. +- **Ease of Use**: For administrators, TDengine significantly reduces the effort to deploy and maintain. For developers, it provides a simple interface, simplified solution and seamless integrations for third party tools. For data users, it gives easy data access. -- **Seamless Integration**: Without a single line of code, TDengine provide seamless, configurable integration with third-party tools such as Telegraf, Grafana, EMQX, Prometheus, StatsD, collectd, etc. More third-party tools are being integrated. +- **Easy Data Analytics**: Through super tables, storage and compute separation, data partitioning by time interval, pre-computation and other means, TDengine makes it easy to explore, format, and get access to data in a highly efficient way. -- **Zero Management**: Installation and cluster setup can be done in seconds. Data partitioning and sharding are executed automatically. TDengine’s running status can be monitored via Grafana or other DevOps tools. +- **Open Source**: TDengine’s core modules, including cluster feature, are all available under open source licenses. It has gathered 18.8k stars on GitHub. There is an active developer community, and over 139k running instances worldwide. -- **Zero Learning Costs**: With SQL as the query language and support for ubiquitous tools like Python, Java, C/C++, Go, Rust, and Node.js connectors, and a REST API, there are zero learning costs. - -- **Interactive Console**: TDengine provides convenient console access to the database, through a CLI, to run ad hoc queries, maintain the database, or manage the cluster, without any programming. - -With TDengine, the total cost of ownership of your time-series data platform can be greatly reduced. 1: With its superior performance, the computing and storage resources are reduced significantly 2: With SQL support, it can be seamlessly integrated with many third party tools, and learning costs/migration costs are reduced significantly 3: With its simple architecture and zero management, the operation and maintenance costs are reduced. +With TDengine, the total cost of ownership of your time-series data platform can be greatly reduced. 1: With its superior performance, the computing and storage resources are reduced significantly；2: With SQL support, it can be seamlessly integrated with many third party tools, and learning costs/migration costs are reduced significantly；3: With its simplified solution and nearly zero management, the operation and maintenance costs are reduced significantly. ## Technical Ecosystem This is how TDengine would be situated, in a typical time-series data processing platform: diff --git a/docs/en/04-concept/index.md b/docs/en/04-concept/index.md index 850f705146c4829db579f14be1a686ef9052f678..44dcad82fc8e77bdb63ed3f8d5a36b9542c72aea 100644 --- a/docs/en/04-concept/index.md +++ b/docs/en/04-concept/index.md @@ -2,7 +2,7 @@ title: Concepts --- -In order to explain the basic concepts and provide some sample code, the TDengine documentation smart meters as a typical time series use case. We assume the following: 1. Each smart meter collects three metrics i.e. current, voltage, and phase 2. There are multiple smart meters, and 3. Each meter has static attributes like location and group ID. Based on this, collected data will look similar to the following table: +In order to explain the basic concepts and provide some sample code, the TDengine documentation smart meters as a typical time series use case. We assume the following: 1. Each smart meter collects three metrics i.e. current, voltage, and phase; 2. There are multiple smart meters; 3. Each meter has static attributes like location and group ID. Based on this, collected data will look similar to the following table:
@@ -116,7 +116,7 @@ Data Collection Point (DCP) refers to hardware or software that collects metrics ## Table -Since time-series data is most likely to be structured data, TDengine adopts the traditional relational database model to process them with a short learning curve. You need to create a database, create tables, then insert data points and execute queries to explore the data. +Since time-series data is most likely to be structured data, TDengine adopts the traditional relational database model to process them with a short learning curve. You need to create a database, create tables, then insert data points and execute queries to explore the data. To make full use of time-series data characteristics, TDengine adopts a strategy of "**One Table for One Data Collection Point**". TDengine requires the user to create a table for each data collection point (DCP) to store collected time-series data. For example, if there are over 10 million smart meters, it means 10 million tables should be created. For the table above, 4 tables should be created for devices D1001, D1002, D1003, and D1004 to store the data collected. This design has several benefits: @@ -125,25 +125,28 @@ To make full use of time-series data characteristics, TDengine adopts a strategy 3. The metric data from a DCP is continuously stored, block by block. If you read data for a period of time, it can greatly reduce random read operations and improve read and query performance by orders of magnitude. 4. Inside a data block for a DCP, columnar storage is used, and different compression algorithms are used for different data types. Metrics generally don't vary as significantly between themselves over a time range as compared to other metrics, which allows for a higher compression rate. -If the metric data of multiple DCPs are traditionally written into a single table, due to uncontrollable network delays, the timing of the data from different DCPs arriving at the server cannot be guaranteed, write operations must be protected by locks, and metric data from one DCP cannot be guaranteed to be continuously stored together. **One table for one data collection point can ensure the best performance of insert and query of a single data collection point to the greatest possible extent.** +If the metric data of multiple DCPs are traditionally written into a single table, due to uncontrollable network delays, the timing of the data from different DCPs arriving at the server cannot be guaranteed, write operations must be protected by locks, and metric data from one DCP cannot be guaranteed to be continuously stored together. ** One table for one data collection point can ensure the best performance of insert and query of a single data collection point to the greatest possible extent.** TDengine suggests using DCP ID as the table name (like D1001 in the above table). Each DCP may collect one or multiple metrics (like the current, voltage, phase as above). Each metric has a corresponding column in the table. The data type for a column can be int, float, string and others. In addition, the first column in the table must be a timestamp. TDengine uses the time stamp as the index, and won’t build the index on any metrics stored. Column wise storage is used. +Complex devices, such as connected cars, may have multiple DCPs. In this case, multiple tables are created for a single device, one table per DCP. + ## Super Table (STable) The design of one table for one data collection point will require a huge number of tables, which is difficult to manage. Furthermore, applications often need to take aggregation operations among DCPs, thus aggregation operations will become complicated. To support aggregation over multiple tables efficiently, the STable(Super Table) concept is introduced by TDengine. STable is a template for a type of data collection point. A STable contains a set of data collection points (tables) that have the same schema or data structure, but with different static attributes (tags). To describe a STable, in addition to defining the table structure of the metrics, it is also necessary to define the schema of its tags. The data type of tags can be int, float, string, and there can be multiple tags, which can be added, deleted, or modified afterward. If the whole system has N different types of data collection points, N STables need to be established. -In the design of TDengine, **a table is used to represent a specific data collection point, and STable is used to represent a set of data collection points of the same type**. +In the design of TDengine, **a table is used to represent a specific data collection point, and STable is used to represent a set of data collection points of the same type**. ## Subtable -When creating a table for a specific data collection point, the user can use a STable as a template and specify the tag values of this specific DCP to create it. **The table created by using a STable as the template is called subtable** in TDengine. The difference between regular table and subtable is: +When creating a table for a specific data collection point, the user can use a STable as a template and specify the tag values of this specific DCP to create it. ** The table created by using a STable as the template is called subtable** in TDengine. The difference between regular table and subtable is: + 1. Subtable is a table, all SQL commands applied on a regular table can be applied on subtable. 2. Subtable is a table with extensions, it has static tags (labels), and these tags can be added, deleted, and updated after it is created. But a regular table does not have tags. 3. A subtable belongs to only one STable, but a STable may have many subtables. Regular tables do not belong to a STable. -4. A regular table can not be converted into a subtable, and vice versa. +4. A regular table can not be converted into a subtable, and vice versa. The relationship between a STable and the subtables created based on this STable is as follows: @@ -151,13 +154,13 @@ The relationship between a STable and the subtables created based on this STable 2. The schema of metrics or labels cannot be adjusted through subtables, and it can only be changed via STable. Changes to the schema of a STable takes effect immediately for all associated subtables. 3. STable defines only one template and does not store any data or label information by itself. Therefore, data cannot be written to a STable, only to subtables. -Queries can be executed on both a table (subtable) and a STable. For a query on a STable, TDengine will treat the data in all its subtables as a whole data set for processing. TDengine will first find the subtables that meet the tag filter conditions, then scan the time-series data of these subtables to perform aggregation operation, which reduces the number of data sets to be scanned which in turn greatly improves the performance of data aggregation across multiple DCPs. +Queries can be executed on both a table (subtable) and a STable. For a query on a STable, TDengine will treat the data in all its subtables as a whole data set for processing. TDengine will first find the subtables that meet the tag filter conditions, then scan the time-series data of these subtables to perform aggregation operation, which reduces the number of data sets to be scanned which in turn greatly improves the performance of data aggregation across multiple DCPs. In essence, querying a supertable is a very efficient aggregate query on multiple DCPs of the same type. -In TDengine, it is recommended to use a subtable instead of a regular table for a DCP. +In TDengine, it is recommended to use a subtable instead of a regular table for a DCP. ## Database -A database is a collection of tables. TDengine allows a running instance to have multiple databases, and each database can be configured with different storage policies. Different types of DCPs often have different data characteristics, including the frequency of data collection, data retention time, the number of replications, the size of data blocks, whether data is allowed to be updated, and so on. In order for TDengine to work with maximum efficiency in various scenarios, TDengine recommends that STables with different data characteristics be created in different databases. +A database is a collection of tables. TDengine allows a running instance to have multiple databases, and each database can be configured with different storage policies. The [characteristics of time-series data](https://www.taosdata.com/blog/2019/07/09/86.html) from different data collection points may be different. Characteristics include collection frequency, retention policy and others which determine how you create and configure the database. For e.g. days to keep, number of replicas, data block size, whether data updates are allowed and other configurable parameters would be determined by the characteristics of your data and your business requirements. In order for TDengine to work with maximum efficiency in various scenarios, TDengine recommends that STables with different data characteristics be created in different databases. In a database, there can be one or more STables, but a STable belongs to only one database. All tables owned by a STable are stored in only one database. @@ -167,4 +170,4 @@ FQDN (Fully Qualified Domain Name) is the full domain name of a specific compute Each node of a TDengine cluster is uniquely identified by an End Point, which consists of an FQDN and a Port, such as h1.tdengine.com:6030. In this way, when the IP changes, we can still use the FQDN to dynamically find the node without changing any configuration of the cluster. In addition, FQDN is used to facilitate unified access to the same cluster from the Intranet and the Internet. -TDengine does not recommend using an IP address to access the cluster. FQDN is recommended for cluster management. +TDengine does not recommend using an IP address to access the cluster. FQDN is recommended for cluster management. diff --git a/docs/en/05-get-started/01-docker.md b/docs/en/05-get-started/01-docker.md index 14f5a8800072971a2ffa8550c838212d7b6a9907..869390cd6c30e8ddc4f0b11ef2402bd7ffb6cf7d 100644 --- a/docs/en/05-get-started/01-docker.md +++ b/docs/en/05-get-started/01-docker.md @@ -1,103 +1,98 @@ --- sidebar_label: Docker -title: 通过 Docker 快速体验 TDengine +title: Quick Install on Docker --- -:::info -如果您希望对 TDengine 贡献代码或对内部实现感兴趣，请参考我们的 [TDengine GitHub 主页](https://github.com/taosdata/TDengine) 下载源码构建和安装. -::: -本节首先介绍如何通过 Docker 快速体验 TDengine，然后介绍如何在 Docker 环境下体验 TDengine 的写入和查询功能。 +This document describes how to install TDengine in a Docker container and perform queries and inserts. To get started with TDengine in a non-containerized environment, see [Quick Install](../../get-started/package). If you want to view the source code, build TDengine yourself, or contribute to the project, see the [TDengine GitHub repository](https://github.com/taosdata/TDengine). -## 启动 TDengine +## Run TDengine -如果已经安装了 docker， 只需执行下面的命令。 +If Docker is already installed on your computer, run the following command: ```shell -docker run -d -p 6030:6030 -p 6041/6041 -p 6043-6049/6043-6049 -p 6043-6049:6043-6049/udp tdengine/tdengine +docker run -d -p 6030:6030 -p 6041:6041 -p 6043-6049:6043-6049 -p 6043-6049:6043-6049/udp tdengine/tdengine ``` -注意：TDengine 3.0 服务端仅使用 6030 TCP 端口。6041 为 taosAdapter 所使用提供 REST 服务端口。6043-6049 为 taosAdapter 提供第三方应用接入所使用端口，可根据需要选择是否打开。 +Note that TDengine Server uses TCP port 6030. Port 6041 is used by taosAdapter for the REST API service. Ports 6043 through 6049 are used by taosAdapter for other connectors. You can open these ports as needed. -确定该容器已经启动并且在正常运行 +Run the following command to ensure that your container is running: ```shell docker ps ``` -进入该容器并执行 bash +Enter the container and open the bash shell: ```shell docker exec -it bash ``` -然后就可以执行相关的 Linux 命令操作和访问 TDengine +You can now access TDengine or run other Linux commands. -## 运行 TDengine CLI +Note: For information about installing docker, see the [official documentation](https://docs.docker.com/get-docker/). -进入容器，执行 taos +## Open the TDengine CLI + +On the container, run the following command to open the TDengine CLI: ``` $ taos -Welcome to the TDengine shell from Linux, Client Version:3.0.0.0 -Copyright (c) 2022 by TAOS Data, Inc. All rights reserved. - -Server is Community Edition. taos> ``` -## 写入数据 +## Insert Data into TDengine -可以使用 TDengine 的自带工具 taosBenchmark 快速体验 TDengine 的写入。 +You can use the `taosBenchmark` tool included with TDengine to write test data into your deployment. -进入容器，启动 taosBenchmark： +To do so, run the following command: ```bash $ taosBenchmark ``` - 该命令将在数据库 test 下面自动创建一张超级表 meters，该超级表下有 1 万张表，表名为 "d0" 到 "d9999"，每张表有 1 万条记录，每条记录有 (ts, current, voltage, phase) 四个字段，时间戳从 "2017-07-14 10:40:00 000" 到 "2017-07-14 10:40:09 999"，每张表带有标签 location 和 groupId，groupId 被设置为 1 到 10， location 被设置为 "San Francisco" 或者 "Los Angeles"等城市名称。 + This command creates the `meters` supertable in the `test` database. In the `meters` supertable, it then creates 10,000 subtables named `d0` to `d9999`. Each table has 10,000 rows and each row has four columns: `ts`, `current`, `voltage`, and `phase`. The timestamps of the data in these columns range from 2017-07-14 10:40:00 000 to 2017-07-14 10:40:09 999. Each table is randomly assigned a `groupId` tag from 1 to ten and a `location` tag of either `California.SanFrancisco` or `California.SanDiego`. - 这条命令很快完成 1 亿条记录的插入。具体时间取决于硬件性能。 + The `taosBenchmark` command creates a deployment with 100 million data points that you can use for testing purposes. The time required depends on the hardware specifications of the local system. - taosBenchmark 命令本身带有很多选项，配置表的数目、记录条数等等，您可以设置不同参数进行体验，请执行 `taosBenchmark --help` 详细列出。taosBenchmark 详细使用方法请参照 [taosBenchmark 参考手册](../../reference/taosbenchmark)。 + You can customize the test deployment that taosBenchmark creates by specifying command-line parameters. For information about command-line parameters, run the `taosBenchmark --help` command. For more information about taosBenchmark, see [taosBenchmark](/reference/taosbenchmark). -## 体验查询 +## Query Data in TDengine -使用上述 taosBenchmark 插入数据后，可以在 TDengine CLI 输入查询命令，体验查询速度。。 +After using taosBenchmark to create your test deployment, you can run queries in the TDengine CLI to test its performance. For example: -查询超级表下记录总条数： +Query the number of rows in the `meters` supertable: ```sql taos> select count(*) from test.meters; ``` -查询 1 亿条记录的平均值、最大值、最小值等： +Query the average, maximum, and minimum values of all 100 million rows of data: ```sql taos> select avg(current), max(voltage), min(phase) from test.meters; ``` -查询 location="San Francisco" 的记录总条数： +Query the number of rows whose `location` tag is `California.SanFrancisco`: ```sql taos> select count(*) from test.meters where location="San Francisco"; ``` -查询 groupId=10 的所有记录的平均值、最大值、最小值等： +Query the average, maximum, and minimum values of all rows whose `groupId` tag is `10`: ```sql taos> select avg(current), max(voltage), min(phase) from test.meters where groupId=10; ``` -对表 d10 按 10s 进行平均值、最大值和最小值聚合统计： +Query the average, maximum, and minimum values for table `d10` in 10 second intervals: ```sql taos> select avg(current), max(voltage), min(phase) from test.d10 interval(10s); ``` -## 其它 +## Additional Information -更多关于在 Docker 环境下使用 TDengine 的细节，请参考 [在 Docker 下使用 TDengine](../../reference/docker) +For more information about deploying TDengine in a Docker environment, see [Using TDengine in Docker](../../reference/docker). diff --git a/docs/en/05-get-started/03-package.md b/docs/en/05-get-started/03-package.md index 6423cc710523a0be95e95ba9e50556e332659dfd..6f6a5087fa7e73f49ed97c5ea5f7db1668cebe9f 100644 --- a/docs/en/05-get-started/03-package.md +++ b/docs/en/05-get-started/03-package.md @@ -1,160 +1,192 @@ --- -sidebar_label: 安装包 -title: 使用安装包立即开始 +sidebar_label: Package +title: Quick Install from Package --- import Tabs from "@theme/Tabs"; import TabItem from "@theme/TabItem"; +import PkgListV3 from "/components/PkgListV3"; -:::info -如果您希望对 TDengine 贡献代码或对内部实现感兴趣，请参考我们的 [TDengine GitHub 主页](https://github.com/taosdata/TDengine) 下载源码构建和安装. +For information about installing TDengine on Docker, see [Quick Install on Docker](../../get-started/docker). If you want to view the source code, build TDengine yourself, or contribute to the project, see the [TDengine GitHub repository](https://github.com/taosdata/TDengine). -::: +The full package of TDengine includes the TDengine Server (`taosd`), TDengine Client (`taosc`), taosAdapter for connecting with third-party systems and providing a RESTful interface, a command-line interface, and some tools. Note that taosAdapter supports Linux only. In addition to connectors for multiple languages, TDengine also provides a [REST API](../../reference/rest-api) through [taosAdapter](../../reference/taosadapter). -TDengine 开源版本提供 deb 和 rpm 格式安装包，用户可以根据自己的运行环境选择合适的安装包。其中 deb 支持 Debian/Ubuntu 及衍生系统，rpm 支持 CentOS/RHEL/SUSE 及衍生系统。同时我们也为企业用户提供 tar.gz 格式安装包，也支持通过 `apt-get` 工具从线上进行安装。 +The standard server installation package includes `taos`, `taosd`, `taosAdapter`, `taosBenchmark`, and sample code. You can also download a lite package that includes only `taosd` and the C/C++ connector. -## 安装 +The TDengine Community Edition is released as .deb and .rpm packages. The .deb package can be installed on Debian, Ubuntu, and derivative systems. The .rpm package can be installed on CentOS, RHEL, SUSE, and derivative systems. A .tar.gz package is also provided for enterprise customers, and you can install TDengine over `apt-get` as well. The .tar.tz package includes `taosdump` and the TDinsight installation script. If you want to use these utilities with the .deb or .rpm package, download and install taosTools separately. TDengine can also be installed on 64-bit Windows servers. + +## Installation - -可以使用 apt-get 工具从官方仓库安装。 + -**安装包仓库** +1. Download the .deb installation package. + +2. In the directory where the package is located, use `dpkg` to install the package: ```bash -wget -qO - http://repos.taosdata.com/tdengine.key | sudo apt-key add - -echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-stable stable main" | sudo tee /etc/apt/sources.list.d/tdengine-stable.list +# Enter the name of the package that you downloaded. +sudo dpkg -i TDengine-server--Linux-x64.deb ``` -如果安装 Beta 版需要安装包仓库 + -```bash -echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-beta beta main" | sudo tee /etc/apt/sources.list.d/tdengine-beta.list -``` + -**使用 apt-get 命令安装** +1. Download the .rpm installation package. + +2. In the directory where the package is located, use rpm to install the package: ```bash -sudo apt-get update -apt-cache policy tdengine -sudo apt-get install tdengine +# Enter the name of the package that you downloaded. +sudo rpm -ivh TDengine-server--Linux-x64.rpm ``` -:::tip -apt-get 方式只适用于 Debian 或 Ubuntu 系统 -:::: - -1、从官网下载获得 deb 安装包，例如 TDengine-server-3.0.0.0-Linux-x64.deb； -2、进入到 TDengine-server-3.0.0.0-Linux-x64.deb 安装包所在目录，执行如下的安装命令： + + +1. Download the .tar.gz installation package. + +2. In the directory where the package is located, use `tar` to decompress the package: ```bash -sudo dpkg -i TDengine-server-3.0.0.0-Linux-x64.deb +# Enter the name of the package that you downloaded. +tar -zxvf TDengine-server--Linux-x64.tar.gz ``` - - - - -1、从官网下载获得 rpm 安装包，例如 TDengine-server-3.0.0.0-Linux-x64.rpm； -2、进入到 TDengine-server-3.0.0.0-Linux-x64.rpm 安装包所在目录，执行如下的安装命令： +In the directory to which the package was decompressed, run `install.sh`: ```bash -sudo rpm -ivh TDengine-server-3.0.0.0-Linux-x64.rpm +sudo ./install.sh ``` +:::info +Users will be prompted to enter some configuration information when install.sh is executing. The interactive mode can be disabled by executing `./install.sh -e no`. `./install.sh -h` can show all parameters with detailed explanation. +::: + - + + +1. Download the Windows installation package. + +2. Run the downloaded package to install TDengine. + + + +You can use `apt-get` to install TDengine from the official package repository. -1、从官网下载获得 tar.gz 安装包，例如 TDengine-server-3.0.0.0-Linux-x64.tar.gz； -2、进入到 TDengine-server-3.0.0.0-Linux-x64.tar.gz 安装包所在目录，先解压文件后，进入子目录，执行其中的 install.sh 安装脚本： +**Configure the package repository** ```bash -tar -zxvf TDengine-server-3.0.0.0-Linux-x64.tar.gz +wget -qO - http://repos.taosdata.com/tdengine.key | sudo apt-key add - +echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-stable stable main" | sudo tee /etc/apt/sources.list.d/tdengine-stable.list ``` -解压后进入相应路径，执行 +You can install beta versions by configuring the following repository: ```bash -sudo ./install.sh +wget -qO - http://repos.taosdata.com/tdengine.key | sudo apt-key add - +echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-beta beta main" | sudo tee /etc/apt/sources.list.d/tdengine-beta.list ``` -:::info -install.sh 安装脚本在执行过程中，会通过命令行交互界面询问一些配置信息。如果希望采取无交互安装方式，那么可以用 -e no 参数来执行 install.sh 脚本。运行 `./install.sh -h` 指令可以查看所有参数的详细说明信息。 +**Install TDengine with `apt-get`** -::: +```bash +sudo apt-get update +apt-cache policy tdengine +sudo apt-get install tdengine +``` +:::tip +This installation method is supported only for Debian and Ubuntu. +:::: +:::info +For information about TDengine releases, see [Release History](../../releases). +::: + :::note -当安装第一个节点时，出现 Enter FQDN：提示的时候，不需要输入任何内容。只有当安装第二个或以后更多的节点时，才需要输入已有集群中任何一个可用节点的 FQDN，支持该新节点加入集群。当然也可以不输入，而是在新节点启动前，配置到新节点的配置文件中。 +On the first node in your TDengine cluster, leave the `Enter FQDN:` prompt blank and press **Enter**. On subsequent nodes, you can enter the end point of the first dnode in the cluster. You can also configure this setting after you have finished installing TDengine. ::: -## 启动 +## Quick Launch -安装后，请使用 `systemctl` 命令来启动 TDengine 的服务进程。 + + + +After the installation is complete, run the following command to start the TDengine service: ```bash systemctl start taosd ``` -检查服务是否正常工作： +Run the following command to confirm that TDengine is running normally: ```bash systemctl status taosd ``` -如果服务进程处于活动状态，则 status 指令会显示如下的相关信息： +Output similar to the following indicates that TDengine is running normally: ``` Active: active (running) ``` -如果后台服务进程处于停止状态，则 status 指令会显示如下的相关信息： +Output similar to the following indicates that TDengine has not started successfully: ``` Active: inactive (dead) ``` -如果 TDengine 服务正常工作，那么您可以通过 TDengine 的命令行程序 `taos` 来访问并体验 TDengine。 +After confirming that TDengine is running, run the `taos` command to access the TDengine CLI. -systemctl 命令汇总： +The following `systemctl` commands can help you manage TDengine: -- 启动服务进程：`systemctl start taosd` +- Start TDengine Server: `systemctl start taosd` -- 停止服务进程：`systemctl stop taosd` +- Stop TDengine Server: `systemctl stop taosd` -- 重启服务进程：`systemctl restart taosd` +- Restart TDengine Server: `systemctl restart taosd` -- 查看服务状态：`systemctl status taosd` +- Check TDengine Server status: `systemctl status taosd` :::info -- systemctl 命令需要 _root_ 权限来运行，如果您非 _root_ 用户，请在命令前添加 sudo 。 -- `systemctl stop taosd` 指令在执行后并不会马上停止 TDengine 服务，而是会等待系统中必要的落盘工作正常完成。在数据量很大的情况下，这可能会消耗较长时间。 -- 如果系统中不支持 `systemd`，也可以用手动运行 `/usr/local/taos/bin/taosd` 方式启动 TDengine 服务。 +- The `systemctl` command requires _root_ privileges. If you are not logged in as the `root` user, use the `sudo` command. +- The `systemctl stop taosd` command does not instantly stop TDengine Server. The server is stopped only after all data in memory is flushed to disk. The time required depends on the cache size. +- If your system does not include `systemd`, you can run `/usr/local/taos/bin/taosd` to start TDengine manually. ::: -## TDengine 命令行 (CLI) + + + + +After the installation is complete, run `C:\TDengine\taosd.exe` to start TDengine Server. + + + + +## Command Line Interface -为便于检查 TDengine 的状态，执行数据库 (Database) 的各种即席(Ad Hoc)查询，TDengine 提供一命令行应用程序(以下简称为 TDengine CLI) taos。要进入 TDengine 命令行，您只要在安装有 TDengine 的 Linux 终端执行 `taos` 即可。 +You can use the TDengine CLI to monitor your TDengine deployment and execute ad hoc queries. To open the CLI, run the following command: ```bash taos ``` -如果连接服务成功，将会打印出欢迎消息和版本信息。如果失败，则会打印错误消息出来（请参考 [FAQ](/train-faq/faq) 来解决终端连接服务端失败的问题）。 TDengine CLI 的提示符号如下： +The TDengine CLI displays a welcome message and version information to indicate that its connection to the TDengine service was successful. If an error message is displayed, see the [FAQ](/train-faq/faq) for troubleshooting information. At the following prompt, you can execute SQL commands. ```cmd taos> ``` -在 TDengine CLI 中，用户可以通过 SQL 命令来创建/删除数据库、表等，并进行数据库(database)插入查询操作。在终端中运行的 SQL 语句需要以分号结束来运行。示例： +For example, you can create and delete databases and tables and run all types of queries. Each SQL command must be end with a semicolon (;). For example: ```sql create database demo; @@ -170,51 +202,51 @@ select * from t; Query OK, 2 row(s) in set (0.003128s) ``` -除执行 SQL 语句外，系统管理员还可以从 TDengine CLI 进行检查系统运行状态、添加删除用户账号等操作。TDengine CLI 连同应用驱动也可以独立安装在 Linux 或 Windows 机器上运行，更多细节请参考 [这里](../../reference/taos-shell/) +You can also can monitor the deployment status, add and remove user accounts, and manage running instances. You can run the TDengine CLI on either Linux or Windows machines. For more information, see [TDengine CLI](../../reference/taos-shell/). -## 使用 taosBenchmark 体验写入速度 +## Test data insert performance -启动 TDengine 的服务，在 Linux 终端执行 `taosBenchmark` （曾命名为 `taosdemo`）： +After your TDengine Server is running normally, you can run the taosBenchmark utility to test its performance: ```bash taosBenchmark ``` -该命令将在数据库 test 下面自动创建一张超级表 meters，该超级表下有 1 万张表，表名为 "d0" 到 "d9999"，每张表有 1 万条记录，每条记录有 (ts, current, voltage, phase) 四个字段，时间戳从 "2017-07-14 10:40:00 000" 到 "2017-07-14 10:40:09 999"，每张表带有标签 location 和 groupId，groupId 被设置为 1 到 10， location 被设置为 "California.SanFrancisco" 或者 "California.LosAngeles"。 +This command creates the `meters` supertable in the `test` database. In the `meters` supertable, it then creates 10,000 subtables named `d0` to `d9999`. Each table has 10,000 rows and each row has four columns: `ts`, `current`, `voltage`, and `phase`. The timestamps of the data in these columns range from 2017-07-14 10:40:00 000 to 2017-07-14 10:40:09 999. Each table is randomly assigned a `groupId` tag from 1 to ten and a `location` tag of either `California.SanFrancisco` or `California.LosAngeles`. -这条命令很快完成 1 亿条记录的插入。具体时间取决于硬件性能，即使在一台普通的 PC 服务器往往也仅需十几秒。 +The `taosBenchmark` command creates a deployment with 100 million data points that you can use for testing purposes. The time required to create the deployment depends on your hardware. On most modern servers, the deployment is created in less than a minute. -taosBenchmark 命令本身带有很多选项，配置表的数目、记录条数等等，您可以设置不同参数进行体验，请执行 `taosBenchmark --help` 详细列出。taosBenchmark 详细使用方法请参照 [如何使用 taosBenchmark 对 TDengine 进行性能测试](https://www.taosdata.com/2021/10/09/3111.html)。 +You can customize the test deployment that taosBenchmark creates by specifying command-line parameters. For information about command-line parameters, run the `taosBenchmark --help` command. For more information about taosBenchmark, see [taosBenchmark](../../reference/taosbenchmark). -## 使用 TDengine CLI 体验查询速度 +## Test data query performance -使用上述 taosBenchmark 插入数据后，可以在 TDengine CLI 输入查询命令，体验查询速度。 +After using taosBenchmark to create your test deployment, you can run queries in the TDengine CLI to test its performance: -查询超级表下记录总条数： +Query the number of rows in the `meters` supertable: ```sql taos> select count(*) from test.meters; ``` -查询 1 亿条记录的平均值、最大值、最小值等： +Query the average, maximum, and minimum values of all 100 million rows of data: ```sql taos> select avg(current), max(voltage), min(phase) from test.meters; ``` -查询 location="California.SanFrancisco" 的记录总条数： +Query the number of rows whose `location` tag is `California.SanFrancisco`: ```sql taos> select count(*) from test.meters where location="California.SanFrancisco"; ``` -查询 groupId=10 的所有记录的平均值、最大值、最小值等： +Query the average, maximum, and minimum values of all rows whose `groupId` tag is `10`: ```sql taos> select avg(current), max(voltage), min(phase) from test.meters where groupId=10; ``` -对表 d10 按 10s 进行平均值、最大值和最小值聚合统计： +Query the average, maximum, and minimum values for table `d10` in 10 second intervals: ```sql taos> select avg(current), max(voltage), min(phase) from test.d10 interval(10s); diff --git a/docs/en/05-get-started/_apt_get_install.mdx b/docs/en/05-get-started/_apt_get_install.mdx index b1bc4a13517bbfdc9eda86a58b89aee8e41fa470..acbcf22122152f970891ca92c20b9b1b7d97fe1a 100644 --- a/docs/en/05-get-started/_apt_get_install.mdx +++ b/docs/en/05-get-started/_apt_get_install.mdx @@ -1,19 +1,19 @@ -可以使用 apt-get 工具从官方仓库安装。 +You can use `apt-get` to install TDengine from the official package repository. -**安装包仓库** +**Configure the package repository** ``` wget -qO - http://repos.taosdata.com/tdengine.key | sudo apt-key add - echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-stable stable main" | sudo tee /etc/apt/sources.list.d/tdengine-stable.list ``` -如果安装 Beta 版需要安装包仓库 +You can install beta versions by configuring the following package repository: ``` echo "deb [arch=amd64] http://repos.taosdata.com/tdengine-beta beta main" | sudo tee /etc/apt/sources.list.d/tdengine-beta.list ``` -**使用 apt-get 命令安装** +**Install TDengine with `apt-get`** ``` sudo apt-get update @@ -22,5 +22,5 @@ sudo apt-get install tdengine ``` :::tip -apt-get 方式只适用于 Debian 或 Ubuntu 系统 +This installation method is supported only for Debian and Ubuntu. :::: diff --git a/docs/en/05-get-started/_category_.yml b/docs/en/05-get-started/_category_.yml index b2348fade63c7bb717eac3e6e6b8dfda3c73b17a..043ae21554ffd8f274c6afe41c5ae5e7da742b26 100644 --- a/docs/en/05-get-started/_category_.yml +++ b/docs/en/05-get-started/_category_.yml @@ -1 +1 @@ -label: 立即开始 +label: Get Started diff --git a/docs/en/05-get-started/_pkg_install.mdx b/docs/en/05-get-started/_pkg_install.mdx index 83c987af8bcf24a9593105b680d32a0421344d5f..32d7c1f376bc6404c45db39fbaa24a5012a11f01 100644 --- a/docs/en/05-get-started/_pkg_install.mdx +++ b/docs/en/05-get-started/_pkg_install.mdx @@ -1,17 +1,17 @@ import PkgList from "/components/PkgList"; -TDengine 的安装非常简单，从下载到安装成功仅仅只要几秒钟。 +TDengine is easy to download and install. -为方便使用，从 2.4.0.10 开始，标准的服务端安装包包含了 taos、taosd、taosAdapter、taosdump、taosBenchmark、TDinsight 安装脚本和示例代码；如果您只需要用到服务端程序和客户端连接的 C/C++ 语言支持，也可以仅下载 lite 版本的安装包。 +The standard server installation package includes `taos`, `taosd`, `taosAdapter`, `taosBenchmark`, and sample code. You can also download a lite package that includes only `taosd` and the C/C++ connector. -在安装包格式上，我们提供 tar.gz, rpm 和 deb 格式，为企业客户提供 tar.gz 格式安装包，以方便在特定操作系统上使用。需要注意的是，rpm 和 deb 包不含 taosdump、taosBenchmark 和 TDinsight 安装脚本，这些工具需要通过安装 taosTool 包获得。 +You can download the TDengine installation package in .rpm, .deb, or .tar.gz format. The .tar.tz package includes `taosdump` and the TDinsight installation script. If you want to use these utilities with the .deb or .rpm package, download and install taosTools separately. -发布版本包括稳定版和 Beta 版，Beta 版含有更多新功能。正式上线或测试建议安装稳定版。您可以根据需要选择下载： +Between official releases, beta versions may be released that contain new features. Do not use beta versions for production or testing environments. Select the installation package appropriate for your system. -具体的安装方法，请参见[安装包的安装和卸载](/operation/pkg-install)。 +For information about installing TDengine, see [Install and Uninstall](/operation/pkg-install). -下载其他组件、最新 Beta 版及之前版本的安装包，请点击[这里](https://www.taosdata.com/all-downloads) +For information about TDengine releases, see [All Downloads](https://tdengine.com/all-downloads) -查看 Release Notes, 请点击[这里](https://github.com/taosdata/TDengine/releases) +and [Release Notes](https://github.com/taosdata/TDengine/releases). diff --git a/docs/en/05-get-started/index.md b/docs/en/05-get-started/index.md index 794081b4e4c438dee2d8cbe125de4094056f190f..cf475a8cd79e15880a4356a89f46c0dd6a8c307d 100644 --- a/docs/en/05-get-started/index.md +++ b/docs/en/05-get-started/index.md @@ -1,11 +1,11 @@ --- -title: 立即开始 -description: '快速设置 TDengine 环境并体验其高效写入和查询' +title: Get Started +description: This article describes how to install TDengine and test its performance. --- -TDengine 完整的软件包包括服务端（taosd）、用于与第三方系统对接并提供 RESTful 接口的 taosAdapter、应用驱动（taosc）、命令行程序 (CLI，taos) 和一些工具软件。TDengine 除了提供多种语言的连接器之外，还通过 [taosAdapter](/reference/taosadapter) 提供 [RESTful 接口](/reference/rest-api)。 +The full package of TDengine includes the TDengine Server (`taosd`), TDengine Client (`taosc`), taosAdapter for connecting with third-party systems and providing a RESTful interface, a command-line interface, and some tools. In addition to connectors for multiple languages, TDengine also provides a [RESTful interface](/reference/rest-api) through [taosAdapter](/reference/taosadapter). -本章主要介绍如何利用 Docker 或者安装包快速设置 TDengine 环境并体验其高效写入和查询。 +You can install and run TDengine on Linux and Windows machines as well as Docker containers. ```mdx-code-block import DocCardList from '@theme/DocCardList'; diff --git a/docs/en/07-develop/01-connect/_connect_php.mdx b/docs/en/07-develop/01-connect/_connect_php.mdx new file mode 100644 index 0000000000000000000000000000000000000000..dbad72bc1988bd5336f1da132dd9e6ba9b8020e6 --- /dev/null +++ b/docs/en/07-develop/01-connect/_connect_php.mdx @@ -0,0 +1,3 @@ +```php title="原生连接" +{{#include docs/examples/php/connect.php}} +``` diff --git a/docs/en/07-develop/01-connect/index.md b/docs/en/07-develop/01-connect/index.md index 1318f4619ba5ff1da23990f881a15f3726401760..017a1a0ee40a242c198994ae6bb432203061ac18 100644 --- a/docs/en/07-develop/01-connect/index.md +++ b/docs/en/07-develop/01-connect/index.md @@ -1,38 +1,39 @@ --- -sidebar_label: Connect title: Connect -description: "This document explains how to establish connections to TDengine, and briefly introduces how to install and use TDengine connectors." +description: "This document explains how to establish connections to TDengine and how to install and use TDengine connectors." --- import Tabs from "@theme/Tabs"; import TabItem from "@theme/TabItem"; -import ConnJava from "./\_connect_java.mdx"; -import ConnGo from "./\_connect_go.mdx"; -import ConnRust from "./\_connect_rust.mdx"; -import ConnNode from "./\_connect_node.mdx"; -import ConnPythonNative from "./\_connect_python.mdx"; -import ConnCSNative from "./\_connect_cs.mdx"; -import ConnC from "./\_connect_c.mdx"; -import ConnR from "./\_connect_r.mdx"; -import InstallOnWindows from "../../14-reference/03-connector/\_linux_install.mdx"; -import InstallOnLinux from "../../14-reference/03-connector/\_windows_install.mdx"; -import VerifyLinux from "../../14-reference/03-connector/\_verify_linux.mdx"; -import VerifyWindows from "../../14-reference/03-connector/\_verify_windows.mdx"; - -Any application programs running on any kind of platform can access TDengine through the REST API provided by TDengine. For details, please refer to [REST API](/reference/rest-api/). Additionally, application programs can use the connectors of multiple programming languages including C/C++, Java, Python, Go, Node.js, C#, Rust to access TDengine. This chapter describes how to establish a connection to TDengine and briefly introduces how to install and use connectors. TDengine community also provides connectors in LUA and PHP languages. For details about the connectors, please refer to [Connectors](/reference/connector/). +import ConnJava from "./_connect_java.mdx"; +import ConnGo from "./_connect_go.mdx"; +import ConnRust from "./_connect_rust.mdx"; +import ConnNode from "./_connect_node.mdx"; +import ConnPythonNative from "./_connect_python.mdx"; +import ConnCSNative from "./_connect_cs.mdx"; +import ConnC from "./_connect_c.mdx"; +import ConnR from "./_connect_r.mdx"; +import ConnPHP from "./_connect_php.mdx"; +import InstallOnWindows from "../../14-reference/03-connector/_linux_install.mdx"; +import InstallOnLinux from "../../14-reference/03-connector/_windows_install.mdx"; +import VerifyLinux from "../../14-reference/03-connector/_verify_linux.mdx"; +import VerifyWindows from "../../14-reference/03-connector/_verify_windows.mdx"; + +Any application running on any platform can access TDengine through the REST API provided by TDengine. For information, see [REST API](/reference/rest-api/). Applications can also use the connectors for various programming languages, including C/C++, Java, Python, Go, Node.js, C#, and Rust, to access TDengine. These connectors support connecting to TDengine clusters using both native interfaces (taosc). Some connectors also support connecting over a REST interface. Community developers have also contributed several unofficial connectors, such as the ADO.NET connector, the Lua connector, and the PHP connector. ## Establish Connection There are two ways for a connector to establish connections to TDengine: -1. Connection through the REST API provided by the taosAdapter component, this way is called "REST connection" hereinafter. -2. Connection through the TDengine client driver (taosc), this way is called "Native connection" hereinafter. +1. REST connection through the REST API provided by the taosAdapter component. +2. Native connection through the TDengine client driver (taosc). + +For REST and native connections, connectors provide similar APIs for performing operations and running SQL statements on your databases. The main difference is the method of establishing the connection, which is not visible to users. Key differences： -1. The TDengine client driver (taosc) has the highest performance with all the features of TDengine like [Parameter Binding](/reference/connector/cpp#parameter-binding-api), [Subscription](/reference/connector/cpp#subscription-and-consumption-api), etc. -2. The TDengine client driver (taosc) is not supported across all platforms, and applications built on taosc may need to be modified when updating taosc to newer versions. 3. The REST connection is more accessible with cross-platform support, however it results in a 30% performance downgrade. +1. The TDengine client driver (taosc) has the highest performance with all the features of TDengine like [Parameter Binding](/reference/connector/cpp#parameter-binding-api), [Subscription](/reference/connector/cpp#subscription-and-consumption-api), etc. ## Install Client Driver taosc @@ -136,19 +137,19 @@ Node.js connector provides different ways of establishing connections by providi 1. Install Node.js Native Connector -``` -npm install @tdengine/client -``` + ``` + npm install @tdengine/client + ``` :::note It's recommend to use Node whose version is between `node-v12.8.0` and `node-v13.0.0`. -::: - +::: + 2. Install Node.js REST Connector -``` -npm install @tdengine/rest -``` + ``` + npm install @tdengine/rest + ``` @@ -236,14 +237,14 @@ phpize && ./configure --enable-swoole && make -j && make install **Enable The Extension:** -Option One: Add `extension=tdengine` in `php.ini` +Option One: Add `extension=tdengine` in `php.ini` Option Two: Specify the extension on CLI `php -d extension=tdengine test.php` -## Establish Connection +## Establish a connection Prior to establishing connection, please make sure TDengine is already running and accessible. The following sample code assumes TDengine is running on the same host as the client program, with FQDN configured to "localhost" and serverPort configured to "6030". @@ -272,6 +273,9 @@ Prior to establishing connection, please make sure TDengine is already running a + + + :::tip diff --git a/docs/en/07-develop/02-model/index.mdx b/docs/en/07-develop/02-model/index.mdx index b647c845d070c26398956f8a9de81864b73120e1..3c16ed2df15940f3f59232a37b6f559010305a96 100644 --- a/docs/en/07-develop/02-model/index.mdx +++ b/docs/en/07-develop/02-model/index.mdx @@ -2,31 +2,36 @@ title: Data Model --- -The data model employed by TDengine is similar to that of a relational database. You have to create databases and tables. You must design the data model based on your own business and application requirements. You should design the STable (an abbreviation for super table) schema to fit your data. This chapter will explain the big picture without getting into syntactical details. +The data model employed by TDengine is similar to that of a relational database. You have to create databases and tables. You must design the data model based on your own business and application requirements. You should design the [STable](/concept/#super-table-stable) (an abbreviation for super table) schema to fit your data. This chapter will explain the big picture without getting into syntactical details. + +Note: before you read this chapter, please make sure you have already read through [Key Concepts](/concept/), since TDengine introduces new concepts like "one table for one [data collection point](/concept/#data-collection-point)" and "[super table](/concept/#super-table-stable)". + + ## Create Database -The [characteristics of time-series data](https://www.taosdata.com/blog/2019/07/09/86.html) from different data collection points may be different. Characteristics include collection frequency, retention policy and others which determine how you create and configure the database. For e.g. days to keep, number of replicas, data block size, whether data updates are allowed and other configurable parameters would be determined by the characteristics of your data and your business requirements. For TDengine to operate with the best performance, we strongly recommend that you create and configure different databases for data with different characteristics. This allows you, for example, to set up different storage and retention policies. When creating a database, there are a lot of parameters that can be configured such as, the days to keep data, the number of replicas, the number of memory blocks, time precision, the minimum and maximum number of rows in each data block, whether compression is enabled, the time range of the data in single data file and so on. Below is an example of the SQL statement to create a database. +The characteristics of time-series data from different data collection points may be different. Characteristics include collection frequency, retention policy and others which determine how you create and configure the database. For e.g. days to keep, number of replicas, data block size, whether data updates are allowed and other configurable parameters would be determined by the characteristics of your data and your business requirements. For TDengine to operate with the best performance, we strongly recommend that you create and configure different databases for data with different characteristics. This allows you, for example, to set up different storage and retention policies. When creating a database, there are a lot of parameters that can be configured such as, the days to keep data, the number of replicas, the size of the cache, time precision, the minimum and maximum number of rows in each data block, whether compression is enabled, the time range of the data in single data file and so on. An example is shown as follows: ```sql -CREATE DATABASE power KEEP 365 DURATION 10 BUFFER 16 VGROUPS 100 WAL 1; +CREATE DATABASE power KEEP 365 DURATION 10 BUFFER 16 WAL_LEVEL 1; ``` In the above SQL statement: -- a database named "power" will be created -- the data in it will be kept for 365 days, which means that data older than 365 days will be deleted automatically +- a database named "power" is created +- the data in it is retained for 365 days, which means that data older than 365 days will be deleted automatically - a new data file will be created every 10 days -- the size of memory cache for writing is 16 MB -- data will be firstly written to WAL without FSYNC - -For more details please refer to [Database](/taos-sql/database). +- the size of the write cache pool on each vnode is 16 MB +- the number of vgroups is 100 +- WAL is enabled but fsync is disabled For more details please refer to [Database](/taos-sql/database). -After creating a database, the current database in use can be switched using SQL command `USE`. For example the SQL statement below switches the current database to `power`. Without the current database specified, table name must be preceded with the corresponding database name. +After creating a database, the current database in use can be switched using SQL command `USE`. For example the SQL statement below switches the current database to `power`. ```sql USE power; ``` +Without the current database specified, table name must be preceded with the corresponding database name. + :::note - Any table or STable must belong to a database. To create a table or STable, the database it belongs to must be ready. @@ -39,14 +44,9 @@ USE power; In a time-series application, there may be multiple kinds of data collection points. For example, in the electrical power system there are meters, transformers, bus bars, switches, etc. For easy and efficient aggregation of multiple tables, one STable needs to be created for each kind of data collection point. For example, for the meters in [table 1](/concept/#model_table1), the SQL statement below can be used to create the super table. ```sql -CREATE STable meters (ts timestamp, current float, voltage int, phase float) TAGS (location binary(64), groupId int); +CREATE STABLE meters (ts timestamp, current float, voltage int, phase float) TAGS (location binary(64), groupId int); ``` -:::note -If you are using versions prior to 2.0.15, the `STable` keyword needs to be replaced with `TABLE`. - -::: - Similar to creating a regular table, when creating a STable, the name and schema need to be provided. In the STable schema, the first column must always be a timestamp (like ts in the example), and the other columns (like current, voltage and phase in the example) are the data collected. The remaining columns can [contain data of type](/taos-sql/data-type/) integer, float, double, string etc. In addition, the schema for tags, like location and groupId in the example, must be provided. The tag type can be integer, float, string, etc. Tags are essentially the static properties of a data collection point. For example, properties like the location, device type, device group ID, manager ID are tags. Tags in the schema can be added, removed or updated. Please refer to [STable](/taos-sql/stable) for more details. For each kind of data collection point, a corresponding STable must be created. There may be many STables in an application. For electrical power system, we need to create a STable respectively for meters, transformers, busbars, switches. There may be multiple kinds of data collection points on a single device, for example there may be one data collection point for electrical data like current and voltage and another data collection point for environmental data like temperature, humidity and wind direction. Multiple STables are required for these kinds of devices. @@ -63,13 +63,8 @@ CREATE TABLE d1001 USING meters TAGS ("California.SanFrancisco", 2); In the above SQL statement, "d1001" is the table name, "meters" is the STable name, followed by the value of tag "Location" and the value of tag "groupId", which are "California.SanFrancisco" and "2" respectively in the example. The tag values can be updated after the table is created. Please refer to [Tables](/taos-sql/table) for details. -In the TDengine system, it's recommended to create a table for a data collection point via STable. A table created via STable is called subtable in some parts of the TDengine documentation. All SQL commands applied on regular tables can be applied on subtables. - -:::tip It's suggested to use the globally unique ID of a data collection point as the table name. For example the device serial number could be used as a unique ID. If a unique ID doesn't exist, multiple IDs that are not globally unique can be combined to form a globally unique ID. It's not recommended to use a globally unique ID as tag value. -::: - ## Create Table Automatically In some circumstances, it's unknown whether the table already exists when inserting rows. The table can be created automatically using the SQL statement below, and nothing will happen if the table already exists. @@ -84,8 +79,6 @@ For more details please refer to [Create Table Automatically](/taos-sql/insert#a ## Single Column vs Multiple Column -A multiple columns data model is supported in TDengine. As long as multiple metrics are collected by the same data collection point at the same time, i.e. the timestamps are identical, these metrics can be put in a single STable as columns. - -However, there is another kind of design, i.e. single column data model in which a table is created for each metric. This means that a STable is required for each kind of metric. For example in a single column model, 3 STables would be required for current, voltage and phase. +A multiple columns data model is supported in TDengine. As long as multiple metrics are collected by the same data collection point at the same time, i.e. the timestamps are identical, these metrics can be put in a single STable as columns. However, there is another kind of design, i.e. single column data model in which a table is created for each metric. This means that a STable is required for each kind of metric. For example in a single column model, 3 STables would be required for current, voltage and phase. It's recommended to use a multiple column data model as much as possible because insert and query performance is higher. In some cases, however, the collected metrics may vary frequently and so the corresponding STable schema needs to be changed frequently too. In such cases, it's more convenient to use single column data model. diff --git a/docs/en/07-develop/03-insert-data/01-sql-writing.mdx b/docs/en/07-develop/03-insert-data/01-sql-writing.mdx index da0e87abbc2c83ca940dd596ffbf5746a6b65823..f2168645ff9e59d60e88c85f86e890945b9f336c 100644 --- a/docs/en/07-develop/03-insert-data/01-sql-writing.mdx +++ b/docs/en/07-develop/03-insert-data/01-sql-writing.mdx @@ -1,5 +1,4 @@ --- -sidebar_label: Insert Using SQL title: Insert Using SQL --- @@ -19,13 +18,14 @@ import CsSQL from "./_cs_sql.mdx"; import CsStmt from "./_cs_stmt.mdx"; import CSQL from "./_c_sql.mdx"; import CStmt from "./_c_stmt.mdx"; +import PhpSQL from "./_php_sql.mdx"; +import PhpStmt from "./_php_stmt.mdx"; ## Introduction Application programs can execute `INSERT` statement through connectors to insert rows. The TDengine CLI can also be used to manually insert data. ### Insert Single Row - The below SQL statement is used to insert one row into table "d1001". ```sql @@ -42,7 +42,7 @@ INSERT INTO d1001 VALUES (1538548684000, 10.2, 220, 0.23) (1538548696650, 10.3, ### Insert into Multiple Tables -Data can be inserted into multiple tables in single SQL statement. The example below inserts 2 rows into table "d1001" and 1 row into table "d1002". +Data can be inserted into multiple tables in the same SQL statement. The example below inserts 2 rows into table "d1001" and 1 row into table "d1002". ```sql INSERT INTO d1001 VALUES (1538548685000, 10.3, 219, 0.31) (1538548695000, 12.6, 218, 0.33) d1002 VALUES (1538548696800, 12.3, 221, 0.31); @@ -52,19 +52,19 @@ For more details about `INSERT` please refer to [INSERT](/taos-sql/insert). :::info -- Inserting in batches can improve performance. Normally, the higher the batch size, the better the performance. Please note that a single row can't exceed 48 KB bytes and each SQL statement can't exceed 1 MB. -- Inserting with multiple threads can also improve performance. However, depending on the system resources on the application side and the server side, when the number of inserting threads grows beyond a specific point the performance may drop instead of improving. The proper number of threads needs to be tested in a specific environment to find the best number. The proper number of threads may be impacted by the system resources on the server side, the system resources on the client side, the table schemas, etc. +- Inserting in batches can improve performance. The higher the batch size, the better the performance. Please note that a single row can't exceed 48K bytes and each SQL statement can't exceed 1MB. +- Inserting with multiple threads can also improve performance. However, at a certain point, increasing the number of threads no longer offers any benefit and can even decrease performance due to the overhead involved in frequent thread switching. The optimal number of threads for a system depends on the processing capabilities and configuration of the server, the configuration of the database, the data schema, and the batch size for writing data. In general, more powerful clients and servers can support higher numbers of concurrently writing threads. Given a sufficiently powerful server, a higher number of vgroups for a database also increases the number of concurrent writes. Finally, a simpler data schema enables more concurrent writes as well. ::: :::warning -- If the timestamp for the row to be inserted already exists in the table, the old data will be overritten by the new values for the columns for which new values are provided, columns for which no new values are provided are not impacted. -- The timestamp to be inserted must be newer than the timestamp of subtracting current time by the parameter `KEEP`. If `KEEP` is set to 3650 days, then the data older than 3650 days ago can't be inserted. The timestamp to be inserted can't be newer than the timestamp of current time plus parameter `DURATION`. If `DAYS` is set to 2, the data newer than 2 days later can't be inserted. +- If the timestamp of a new record already exists in a table, columns with new data for that timestamp replace old data with new data, while columns without new data are not affected. +- The timestamp to be inserted must be newer than the timestamp of subtracting current time by the parameter `KEEP`. If `KEEP` is set to 3650 days, then the data older than 3650 days ago can't be inserted. The timestamp to be inserted cannot be newer than the timestamp of current time plus parameter `DURATION`. If `DURATION` is set to 2, the data newer than 2 days later can't be inserted. ::: -## Examples +## Sample program ### Insert Using SQL @@ -90,6 +90,9 @@ For more details about `INSERT` please refer to [INSERT](/taos-sql/insert). + + + :::note @@ -101,7 +104,7 @@ For more details about `INSERT` please refer to [INSERT](/taos-sql/insert). ### Insert with Parameter Binding -TDengine also provides API support for parameter binding. Similar to MySQL, only `?` can be used in these APIs to represent the parameters to bind. Parameter binding support for inserting data has improved significantly to improve the insert performance by avoiding the cost of parsing SQL statements. +TDengine also provides API support for parameter binding. Similar to MySQL, only `?` can be used in these APIs to represent the parameters to bind. This avoids the resource consumption of SQL syntax parsing when writing data through the parameter binding interface, thus significantly improving write performance in most cases. Parameter binding is available only with native connection. @@ -127,4 +130,8 @@ Parameter binding is available only with native connection. + + + + diff --git a/docs/en/07-develop/03-insert-data/02-influxdb-line.mdx b/docs/en/07-develop/03-insert-data/02-influxdb-line.mdx index 41109937053c31f0a141fcc90016397863152e57..11db3daeb054b2cac29c6a0ccde2add27774f3da 100644 --- a/docs/en/07-develop/03-insert-data/02-influxdb-line.mdx +++ b/docs/en/07-develop/03-insert-data/02-influxdb-line.mdx @@ -21,15 +21,15 @@ In the InfluxDB Line protocol format, a single line of text is used to represent measurement,tag_set field_set timestamp ``` -- `measurement` will be used as the name of the STable -- `tag_set` will be used as tags, with format like `=,=` -- `field_set`will be used as data columns, with format like `=,=` +- `measurement` will be used as the name of the STable Enter a comma (,) between `measurement` and `tag_set`. +- `tag_set` will be used as tags, with format like `=,=` Enter a space between `tag_set` and `field_set`. +- `field_set`will be used as data columns, with format like `=,=` Enter a space between `field_set` and `timestamp`. - `timestamp` is the primary key timestamp corresponding to this row of data For example: ``` -meters,location=California.LoSangeles,groupid=2 current=13.4,voltage=223,phase=0.29 1648432611249500 +meters,location=California.LosAngeles,groupid=2 current=13.4,voltage=223,phase=0.29 1648432611249500 ``` :::note @@ -42,7 +42,6 @@ meters,location=California.LoSangeles,groupid=2 current=13.4,voltage=223,phase=0 For more details please refer to [InfluxDB Line Protocol](https://docs.influxdata.com/influxdb/v2.0/reference/syntax/line-protocol/) and [TDengine Schemaless](/reference/schemaless/#Schemaless-Line-Protocol) - ## Examples diff --git a/docs/en/07-develop/03-insert-data/03-opentsdb-telnet.mdx b/docs/en/07-develop/03-insert-data/03-opentsdb-telnet.mdx index 81e1169489d4188e14f4c5338ca322041bba80fb..db9bfd73649709cf806ae6499513191db8321107 100644 --- a/docs/en/07-develop/03-insert-data/03-opentsdb-telnet.mdx +++ b/docs/en/07-develop/03-insert-data/03-opentsdb-telnet.mdx @@ -17,19 +17,19 @@ import CTelnet from "./_c_opts_telnet.mdx"; A single line of text is used in OpenTSDB line protocol to represent one row of data. OpenTSDB employs a single column data model, so each line can only contain a single data column. There can be multiple tags. Each line contains 4 parts as below: -``` +```txt =[ =] ``` - `metric` will be used as the STable name. - `timestamp` is the timestamp of current row of data. The time precision will be determined automatically based on the length of the timestamp. Second and millisecond time precision are supported. -- `value` is a metric which must be a numeric value, the corresponding column name is "_value". +- `value` is a metric which must be a numeric value, The corresponding column name is "value". - The last part is the tag set separated by spaces, all tags will be converted to nchar type automatically. For example: ```txt -meters.current 1648432611250 11.3 location=California.LoSangeles groupid=3 +meters.current 1648432611250 11.3 location=California.LosAngeles groupid=3 ``` Please refer to [OpenTSDB Telnet API](http://opentsdb.net/docs/build/html/api_telnet/put.html) for more details. @@ -63,7 +63,7 @@ Please refer to [OpenTSDB Telnet API](http://opentsdb.net/docs/build/html/api_te taos> use test; Database changed. -taos> show STables; +taos> show stables; name | created_time | columns | tags | tables | ============================================================================================ meters.current | 2022-03-30 17:04:10.877 | 2 | 2 | 2 | @@ -73,8 +73,8 @@ Query OK, 2 row(s) in set (0.002544s) taos> select tbname, * from `meters.current`; tbname | _ts | _value | groupid | location | ================================================================================================================================== - t_0e7bcfa21a02331c06764f275... | 2022-03-28 09:56:51.249 | 10.800000000 | 3 | California.LoSangeles | - t_0e7bcfa21a02331c06764f275... | 2022-03-28 09:56:51.250 | 11.300000000 | 3 | California.LoSangeles | + t_0e7bcfa21a02331c06764f275... | 2022-03-28 09:56:51.249 | 10.800000000 | 3 | California.LosAngeles | + t_0e7bcfa21a02331c06764f275... | 2022-03-28 09:56:51.250 | 11.300000000 | 3 | California.LosAngeles | t_7e7b26dd860280242c6492a16... | 2022-03-28 09:56:51.249 | 10.300000000 | 2 | California.SanFrancisco | t_7e7b26dd860280242c6492a16... | 2022-03-28 09:56:51.250 | 12.600000000 | 2 | California.SanFrancisco | Query OK, 4 row(s) in set (0.005399s) diff --git a/docs/en/07-develop/03-insert-data/04-opentsdb-json.mdx b/docs/en/07-develop/03-insert-data/04-opentsdb-json.mdx index aad94c3d913a128b344757162c231affc6a64651..23703f4087483373a15e9cf7604bb67ca62888f5 100644 --- a/docs/en/07-develop/03-insert-data/04-opentsdb-json.mdx +++ b/docs/en/07-develop/03-insert-data/04-opentsdb-json.mdx @@ -15,37 +15,37 @@ import CJson from "./_c_opts_json.mdx"; ## Introduction -A JSON string is used in OpenTSDB JSON to represent one or more rows of data, for example: +A JSON string is used in OpenTSDB JSON to represent one or more rows of data, for example: For example: ```json [ - { - "metric": "sys.cpu.nice", - "timestamp": 1346846400, - "value": 18, - "tags": { - "host": "web01", - "dc": "lga" - } - }, - { - "metric": "sys.cpu.nice", - "timestamp": 1346846400, - "value": 9, - "tags": { - "host": "web02", - "dc": "lga" - } + { + "metric": "sys.cpu.nice", + "timestamp": 1346846400, + "value": 18, + "tags": { + "host": "web01", + "dc": "lga" } + }, + { + "metric": "sys.cpu.nice", + "timestamp": 1346846400, + "value": 9, + "tags": { + "host": "web02", + "dc": "lga" + } + } ] ``` Similar to OpenTSDB line protocol, `metric` will be used as the STable name, `timestamp` is the timestamp to be used, `value` represents the metric collected, `tags` are the tag sets. - Please refer to [OpenTSDB HTTP API](http://opentsdb.net/docs/build/html/api_http/put.html) for more details. :::note + - In JSON protocol, strings will be converted to nchar type and numeric values will be converted to double type. - Only data in array format is accepted and so an array must be used even if there is only one row. @@ -74,13 +74,13 @@ Please refer to [OpenTSDB HTTP API](http://opentsdb.net/docs/build/html/api_http -The above sample code will created 2 STables automatically while each STable has 2 rows of data. +2 STables will be created automatically and each STable has 2 rows of data in the above sample code. ```cmd taos> use test; Database changed. -taos> show STables; +taos> show stables; name | created_time | columns | tags | tables | ============================================================================================ meters.current | 2022-03-29 16:05:25.193 | 2 | 2 | 1 | diff --git a/docs/en/07-develop/03-insert-data/_go_stmt.mdx b/docs/en/07-develop/03-insert-data/_go_stmt.mdx index ab519c9a806345c2f14337f62c74728da955d2e0..0bdcffc16fa18a1d0afcd50548ed6076f6154838 100644 --- a/docs/en/07-develop/03-insert-data/_go_stmt.mdx +++ b/docs/en/07-develop/03-insert-data/_go_stmt.mdx @@ -3,6 +3,6 @@ ``` :::tip -`github.com/taosdata/driver-go/v2/wrapper` module in driver-go is the wrapper for C API, it can be used to insert data with parameter binding. +`github.com/taosdata/driver-go/v3/wrapper` module in driver-go is the wrapper for C API, it can be used to insert data with parameter binding. ::: diff --git a/docs/en/07-develop/03-insert-data/_php_sql.mdx b/docs/en/07-develop/03-insert-data/_php_sql.mdx new file mode 100644 index 0000000000000000000000000000000000000000..78cd663ec219dabc2eeb81c7e67426eda41d7762 --- /dev/null +++ b/docs/en/07-develop/03-insert-data/_php_sql.mdx @@ -0,0 +1,3 @@ +```php +{{#include docs/examples/php/insert.php}} +``` diff --git a/docs/en/07-develop/03-insert-data/_php_stmt.mdx b/docs/en/07-develop/03-insert-data/_php_stmt.mdx new file mode 100644 index 0000000000000000000000000000000000000000..3bb7b2f8da9887c1063822e69bfdff599aa50b7b --- /dev/null +++ b/docs/en/07-develop/03-insert-data/_php_stmt.mdx @@ -0,0 +1,3 @@ +```php +{{#include docs/examples/php/insert_stmt.php}} +``` diff --git a/docs/en/07-develop/04-query-data/_php.mdx b/docs/en/07-develop/04-query-data/_php.mdx new file mode 100644 index 0000000000000000000000000000000000000000..bcafd1cfbcb1bbb55b03f6fe198e6fa1b5251b19 --- /dev/null +++ b/docs/en/07-develop/04-query-data/_php.mdx @@ -0,0 +1,3 @@ +```go +{{#include docs/examples/php/query.php}} +``` diff --git a/docs/en/07-develop/04-query-data/index.mdx b/docs/en/07-develop/04-query-data/index.mdx index a212fa9529215fc24c55c95a166cfc1a407359b2..38dc98d1ff262c7f8ec4951297e6f42e436682c8 100644 --- a/docs/en/07-develop/04-query-data/index.mdx +++ b/docs/en/07-develop/04-query-data/index.mdx @@ -1,6 +1,5 @@ --- -Sidebar_label: Query data -title: Query data +title: Query Data description: "This chapter introduces major query functionalities and how to perform sync and async query using connectors." --- @@ -13,6 +12,7 @@ import RustQuery from "./_rust.mdx"; import NodeQuery from "./_js.mdx"; import CsQuery from "./_cs.mdx"; import CQuery from "./_c.mdx"; +import PhpQuery from "./_php.mdx"; import PyAsync from "./_py_async.mdx"; import NodeAsync from "./_js_async.mdx"; import CsAsync from "./_cs_async.mdx"; @@ -24,9 +24,8 @@ SQL is used by TDengine as its query language. Application programs can send SQL - Query on single column or multiple columns - Filter on tags or data columns：>, <, =, <\>, like -- Grouping of results: `Group By` -- Sorting of results: `Order By` -- Limit the number of results: `Limit/Offset` +- Grouping of results: `Group By` - Sorting of results: `Order By` - Limit the number of results: `Limit/Offset` +- Windowed aggregate queries for time windows (interval), session windows (session), and state windows (state_window) - Arithmetic on columns of numeric types or aggregate results - Join query with timestamp alignment - Aggregate functions: count, max, min, avg, sum, twa, stddev, leastsquares, top, bottom, first, last, percentile, apercentile, last_row, spread, diff @@ -34,10 +33,6 @@ SQL is used by TDengine as its query language. Application programs can send SQL For example, the SQL statement below can be executed in TDengine CLI `taos` to select records with voltage greater than 215 and limit the output to only 2 rows. ```sql -select * from d1001 where voltage > 215 order by ts desc limit 2; -``` - -```title=Output taos> select * from d1001 where voltage > 215 order by ts desc limit 2; ts | current | voltage | phase | ====================================================================================== @@ -46,89 +41,88 @@ taos> select * from d1001 where voltage > 215 order by ts desc limit 2; Query OK, 2 row(s) in set (0.001100s) ``` -To meet the requirements of varied use cases, some special functions have been added in TDengine. Some examples are `twa` (Time Weighted Average), `spread` (The difference between the maximum and the minimum), and `last_row` (the last row). Furthermore, continuous query is also supported in TDengine. +To meet the requirements of varied use cases, some special functions have been added in TDengine. Some examples are `twa` (Time Weighted Average), `spread` (The difference between the maximum and the minimum), and `last_row` (the last row). -For detailed query syntax please refer to [Select](/taos-sql/select). +For detailed query syntax, see [Select](../../taos-sql/select). ## Aggregation among Tables In most use cases, there are always multiple kinds of data collection points. A new concept, called STable (abbreviation for super table), is used in TDengine to represent one type of data collection point, and a subtable is used to represent a specific data collection point of that type. Tags are used by TDengine to represent the static properties of data collection points. A specific data collection point has its own values for static properties. By specifying filter conditions on tags, aggregation can be performed efficiently among all the subtables created via the same STable, i.e. same type of data collection points. Aggregate functions applicable for tables can be used directly on STables; the syntax is exactly the same. -In summary, records across subtables can be aggregated by a simple query on their STable. It is like a join operation. However, tables belonging to different STables can not be aggregated. - ### Example 1 In TDengine CLI `taos`, use the SQL below to get the average voltage of all the meters in California grouped by location. ``` -taos> SELECT AVG(voltage) FROM meters GROUP BY location; - avg(voltage) | location | -============================================================= - 222.000000000 | California.LosAngeles | - 219.200000000 | California.SanFrancisco | -Query OK, 2 row(s) in set (0.002136s) +taos> SELECT AVG(voltage), location FROM meters GROUP BY location; + avg(voltage) | location | +=============================================================================================== + 219.200000000 | California.SanFrancisco | + 221.666666667 | California.LosAngeles | +Query OK, 2 rows in database (0.005995s) ``` ### Example 2 -In TDengine CLI `taos`, use the SQL below to get the number of rows and the maximum current in the past 24 hours from meters whose groupId is 2. +In TDengine CLI `taos`, use the SQL below to get the number of rows and the maximum current from meters whose groupId is 2. ``` -taos> SELECT count(*), max(current) FROM meters where groupId = 2 and ts > now - 24h; +taos> SELECT count(*), max(current) FROM meters where groupId = 2; count(*) | max(current) | ================================== 5 | 13.4 | Query OK, 1 row(s) in set (0.002136s) ``` -Join queries are only allowed between subtables of the same STable. In [Select](/taos-sql/select), all query operations are marked as to whether they support STables or not. +In [Select](../../taos-sql/select), all query operations are marked as to whether they support STables or not. ## Down Sampling and Interpolation In IoT use cases, down sampling is widely used to aggregate data by time range. The `INTERVAL` keyword in TDengine can be used to simplify the query by time window. For example, the SQL statement below can be used to get the sum of current every 10 seconds from meters table d1001. ``` -taos> SELECT sum(current) FROM d1001 INTERVAL(10s); - ts | sum(current) | +taos> SELECT _wstart, sum(current) FROM d1001 INTERVAL(10s); + _wstart | sum(current) | ====================================================== 2018-10-03 14:38:00.000 | 10.300000191 | 2018-10-03 14:38:10.000 | 24.900000572 | -Query OK, 2 row(s) in set (0.000883s) +Query OK, 2 rows in database (0.003139s) ``` Down sampling can also be used for STable. For example, the below SQL statement can be used to get the sum of current from all meters in California. ``` -taos> SELECT SUM(current) FROM meters where location like "California%" INTERVAL(1s); - ts | sum(current) | +taos> SELECT _wstart, SUM(current) FROM meters where location like "California%" INTERVAL(1s); + _wstart | sum(current) | ====================================================== 2018-10-03 14:38:04.000 | 10.199999809 | - 2018-10-03 14:38:05.000 | 32.900000572 | + 2018-10-03 14:38:05.000 | 23.699999809 | 2018-10-03 14:38:06.000 | 11.500000000 | 2018-10-03 14:38:15.000 | 12.600000381 | - 2018-10-03 14:38:16.000 | 36.000000000 | -Query OK, 5 row(s) in set (0.001538s) + 2018-10-03 14:38:16.000 | 34.400000572 | +Query OK, 5 rows in database (0.007413s) ``` Down sampling also supports time offset. For example, the below SQL statement can be used to get the sum of current from all meters but each time window must start at the boundary of 500 milliseconds. ``` -taos> SELECT SUM(current) FROM meters INTERVAL(1s, 500a); - ts | sum(current) | +taos> SELECT _wstart, SUM(current) FROM meters INTERVAL(1s, 500a); + _wstart | sum(current) | ====================================================== - 2018-10-03 14:38:04.500 | 11.189999809 | - 2018-10-03 14:38:05.500 | 31.900000572 | - 2018-10-03 14:38:06.500 | 11.600000000 | - 2018-10-03 14:38:15.500 | 12.300000381 | - 2018-10-03 14:38:16.500 | 35.000000000 | -Query OK, 5 row(s) in set (0.001521s) + 2018-10-03 14:38:03.500 | 10.199999809 | + 2018-10-03 14:38:04.500 | 10.300000191 | + 2018-10-03 14:38:05.500 | 13.399999619 | + 2018-10-03 14:38:06.500 | 11.500000000 | + 2018-10-03 14:38:14.500 | 12.600000381 | + 2018-10-03 14:38:16.500 | 34.400000572 | +Query OK, 6 rows in database (0.005515s) ``` In many use cases, it's hard to align the timestamp of the data collected by each collection point. However, a lot of algorithms like FFT require the data to be aligned with same time interval and application programs have to handle this by themselves. In TDengine, it's easy to achieve the alignment using down sampling. Interpolation can be performed in TDengine if there is no data in a time range. -For more details please refer to [Aggregate by Window](/taos-sql/interval). +For more information, see [Aggregate by Window](../../taos-sql/distinguished). ## Examples @@ -158,6 +152,9 @@ In the section describing [Insert](/develop/insert-data/sql-writing), a database + + + :::note diff --git a/docs/en/07-develop/06-stream.md b/docs/en/07-develop/06-stream.md new file mode 100644 index 0000000000000000000000000000000000000000..36f903ee9a4f2d210e63d0b79e702bc199f790ed --- /dev/null +++ b/docs/en/07-develop/06-stream.md @@ -0,0 +1,113 @@ +--- +sidebar_label: Stream Processing +description: "The TDengine stream processing engine combines data inserts, preprocessing, analytics, real-time computation, and alerting into a single component." +title: Stream Processing +--- + +Raw time-series data is often cleaned and preprocessed before being permanently stored in a database. In a traditional time-series solution, this generally requires the deployment of stream processing systems such as Kafka or Flink. However, the complexity of such systems increases the cost of development and maintenance. + +With the stream processing engine built into TDengine, you can process incoming data streams in real time and define stream transformations in SQL. Incoming data is automatically processed, and the results are pushed to specified tables based on triggering rules that you define. This is a lightweight alternative to complex processing engines that returns computation results in milliseconds even in high throughput scenarios. + +The stream processing engine includes data filtering, scalar function computation (including user-defined functions), and window aggregation, with support for sliding windows, session windows, and event windows. Stream processing can write data to supertables from other supertables, standard tables, or subtables. When you create a stream, the target supertable is automatically created. New data is then processed and written to that supertable according to the rules defined for the stream. You can use PARTITION BY statements to partition the data by table name or tag. Separate partitions are then written to different subtables within the target supertable. + +TDengine stream processing supports the aggregation of supertables that are deployed across multiple vnodes. It can also handle out-of-order writes and includes a watermark mechanism that determines the extent to which out-of-order data is accepted by the system. You can configure whether to drop or reprocess out-of-order data through the **ignore expired** parameter. + +For more information, see [Stream Processing](../../taos-sql/stream). + + +## Create a Stream + +```sql +CREATE STREAM [IF NOT EXISTS] stream_name [stream_options] INTO stb_name AS subquery +stream_options: { + TRIGGER [AT_ONCE | WINDOW_CLOSE | MAX_DELAY time] + WATERMARK time + IGNORE EXPIRED [0 | 1] +} +``` + +For more information, see [Stream Processing](../../taos-sql/stream). + +## Usage Scenario 1 + +It is common that smart electrical meter systems for businesses generate millions of data points that are widely dispersed and not ordered. The time required to clean and convert this data makes efficient, real-time processing impossible for traditional solutions. This scenario shows how you can configure TDengine stream processing to drop data points over 220 V, find the maximum voltage for 5 second windows, and output this data to a table. + +### Create a Database for Raw Data + +A database including one supertable and four subtables is created as follows: + +```sql +DROP DATABASE IF EXISTS power; +CREATE DATABASE power; +USE power; + +CREATE STABLE meters (ts timestamp, current float, voltage int, phase float) TAGS (location binary(64), groupId int); + +CREATE TABLE d1001 USING meters TAGS ("California.SanFrancisco", 2); +CREATE TABLE d1002 USING meters TAGS ("California.SanFrancisco", 3); +CREATE TABLE d1003 USING meters TAGS ("California.LosAngeles", 2); +CREATE TABLE d1004 USING meters TAGS ("California.LosAngeles", 3); +``` + +### Create a Stream + +```sql +create stream current_stream into current_stream_output_stb as select _wstart as start, _wend as end, max(current) as max_current from meters where voltage <= 220 interval (5s); +``` + +### Write Data +```sql +insert into d1001 values("2018-10-03 14:38:05.000", 10.30000, 219, 0.31000); +insert into d1001 values("2018-10-03 14:38:15.000", 12.60000, 218, 0.33000); +insert into d1001 values("2018-10-03 14:38:16.800", 12.30000, 221, 0.31000); +insert into d1002 values("2018-10-03 14:38:16.650", 10.30000, 218, 0.25000); +insert into d1003 values("2018-10-03 14:38:05.500", 11.80000, 221, 0.28000); +insert into d1003 values("2018-10-03 14:38:16.600", 13.40000, 223, 0.29000); +insert into d1004 values("2018-10-03 14:38:05.000", 10.80000, 223, 0.29000); +insert into d1004 values("2018-10-03 14:38:06.500", 11.50000, 221, 0.35000); +``` + +### Query the Results + +```sql +taos> select start, end, max_current from current_stream_output_stb; + start | end | max_current | +=========================================================================== + 2018-10-03 14:38:05.000 | 2018-10-03 14:38:10.000 | 10.30000 | + 2018-10-03 14:38:15.000 | 2018-10-03 14:38:20.000 | 12.60000 | +Query OK, 2 rows in database (0.018762s) +``` + +## Usage Scenario 2 + +In this scenario, the active power and reactive power are determined from the data gathered in the previous scenario. The location and name of each meter are concatenated with a period (.) between them, and the data set is partitioned by meter name and written to a new database. + +### Create a Database for Raw Data + +The procedure from the previous scenario is used to create the database. + +### Create a Stream + +```sql +create stream power_stream into power_stream_output_stb as select ts, concat_ws(".", location, tbname) as meter_location, current*voltage*cos(phase) as active_power, current*voltage*sin(phase) as reactive_power from meters partition by tbname; +``` + +### Write data + +The procedure from the previous scenario is used to write the data. + +### Query the Results +```sql +taos> select ts, meter_location, active_power, reactive_power from power_stream_output_stb; + ts | meter_location | active_power | reactive_power | +=================================================================================================================== + 2018-10-03 14:38:05.000 | California.LosAngeles.d1004 | 2307.834596289 | 688.687331847 | + 2018-10-03 14:38:06.500 | California.LosAngeles.d1004 | 2387.415754896 | 871.474763418 | + 2018-10-03 14:38:05.500 | California.LosAngeles.d1003 | 2506.240411679 | 720.680274962 | + 2018-10-03 14:38:16.600 | California.LosAngeles.d1003 | 2863.424274422 | 854.482390839 | + 2018-10-03 14:38:05.000 | California.SanFrancisco.d1001 | 2148.178871730 | 688.120784090 | + 2018-10-03 14:38:15.000 | California.SanFrancisco.d1001 | 2598.589176205 | 890.081451418 | + 2018-10-03 14:38:16.800 | California.SanFrancisco.d1001 | 2588.728381186 | 829.240910475 | + 2018-10-03 14:38:16.650 | California.SanFrancisco.d1002 | 2175.595991997 | 555.520860397 | +Query OK, 8 rows in database (0.014753s) +``` diff --git a/docs/en/07-develop/07-tmq.mdx b/docs/en/07-develop/07-tmq.mdx new file mode 100644 index 0000000000000000000000000000000000000000..ceeea64fca91473ea62de404fb9e92c179f7e6d4 --- /dev/null +++ b/docs/en/07-develop/07-tmq.mdx @@ -0,0 +1,841 @@ +--- +sidebar_label: Data Subscription +description: "The TDengine data subscription service automatically pushes data written in TDengine to subscribing clients." +title: Data Subscription +--- + +import Tabs from "@theme/Tabs"; +import TabItem from "@theme/TabItem"; +import Java from "./_sub_java.mdx"; +import Python from "./_sub_python.mdx"; +import Go from "./_sub_go.mdx"; +import Rust from "./_sub_rust.mdx"; +import Node from "./_sub_node.mdx"; +import CSharp from "./_sub_cs.mdx"; +import CDemo from "./_sub_c.mdx"; + +TDengine provides data subscription and consumption interfaces similar to message queue products. These interfaces make it easier for applications to obtain data written to TDengine either in real time and to process data in the order that events occurred. This simplifies your time-series data processing systems and reduces your costs because it is no longer necessary to deploy a message queue product such as Kafka. + +To use TDengine data subscription, you define topics like in Kafka. However, a topic in TDengine is based on query conditions for an existing supertable, standard table, or subtable - in other words, a SELECT statement. You can use SQL to filter data by tag, table name, column, or expression and then perform a scalar function or user-defined function on the data. Aggregate functions are not supported. This gives TDengine data subscription more flexibility than similar products. The granularity of data can be controlled on demand by applications, while filtering and preprocessing are handled by TDengine instead of the application layer. This implementation reduces the amount of data transmitted and the complexity of applications. + +By subscribing to a topic, a consumer can obtain the latest data in that topic in real time. Multiple consumers can be formed into a consumer group that consumes messages together. Consumer groups enable faster speed through multi-threaded, distributed data consumption. Note that consumers in different groups that are subscribed to the same topic do not consume messages together. A single consumer can subscribe to multiple topics. If the data in a supertable is sharded across multiple vnodes, consumer groups can consume it much more efficiently than single consumers. TDengine also includes an acknowledgement mechanism that ensures at-least-once delivery in complicated environments where machines may crash or restart. + +To implement these features, TDengine indexes its write-ahead log (WAL) file for fast random access and provides configurable methods for replacing and retaining this file. You can define a retention period and size for this file. For information, see the CREATE DATABASE statement. In this way, the WAL file is transformed into a persistent storage engine that remembers the order in which events occur. However, note that configuring an overly long retention period for your WAL files makes database compression inefficient. TDengine then uses the WAL file instead of the time-series database as its storage engine for queries in the form of topics. TDengine reads the data from the WAL file; uses a unified query engine instance to perform filtering, transformations, and other operations; and finally pushes the data to consumers. + + + +## Data Schema and API + +The related schemas and APIs in various languages are described as follows: + + + + +```c +typedef struct tmq_t tmq_t; +typedef struct tmq_conf_t tmq_conf_t; +typedef struct tmq_list_t tmq_list_t; + +typedef void(tmq_commit_cb(tmq_t *, int32_t code, void *param)); + +DLL_EXPORT tmq_list_t *tmq_list_new(); +DLL_EXPORT int32_t tmq_list_append(tmq_list_t *, const char *); +DLL_EXPORT void tmq_list_destroy(tmq_list_t *); +DLL_EXPORT tmq_t *tmq_consumer_new(tmq_conf_t *conf, char *errstr, int32_t errstrLen); +DLL_EXPORT const char *tmq_err2str(int32_t code); + +DLL_EXPORT int32_t tmq_subscribe(tmq_t *tmq, const tmq_list_t *topic_list); +DLL_EXPORT int32_t tmq_unsubscribe(tmq_t *tmq); +DLL_EXPORT TAOS_RES *tmq_consumer_poll(tmq_t *tmq, int64_t timeout); +DLL_EXPORT int32_t tmq_consumer_close(tmq_t *tmq); +DLL_EXPORT int32_t tmq_commit_sync(tmq_t *tmq, const TAOS_RES *msg); +DLL_EXPORT void tmq_commit_async(tmq_t *tmq, const TAOS_RES *msg, tmq_commit_cb *cb, void *param); + +enum tmq_conf_res_t { + TMQ_CONF_UNKNOWN = -2, + TMQ_CONF_INVALID = -1, + TMQ_CONF_OK = 0, +}; +typedef enum tmq_conf_res_t tmq_conf_res_t; + +DLL_EXPORT tmq_conf_t *tmq_conf_new(); +DLL_EXPORT tmq_conf_res_t tmq_conf_set(tmq_conf_t *conf, const char *key, const char *value); +DLL_EXPORT void tmq_conf_destroy(tmq_conf_t *conf); +DLL_EXPORT void tmq_conf_set_auto_commit_cb(tmq_conf_t *conf, tmq_commit_cb *cb, void *param); +``` + +For more information, see [C/C++ Connector](/reference/connector/cpp). + +The following example is based on the smart meter table described in Data Models. For complete sample code, see the C language section below. + + + + +```java +void subscribe(Collection topics) throws SQLException; + +void unsubscribe() throws SQLException; + +Set subscription() throws SQLException; + +ConsumerRecords poll(Duration timeout) throws SQLException; + +void commitAsync(); + +void commitAsync(OffsetCommitCallback callback); + +void commitSync() throws SQLException; + +void close() throws SQLException; +``` + + + + + +```python +class TaosConsumer(): + def __init__(self, *topics, **configs) + + def __iter__(self) + + def __next__(self) + + def sync_next(self) + + def subscription(self) + + def unsubscribe(self) + + def close(self) + + def __del__(self) +``` + + + + + +```go +func NewConsumer(conf *Config) (*Consumer, error) + +func (c *Consumer) Close() error + +func (c *Consumer) Commit(ctx context.Context, message unsafe.Pointer) error + +func (c *Consumer) FreeMessage(message unsafe.Pointer) + +func (c *Consumer) Poll(timeout time.Duration) (*Result, error) + +func (c *Consumer) Subscribe(topics []string) error + +func (c *Consumer) Unsubscribe() error +``` + + + + + +```rust +impl TBuilder for TmqBuilder + fn from_dsn(dsn: D) -> Result + fn build(&self) -> Result + +impl AsAsyncConsumer for Consumer + async fn subscribe, I: IntoIterator + Send>( + &mut self, + topics: I, + ) -> Result<(), Self::Error>; + fn stream( + &self, + ) -> Pin< + Box< + dyn '_ + + Send + + futures::Stream< + Item = Result<(Self::Offset, MessageSet), Self::Error>, + >, + >, + >; + async fn commit(&self, offset: Self::Offset) -> Result<(), Self::Error>; + + async fn unsubscribe(self); +``` + +For more information, see [Crate taos](https://docs.rs/taos). + + + + + +```js +function TMQConsumer(config) + +function subscribe(topic) + +function consume(timeout) + +function subscription() + +function unsubscribe() + +function commit(msg) + +function close() +``` + + + + + +```csharp +ConsumerBuilder(IEnumerable> config) + +virtual IConsumer Build() + +Consumer(ConsumerBuilder builder) + +void Subscribe(IEnumerable topics) + +void Subscribe(string topic) + +ConsumeResult Consume(int millisecondsTimeout) + +List Subscription() + +void Unsubscribe() + +void Commit(ConsumeResult consumerResult) + +void Close() +``` + + + + +## Insert Data into TDengine + +A database including one supertable and two subtables is created as follows: + +```sql +DROP DATABASE IF EXISTS tmqdb; +CREATE DATABASE tmqdb; +CREATE TABLE tmqdb.stb (ts TIMESTAMP, c1 INT, c2 FLOAT, c3 VARCHAR(16) TAGS(t1 INT, t3 VARCHAR(16)); +CREATE TABLE tmqdb.ctb0 USING tmqdb.stb TAGS(0, "subtable0"); +CREATE TABLE tmqdb.ctb1 USING tmqdb.stb TAGS(1, "subtable1"); +INSERT INTO tmqdb.ctb0 VALUES(now, 0, 0, 'a0')(now+1s, 0, 0, 'a00'); +INSERT INTO tmqdb.ctb1 VALUES(now, 1, 1, 'a1')(now+1s, 11, 11, 'a11'); +``` + +## Create a Topic + +The following SQL statement creates a topic in TDengine: + +```sql +CREATE TOPIC topic_name AS SELECT ts, c1, c2, c3 FROM tmqdb.stb WHERE c1 > 1; +``` + +Multiple subscription types are supported. + +#### Subscribe to a Column + +Syntax: + +```sql +CREATE TOPIC topic_name as subquery +``` + +You can subscribe to a topic through a SELECT statement. Statements that specify columns, such as `SELECT *` and `SELECT ts, cl` are supported, as are filtering conditions and scalar functions. Aggregate functions and time window aggregation are not supported. Note: + +- The schema of topics created in this manner is determined by the subscribed data. +- You cannot modify (`ALTER
MODIFY`) or delete (`ALTER
DROP`) columns or tags that are used in a subscription or calculation. +- Columns added to a table after the subscription is created are not displayed in the results. Deleting columns will cause an error. + +### Subscribe to a Supertable + +Syntax: + +```sql +CREATE TOPIC topic_name AS STABLE stb_name +``` + +Creating a topic in this manner differs from a `SELECT * from stbName` statement as follows: + +- The table schema can be modified. +- Unstructured data is returned. The format of the data returned changes based on the supertable schema. +- A different table schema may exist for every data block to be processed. +- The data returned does not include tags. + +### Subscribe to a Database + +Syntax: + +```sql +CREATE TOPIC topic_name [WITH META] AS DATABASE db_name; +``` + +This SQL statement creates a subscription to all tables in the database. You can add the `WITH META` parameter to include schema changes in the subscription, including creating and deleting supertables; adding, deleting, and modifying columns; and creating, deleting, and modifying the tags of subtables. Consumers can determine the message type from the API. Note that this differs from Kafka. + +## Create a Consumer + +You configure the following parameters when creating a consumer: + +| Parameter | Type | Description | Remarks | +| :----------------------------: | :-----: | -------------------------------------------------------- | ------------------------------------------- | +| `td.connect.ip` | string | Used in establishing a connection; same as `taos_connect` | | +| `td.connect.user` | string | Used in establishing a connection; same as `taos_connect` | | +| `td.connect.pass` | string | Used in establishing a connection; same as `taos_connect` | | +| `td.connect.port` | string | Used in establishing a connection; same as `taos_connect` | | +| `group.id` | string | Consumer group ID; consumers with the same ID are in the same group | **Required**. Maximum length: 192. | +| `client.id` | string | Client ID | Maximum length: 192. | +| `auto.offset.reset` | enum | Initial offset for the consumer group | Specify `earliest`, `latest`, or `none`(default) | +| `enable.auto.commit` | boolean | Commit automatically | Specify `true` or `false`. | +| `auto.commit.interval.ms` | integer | Interval for automatic commits, in milliseconds | +| `enable.heartbeat.background` | boolean | Backend heartbeat; if enabled, the consumer does not go offline even if it has not polled for a long time | | +| `experimental.snapshot.enable` | boolean | Specify whether to consume messages from the WAL or from TSBS | | +| `msg.with.table.name` | boolean | Specify whether to deserialize table names from messages | + +The method of specifying these parameters depends on the language used: + + + + +```c +/* Create consumer groups on demand (group.id) and enable automatic commits (enable.auto.commit), + an automatic commit interval (auto.commit.interval.ms), and a username (td.connect.user) and password (td.connect.pass) */ +tmq_conf_t* conf = tmq_conf_new(); +tmq_conf_set(conf, "enable.auto.commit", "true"); +tmq_conf_set(conf, "auto.commit.interval.ms", "1000"); +tmq_conf_set(conf, "group.id", "cgrpName"); +tmq_conf_set(conf, "td.connect.user", "root"); +tmq_conf_set(conf, "td.connect.pass", "taosdata"); +tmq_conf_set(conf, "auto.offset.reset", "earliest"); +tmq_conf_set(conf, "experimental.snapshot.enable", "true"); +tmq_conf_set(conf, "msg.with.table.name", "true"); +tmq_conf_set_auto_commit_cb(conf, tmq_commit_cb_print, NULL); + +tmq_t* tmq = tmq_consumer_new(conf, NULL, 0); +tmq_conf_destroy(conf); +``` + + + + +Java programs use the following parameters: + +| Parameter | Type | Description | Remarks | +| ----------------------------- | ------ | ----------------------------------------------------------------------------------------------------------------------------- | +| `bootstrap.servers` | string |Connection address, such as `localhost:6030` | +| `value.deserializer` | string | Value deserializer; to use this method, implement the `com.taosdata.jdbc.tmq.Deserializer` interface or inherit the `com.taosdata.jdbc.tmq.ReferenceDeserializer` type | +| `value.deserializer.encoding` | string | Specify the encoding for string deserialization | | + +Note: The `bootstrap.servers` parameter is used instead of `td.connect.ip` and `td.connect.port` to provide an interface that is consistent with Kafka. + +```java +Properties properties = new Properties(); +properties.setProperty("enable.auto.commit", "true"); +properties.setProperty("auto.commit.interval.ms", "1000"); +properties.setProperty("group.id", "cgrpName"); +properties.setProperty("bootstrap.servers", "127.0.0.1:6030"); +properties.setProperty("td.connect.user", "root"); +properties.setProperty("td.connect.pass", "taosdata"); +properties.setProperty("auto.offset.reset", "earliest"); +properties.setProperty("msg.with.table.name", "true"); +properties.setProperty("value.deserializer", "com.taos.example.MetersDeserializer"); + +TaosConsumer consumer = new TaosConsumer<>(properties); + +/* value deserializer definition. */ +import com.taosdata.jdbc.tmq.ReferenceDeserializer; + +public class MetersDeserializer extends ReferenceDeserializer { +} +``` + + + + + +```go +config := tmq.NewConfig() +defer config.Destroy() +err = config.SetGroupID("test") +if err != nil { + panic(err) +} +err = config.SetAutoOffsetReset("earliest") +if err != nil { + panic(err) +} +err = config.SetConnectIP("127.0.0.1") +if err != nil { + panic(err) +} +err = config.SetConnectUser("root") +if err != nil { + panic(err) +} +err = config.SetConnectPass("taosdata") +if err != nil { + panic(err) +} +err = config.SetConnectPort("6030") +if err != nil { + panic(err) +} +err = config.SetMsgWithTableName(true) +if err != nil { + panic(err) +} +err = config.EnableHeartBeat() +if err != nil { + panic(err) +} +err = config.EnableAutoCommit(func(result *wrapper.TMQCommitCallbackResult) { + if result.ErrCode != 0 { + errStr := wrapper.TMQErr2Str(result.ErrCode) + err := errors.NewError(int(result.ErrCode), errStr) + panic(err) + } +}) +if err != nil { + panic(err) +} +``` + + + + + +```rust +let mut dsn: Dsn = "taos://".parse()?; +dsn.set("group.id", "group1"); +dsn.set("client.id", "test"); +dsn.set("auto.offset.reset", "earliest"); + +let tmq = TmqBuilder::from_dsn(dsn)?; + +let mut consumer = tmq.build()?; +``` + + + + + +Python programs use the following parameters: + +| Parameter | Type | Description | Remarks | +| :----------------------------: | :----: | -------------------------------------------------------- | ------------------------------------------- | +| `td_connect_ip` | string | Used in establishing a connection; same as `taos_connect` | | +| `td_connect_user` | string | Used in establishing a connection; same as `taos_connect` | | +| `td_connect_pass` | string | Used in establishing a connection; same as `taos_connect` | | +| `td_connect_port` | string | Used in establishing a connection; same as `taos_connect` | | +| `group_id` | string | Consumer group ID; consumers with the same ID are in the same group | **Required**. Maximum length: 192. | +| `client_id` | string | Client ID | Maximum length: 192. | +| `auto_offset_reset` | string | Initial offset for the consumer group | Specify `earliest`, `latest`, or `none`(default) | +| `enable_auto_commit` | string | Commit automatically | Specify `true` or `false`. | +| `auto_commit_interval_ms` | string | Interval for automatic commits, in milliseconds | +| `enable_heartbeat_background` | string | Backend heartbeat; if enabled, the consumer does not go offline even if it has not polled for a long time | Specify `true` or `false`. | +| `experimental_snapshot_enable` | string | Specify whether to consume messages from the WAL or from TSBS | Specify `true` or `false`. | +| `msg_with_table_name` | string | Specify whether to deserialize table names from messages | Specify `true` or `false`. +| `timeout` | int | Consumer pull timeout | | + + + + + +```js +// Create consumer groups on demand (group.id) and enable automatic commits (enable.auto.commit), +// an automatic commit interval (auto.commit.interval.ms), and a username (td.connect.user) and password (td.connect.pass) + +let consumer = taos.consumer({ + 'enable.auto.commit': 'true', + 'auto.commit.interval.ms','1000', + 'group.id': 'tg2', + 'td.connect.user': 'root', + 'td.connect.pass': 'taosdata', + 'auto.offset.reset','earliest', + 'msg.with.table.name': 'true', + 'td.connect.ip','127.0.0.1', + 'td.connect.port','6030' + }); +``` + + + + + +```csharp +using TDengineTMQ; + +// Create consumer groups on demand (GourpID) and enable automatic commits (EnableAutoCommit), +// an automatic commit interval (AutoCommitIntervalMs), and a username (TDConnectUser) and password (TDConnectPasswd) +var cfg = new ConsumerConfig + { + EnableAutoCommit = "true" + AutoCommitIntervalMs = "1000" + GourpId = "TDengine-TMQ-C#", + TDConnectUser = "root", + TDConnectPasswd = "taosdata", + AutoOffsetReset = "earliest" + MsgWithTableName = "true", + TDConnectIp = "127.0.0.1", + TDConnectPort = "6030" + }; + +var consumer = new ConsumerBuilder(cfg).Build(); + +``` + + + + + +A consumer group is automatically created when multiple consumers are configured with the same consumer group ID. + +## Subscribe to a Topic + +A single consumer can subscribe to multiple topics. + + + + +```c +// Create a list of subscribed topics +tmq_list_t* topicList = tmq_list_new(); +tmq_list_append(topicList, "topicName"); +// Enable subscription +tmq_subscribe(tmq, topicList); +tmq_list_destroy(topicList); + +``` + + + + +```java +List topics = new ArrayList<>(); +topics.add("tmq_topic"); +consumer.subscribe(topics); +``` + + + + +```go +consumer, err := tmq.NewConsumer(config) +if err != nil { + panic(err) +} +err = consumer.Subscribe([]string{"example_tmq_topic"}) +if err != nil { + panic(err) +} +``` + + + + +```rust +consumer.subscribe(["tmq_meters"]).await?; +``` + + + + + +```python +consumer = TaosConsumer('topic_ctb_column', group_id='vg2') +``` + + + + + +```js +// Create a list of subscribed topics +let topics = ['topic_test'] + +// Enable subscription +consumer.subscribe(topics); +``` + + + + + +```csharp +// Create a list of subscribed topics +List topics = new List(); +topics.add("tmq_topic"); +// Enable subscription +consumer.Subscribe(topics); +``` + + + + + +## Consume messages + +The following code demonstrates how to consume the messages in a queue. + + + + +```c +## Consume data +while (running) { + TAOS_RES* msg = tmq_consumer_poll(tmq, timeOut); + msg_process(msg); +} +``` + +The `while` loop obtains a message each time it calls `tmq_consumer_poll()`. This message is exactly the same as the result returned by a query, and the same deserialization API can be used on it. + + + + +```java +while(running){ + ConsumerRecords meters = consumer.poll(Duration.ofMillis(100)); + for (Meters meter : meters) { + processMsg(meter); + } +} +``` + + + + + +```go +for { + result, err := consumer.Poll(time.Second) + if err != nil { + panic(err) + } + fmt.Println(result) + consumer.Commit(context.Background(), result.Message) + consumer.FreeMessage(result.Message) +} +``` + + + + + +```rust +{ + let mut stream = consumer.stream(); + + while let Some((offset, message)) = stream.try_next().await? { + // get information from offset + + // the topic + let topic = offset.topic(); + // the vgroup id, like partition id in kafka. + let vgroup_id = offset.vgroup_id(); + println!("* in vgroup id {vgroup_id} of topic {topic}\n"); + + if let Some(data) = message.into_data() { + while let Some(block) = data.fetch_raw_block().await? { + // one block for one table, get table name if needed + let name = block.table_name(); + let records: Vec = block.deserialize().try_collect()?; + println!( + "** table: {}, got {} records: {:#?}\n", + name.unwrap(), + records.len(), + records + ); + } + } + consumer.commit(offset).await?; + } +} +``` + + + + +```python +for msg in consumer: + for row in msg: + print(row) +``` + + + + + +```js +while(true){ + msg = consumer.consume(200); + // process message(consumeResult) + console.log(msg.topicPartition); + console.log(msg.block); + console.log(msg.fields) +} +``` + + + + + +```csharp +## Consume data +while (true) +{ + var consumerRes = consumer.Consume(100); + // process ConsumeResult + ProcessMsg(consumerRes); + consumer.Commit(consumerRes); +} +``` + + + + + +## Close the consumer + +After message consumption is finished, the consumer is unsubscribed. + + + + +```c +/* Unsubscribe */ +tmq_unsubscribe(tmq); + +/* Close consumer object */ +tmq_consumer_close(tmq); +``` + + + + +```java +/* Unsubscribe */ +consumer.unsubscribe(); + +/* Close consumer */ +consumer.close(); +``` + + + + + +```go +consumer.Close() +``` + + + + + +```rust +consumer.unsubscribe().await; +``` + + + + + +```py +# Unsubscribe +consumer.unsubscribe() +# Close consumer +consumer.close() +``` + + + + +```js +consumer.unsubscribe(); +consumer.close(); +``` + + + + + +```csharp +// Unsubscribe +consumer.Unsubscribe(); + +// Close consumer +consumer.Close(); +``` + + + + + +## Delete a Topic + +You can delete topics that are no longer useful. Note that you must unsubscribe all consumers from a topic before deleting it. + +```sql +/* Delete topic/ +DROP TOPIC topic_name; +``` + +## Check Status + +1. Query all existing topics. + +```sql +SHOW TOPICS; +``` + +2. Query the status and subscribed topics of all consumers. + +```sql +SHOW CONSUMERS; +``` + +3. Query the relationships between consumers and vgroups. + +```sql +SHOW SUBSCRIPTIONS; +``` + +## Examples + +The following section shows sample code in various languages. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + diff --git a/docs/en/07-develop/08-cache.md b/docs/en/07-develop/08-cache.md index 8e86eff7414a02ad36a965eb092b8b9b65343301..4892c21c9ddb97b3f967053ee64be24f8cb78c85 100644 --- a/docs/en/07-develop/08-cache.md +++ b/docs/en/07-develop/08-cache.md @@ -1,52 +1,49 @@ --- -sidebar_label: Cache -title: Cache -description: "Caching System inside TDengine" +sidebar_label: Caching +title: Caching +description: "This document describes the caching component of TDengine." --- -To achieve the purpose of high performance data writing and querying, TDengine employs a lot of caching technologies in both server side and client side. +TDengine uses various kinds of caching techniques to efficiently write and query data. This document describes the caching component of TDengine. ## Write Cache -The cache management policy in TDengine is First-In-First-Out (FIFO). FIFO is also known as insert driven cache management policy and it is different from read driven cache management, which is more commonly known as Least-Recently-Used (LRU). FIFO simply stores the latest data in cache and flushes the oldest data in cache to disk, when the cache usage reaches a threshold. In IoT use cases, it is the current state i.e. the latest or most recent data that is important. The cache policy in TDengine, like much of the design and architecture of TDengine, is based on the nature of IoT data. +TDengine uses an insert-driven cache management policy, known as first in, first out (FIFO). This policy differs from read-driven "least recently used (LRU)" cache management. A FIFO policy stores the latest data in cache and flushes the oldest data from cache to disk when the cache usage reaches a threshold. In IoT use cases, the most recent data or the current state is most important. The cache policy in TDengine, like much of the design and architecture of TDengine, is based on the nature of IoT data. -The memory space used by each vnode as write cache is determined when creating a database. Parameter `vgroups` and `buffer` can be used to specify the number of vnode and the size of write cache for each vnode when creating the database. Then, the total size of write cache for this database is `vgroups * buffer`. +When you create a database, you can configure the size of the write cache on each vnode. The **vgroups** parameter determines the number of vgroups that process data in the database, and the **buffer** parameter determines the size of the write cache for each vnode. ```sql create database db0 vgroups 100 buffer 16MB ``` -The above statement creates a database of 100 vnodes while each vnode has a write cache of 16MB. - -Even though in theory it's always better to have a larger cache, the extra effect would be very minor once the size of cache grows beyond a threshold. So normally it's enough to use the default value of `buffer` parameter. +In theory, larger cache sizes are always better. However, at a certain point, it becomes impossible to improve performance by increasing cache size. In most scenarios, you can retain the default cache settings. ## Read Cache -When creating a database, it's also possible to specify whether to cache the latest data of each sub table, using parameter `cachelast`. There are 3 cases: -- 0: No cache for latest data -- 1: The last row of each table is cached, `last_row` function can benefit significantly from it -- 2: The latest non-NULL value of each column for each table is cached, `last` function can benefit very much when there is no `where`, `group by`, `order by` or `interval` clause -- 3: Bot hthe last row and the latest non-NULL value of each column for each table are cached, identical to the behavior of both 1 and 2 are set together - +When you create a database, you can configure whether the latest data from every subtable is cached. To do so, set the *cachelast* parameter as follows: +- 0: Caching is disabled. +- 1: The latest row of data in each subtable is cached. This option significantly improves the performance of the `LAST_ROW` function +- 2: The latest non-null value in each column of each subtable is cached. This option significantly improves the performance of the `LAST` function in normal situations, such as WHERE, ORDER BY, GROUP BY, and INTERVAL statements. +- 3: Rows and columns are both cached. This option is equivalent to simultaneously enabling options 1 and 2. -## Meta Cache +## Metadata Cache -To process data writing and querying efficiently, each vnode caches the metadata that's already retrieved. Parameters `pages` and `pagesize` are used to specify the size of metadata cache for each vnode. +To improve query and write performance, each vnode caches the metadata that it receives. When you create a database, you can configure the size of the metadata cache through the *pages* and *pagesize* parameters. ```sql create database db0 pages 128 pagesize 16kb ``` -The above statement will create a database db0 each of whose vnode is allocated a meta cache of `128 * 16 KB = 2 MB` . +The preceding SQL statement creates 128 pages on each vnode in the `db0` database. Each page has a 16 KB metadata cache. ## File System Cache -TDengine utilizes WAL to provide basic reliability. The essential of WAL is to append data in a disk file, so the file system cache also plays an important role in the writing performance. Parameter `wal` can be used to specify the policy of writing WAL, there are 2 cases: -- 1: Write data to WAL without calling fsync, the data is actually written to the file system cache without flushing immediately, in this way you can get better write performance -- 2: Write data to WAL and invoke fsync, the data is immediately flushed to disk, in this way you can get higher reliability +TDengine implements data reliability by means of a write-ahead log (WAL). Writing data to the WAL is essentially writing data to the disk in an ordered, append-only manner. For this reason, the file system cache plays an important role in write performance. When you create a database, you can use the *wal* parameter to choose higher performance or higher reliability. +- 1: This option writes to the WAL but does not enable fsync. New data written to the WAL is stored in the file system cache but not written to disk. This provides better performance. +- 2: This option writes to the WAL and enables fsync. New data written to the WAL is immediately written to disk. This provides better data reliability. ## Client Cache -To improve the overall efficiency of processing data, besides the above caches, the core library `libtaos.so` (also referred to as `taosc`) which all client programs depend on also has its own cache. `taosc` caches the metadata of the databases, super tables, tables that the invoking client has accessed, plus other critical metadata such as the cluster topology. +In addition to the server-side caching discussed previously, the core client library `libtaos.so` also makes use of caching. TDengine Client caches the metadata of all databases, supertables, and subtables that it has accessed, as well as the cluster topology. -When multiple client programs are accessing a TDengine cluster, if one of the clients modifies some metadata, the cache may become invalid in other clients. If this case happens, the client programs need to "reset query cache" to invalidate the whole cache so that `taosc` is enforced to repull the metadata it needs to rebuild the cache. +If a client modifies certain metadata while multiple clients are simultaneously accessing a TDengine cluster, the metadata caches on each client may fail or become out of sync. If this occurs, run the `reset query cache` command on the affected clientsto force them to obtain fresh metadata and reset their caches. diff --git a/docs/en/07-develop/09-udf.md b/docs/en/07-develop/09-udf.md index 49bc95bd91a4c31d42d2b21ef05d69225f1bd963..deb9c4cdb5b50edf7b48537f607ac47edc1246fd 100644 --- a/docs/en/07-develop/09-udf.md +++ b/docs/en/07-develop/09-udf.md @@ -1,240 +1,245 @@ --- sidebar_label: UDF -title: User Defined Functions(UDF) -description: "Scalar functions and aggregate functions developed by users can be utilized by the query framework to expand query capability" +title: User-Defined Functions (UDF) +description: "You can define your own scalar and aggregate functions to expand the query capabilities of TDengine." --- -In some use cases, built-in functions are not adequate for the query capability required by application programs. With UDF, the functions developed by users can be utilized by the query framework to meet business and application requirements. UDF normally takes one column of data as input, but can also support the result of a sub-query as input. +The built-in functions of TDengine may not be sufficient for the use cases of every application. In this case, you can define custom functions for use in TDengine queries. These are known as user-defined functions (UDF). A user-defined function takes one column of data or the result of a subquery as its input. -From version 2.2.0.0, UDF written in C/C++ are supported by TDengine. +TDengine supports user-defined functions written in C or C++. This document describes the usage of user-defined functions. +User-defined functions can be scalar functions or aggregate functions. Scalar functions, such as `abs`, `sin`, and `concat`, output a value for every row of data. Aggregate functions, such as `avg` and `max` output one value for multiple rows of data. -## Types of UDF +When you create a user-defined function, you must implement standard interface functions: +- For scalar functions, implement the `scalarfn` interface function. +- For aggregate functions, implement the `aggfn_start`, `aggfn`, and `aggfn_finish` interface functions. +- To initialize your function, implement the `udf_init` function. To terminate your function, implement the `udf_destroy` function. -Two kinds of functions can be implemented by UDF: scalar functions and aggregate functions. +There are strict naming conventions for these interface functions. The names of the start, finish, init, and destroy interfaces must be _start, _finish, _init, and _destroy, respectively. Replace `scalarfn`, `aggfn`, and `udf` with the name of your user-defined function. -Scalar functions return multiple rows and aggregate functions return either 0 or 1 row. - -In the case of a scalar function you only have to implement the "normal" function template. - -In the case of an aggregate function, in addition to the "normal" function, you also need to implement the "merge" and "finalize" function templates even if the implementation is empty. This will become clear in the sections below. - -### Scalar Function - -As mentioned earlier, a scalar UDF only has to implement the "normal" function template. The function template below can be used to define your own scalar function. - -`void udfNormalFunc(char* data, short itype, short ibytes, int numOfRows, long long* ts, char* dataOutput, char* interBuf, char* tsOutput, int* numOfOutput, short otype, short obytes, SUdfInit* buf)` - -`udfNormalFunc` is the place holder for a function name. A function implemented based on the above template can be used to perform scalar computation on data rows. The parameters are fixed to control the data exchange between UDF and TDengine. - -- Definitions of the parameters: - - - data：input data - - itype：the type of input data, for details please refer to [type definition in column_meta](/reference/rest-api/), for example 4 represents INT - - iBytes：the number of bytes consumed by each value in the input data - - oType：the type of output data, similar to iType - - oBytes：the number of bytes consumed by each value in the output data - - numOfRows：the number of rows in the input data - - ts: the column of timestamp corresponding to the input data - - dataOutput：the buffer for output data, total size is `oBytes * numberOfRows` - - interBuf：the buffer for an intermediate result. Its size is specified by the `BUFSIZE` parameter when creating a UDF. It's normally used when the intermediate result is not same as the final result. This buffer is allocated and freed by TDengine. - - tsOutput：the column of timestamps corresponding to the output data; it can be used to output timestamp together with the output data if it's not NULL - - numOfOutput：the number of rows in output data - - buf：for the state exchange between UDF and TDengine +## Implementing a Scalar Function +The implementation of a scalar function is described as follows: +```c +#include "taos.h" +#include "taoserror.h" +#include "taosudf.h" + +// initialization function. if no initialization, we can skip definition of it. The initialization function shall be concatenation of the udf name and _init suffix +// @return error number defined in taoserror.h +int32_t scalarfn_init() { + // initialization. + return TSDB_CODE_SUCCESS; +} + +// scalar function main computation function +// @param inputDataBlock, input data block composed of multiple columns with each column defined by SUdfColumn +// @param resultColumn, output column +// @return error number defined in taoserror.h +int32_t scalarfn(SUdfDataBlock* inputDataBlock, SUdfColumn* resultColumn) { + // read data from inputDataBlock and process, then output to resultColumn. + return TSDB_CODE_SUCCESS; +} + +// cleanup function. if no cleanup related processing, we can skip definition of it. The destroy function shall be concatenation of the udf name and _destroy suffix. +// @return error number defined in taoserror.h +int32_t scalarfn_destroy() { + // clean up + return TSDB_CODE_SUCCESS; +} +``` +Replace `scalarfn` with the name of your function. - [add_one.c](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/add_one.c) is one example of a very simple UDF implementation, i.e. one instance of the above `udfNormalFunc` template. It adds one to each value of a passed in column, which can be filtered using the `where` clause, and outputs the result. +## Implementing an Aggregate Function -### Aggregate Function +The implementation of an aggregate function is described as follows: +```c +#include "taos.h" +#include "taoserror.h" +#include "taosudf.h" + +// Initialization function. if no initialization, we can skip definition of it. The initialization function shall be concatenation of the udf name and _init suffix +// @return error number defined in taoserror.h +int32_t aggfn_init() { + // initialization. + return TSDB_CODE_SUCCESS; +} + +// aggregate start function. The intermediate value or the state(@interBuf) is initialized in this function. The function name shall be concatenation of udf name and _start suffix +// @param interbuf intermediate value to intialize +// @return error number defined in taoserror.h +int32_t aggfn_start(SUdfInterBuf* interBuf) { + // initialize intermediate value in interBuf + return TSDB_CODE_SUCESS; +} + +// aggregate reduce function. This function aggregate old state(@interbuf) and one data bock(inputBlock) and output a new state(@newInterBuf). +// @param inputBlock input data block +// @param interBuf old state +// @param newInterBuf new state +// @return error number defined in taoserror.h +int32_t aggfn(SUdfDataBlock* inputBlock, SUdfInterBuf *interBuf, SUdfInterBuf *newInterBuf) { + // read from inputBlock and interBuf and output to newInterBuf + return TSDB_CODE_SUCCESS; +} + +// aggregate function finish function. This function transforms the intermediate value(@interBuf) into the final output(@result). The function name must be concatenation of aggfn and _finish suffix. +// @interBuf : intermediate value +// @result: final result +// @return error number defined in taoserror.h +int32_t int32_t aggfn_finish(SUdfInterBuf* interBuf, SUdfInterBuf *result) { + // read data from inputDataBlock and process, then output to result + return TSDB_CODE_SUCCESS; +} + +// cleanup function. if no cleanup related processing, we can skip definition of it. The destroy function shall be concatenation of the udf name and _destroy suffix. +// @return error number defined in taoserror.h +int32_t aggfn_destroy() { + // clean up + return TSDB_CODE_SUCCESS; +} +``` +Replace `aggfn` with the name of your function. -For aggregate UDF, as mentioned earlier you must implement a "normal" function template (described above) and also implement the "merge" and "finalize" templates. +## Interface Functions -#### Merge Function Template +There are strict naming conventions for interface functions. The names of the start, finish, init, and destroy interfaces must be _start, _finish, _init, and _destroy, respectively. Replace `scalarfn`, `aggfn`, and `udf` with the name of your user-defined function. -The function template below can be used to define your own merge function for an aggregate UDF. +Interface functions return a value that indicates whether the operation was successful. If an operation fails, the interface function returns an error code. Otherwise, it returns TSDB_CODE_SUCCESS. The error codes are defined in `taoserror.h` and in the common API error codes in `taos.h`. For example, TSDB_CODE_UDF_INVALID_INPUT indicates invalid input. TSDB_CODE_OUT_OF_MEMORY indicates insufficient memory. -`void udfMergeFunc(char* data, int32_t numOfRows, char* dataOutput, int32_t* numOfOutput, SUdfInit* buf)` +For information about the parameters for interface functions, see Data Model -`udfMergeFunc` is the place holder for a function name. The function implemented with the above template is used to aggregate intermediate results and can only be used in the aggregate query for STable. +### Interfaces for Scalar Functions -Definitions of the parameters: + `int32_t scalarfn(SUdfDataBlock* inputDataBlock, SUdfColumn *resultColumn)` + + Replace `scalarfn` with the name of your function. This function performs scalar calculations on data blocks. You can configure a value through the parameters in the `resultColumn` structure. -- data：array of output data, if interBuf is used it's an array of interBuf -- numOfRows：number of rows in `data` -- dataOutput：the buffer for output data, the size is same as that of the final result; If the result is not final, it can be put in the interBuf, i.e. `data`. -- numOfOutput：number of rows in the output data -- buf：for the state exchange between UDF and TDengine +The parameters in the function are defined as follows: + - inputDataBlock: The data block to input. + - resultColumn: The column to output. The column to output. -#### Finalize Function Template +### Interfaces for Aggregate Functions -The function template below can be used to finalize the result of your own UDF, normally used when interBuf is used. +`int32_t aggfn_start(SUdfInterBuf *interBuf)` -`void udfFinalizeFunc(char* dataOutput, char* interBuf, int* numOfOutput, SUdfInit* buf)` +`int32_t aggfn(SUdfDataBlock* inputBlock, SUdfInterBuf *interBuf, SUdfInterBuf *newInterBuf)` -`udfFinalizeFunc` is the place holder of function name, definitions of the parameter are as below: +`int32_t aggfn_finish(SUdfInterBuf* interBuf, SUdfInterBuf *result)` -- dataOutput：buffer for output data -- interBuf：buffer for intermediate result, can be used as input for next processing step -- numOfOutput：number of output data, can only be 0 or 1 for aggregate function -- buf：for state exchange between UDF and TDengine +Replace `aggfn` with the name of your function. In the function, aggfn_start is called to generate a result buffer. Data is then divided between multiple blocks, and aggfn is called on each block to update the result. Finally, aggfn_finish is called to generate final results from the intermediate results. The final result contains only one or zero data points. -### Example abs_max.c +The parameters in the function are defined as follows: + - interBuf: The intermediate result buffer. + - inputBlock: The data block to input. + - newInterBuf: The new intermediate result buffer. + - result: The final result. -[abs_max.c](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/abs_max.c) is an example of a user defined aggregate function to get the maximum from the absolute values of a column. -The internal processing happens as follows. The results of the select statement are divided into multiple row blocks and `udfNormalFunc`, i.e. `abs_max` in this case, is performed on each row block to generate the intermediate results for each sub table. Then `udfMergeFunc`, i.e. `abs_max_merge` in this case, is performed on the intermediate result of sub tables to aggregate and generate the final or intermediate result of STable. The intermediate result of STable is finally processed by `udfFinalizeFunc`, i.e. `abs_max_finalize` in this example, to generate the final result, which contains either 0 or 1 row. +### Initializing and Terminating User-Defined Functions +`int32_t udf_init()` -Other typical aggregation functions such as covariance, can also be implemented using aggregate UDF. +`int32_t udf_destroy()` -## UDF Naming Conventions +Replace `udf`with the name of your function. udf_init initializes the function. udf_destroy terminates the function. If it is not necessary to initialize your function, udf_init is not required. If it is not necessary to terminate your function, udf_destroy is not required. -The naming convention for the 3 kinds of function templates required by UDF is as follows: - - udfNormalFunc, udfMergeFunc, and udfFinalizeFunc are required to have same prefix, i.e. the actual name of udfNormalFunc. The udfNormalFunc doesn't need a suffix following the function name. - - udfMergeFunc should be udfNormalFunc followed by `_merge` - - udfFinalizeFunc should be udfNormalFunc followed by `_finalize`. - -The naming convention is part of TDengine's UDF framework. TDengine follows this convention to invoke the corresponding actual functions. -Depending on whether you are creating a scalar UDF or aggregate UDF, the functions that you need to implement are different. +## Data Structure of User-Defined Functions +```c +typedef struct SUdfColumnMeta { + int16_t type; + int32_t bytes; + uint8_t precision; + uint8_t scale; +} SUdfColumnMeta; + +typedef struct SUdfColumnData { + int32_t numOfRows; + int32_t rowsAlloc; + union { + struct { + int32_t nullBitmapLen; + char *nullBitmap; + int32_t dataLen; + char *data; + } fixLenCol; + + struct { + int32_t varOffsetsLen; + int32_t *varOffsets; + int32_t payloadLen; + char *payload; + int32_t payloadAllocLen; + } varLenCol; + }; +} SUdfColumnData; + +typedef struct SUdfColumn { + SUdfColumnMeta colMeta; + bool hasNull; + SUdfColumnData colData; +} SUdfColumn; + +typedef struct SUdfDataBlock { + int32_t numOfRows; + int32_t numOfCols; + SUdfColumn **udfCols; +} SUdfDataBlock; + +typedef struct SUdfInterBuf { + int32_t bufLen; + char* buf; + int8_t numOfResult; //zero or one +} SUdfInterBuf; +``` +The data structure is described as follows: -- Scalar function：udfNormalFunc is required. -- Aggregate function：udfNormalFunc, udfMergeFunc (if query on STable) and udfFinalizeFunc are required. +- The SUdfDataBlock block includes the number of rows (numOfRows) and number of columns (numCols). udfCols[i] (0 <= i <= numCols-1) indicates that each column is of type SUdfColumn. +- SUdfColumn includes the definition of the data type of the column (colMeta) and the data in the column (colData). +- The member definitions of SUdfColumnMeta are the same as the data type definitions in `taos.h`. +- The data in SUdfColumnData can become longer. varLenCol indicates variable-length data, and fixLenCol indicates fixed-length data. +- SUdfInterBuf defines the intermediate structure `buffer` and the number of results in the buffer `numOfResult`. -For clarity, assuming we want to implement a UDF named "foo": -- If the function is a scalar function, we only need to implement the "normal" function template and it should be named simply `foo`. -- If the function is an aggregate function, we need to implement `foo`, `foo_merge`, and `foo_finalize`. Note that for aggregate UDF, even though one of the three functions is not necessary, there must be an empty implementation. +Additional functions are defined in `taosudf.h` to make it easier to work with these structures. ## Compile UDF -The source code of UDF in C can't be utilized by TDengine directly. UDF can only be loaded into TDengine after compiling to dynamically linked library (DLL). +To use your user-defined function in TDengine, first compile it to a dynamically linked library (DLL). -For example, the example UDF `add_one.c` mentioned earlier, can be compiled into DLL using the command below, in a Linux Shell. +For example, the sample UDF `add_one.c` can be compiled into a DLL as follows: ```bash gcc -g -O0 -fPIC -shared add_one.c -o add_one.so ``` -The generated DLL file `add_one.so` can be used later when creating a UDF. It's recommended to use GCC not older than 7.5. - -## Create and Use UDF - -When a UDF is created in a TDengine instance, it is available across the databases in that instance. - -### Create UDF - -SQL command can be executed on the host where the generated UDF DLL resides to load the UDF DLL into TDengine. This operation cannot be done through REST interface or web console. Once created, any client of the current TDengine can use these UDF functions in their SQL commands. UDF are stored in the management node of TDengine. The UDFs loaded in TDengine would be still available after TDengine is restarted. - -When creating UDF, the type of UDF, i.e. a scalar function or aggregate function must be specified. If the specified type is wrong, the SQL statements using the function would fail with errors. The input type and output type don't need to be the same in UDF, but the input data type and output data type must be consistent with the UDF definition. - -- Create Scalar Function - -```sql -CREATE FUNCTION userDefinedFunctionName AS "/absolute/path/to/userDefinedFunctionName.so" OUTPUTTYPE [BUFSIZE B]; -``` - -- userDefinedFunctionName：The function name to be used in SQL statement which must be consistent with the function name defined by `udfNormalFunc` and is also the name of the compiled DLL (.so file). -- path：The absolute path of the DLL file including the name of the shared object file (.so). The path must be quoted with single or double quotes. -- outputtype：The output data type, the value is the literal string of the supported TDengine data type. -- B：the size of intermediate buffer, in bytes; it is an optional parameter and the range is [0,512]. - -For example, below SQL statement can be used to create a UDF from `add_one.so`. - -```sql -CREATE FUNCTION add_one AS "/home/taos/udf_example/add_one.so" OUTPUTTYPE INT; -``` - -- Create Aggregate Function - -```sql -CREATE AGGREGATE FUNCTION userDefinedFunctionName AS "/absolute/path/to/userDefinedFunctionName.so" OUTPUTTYPE [ BUFSIZE B ]; -``` - -- userDefinedFunctionName：the function name to be used in SQL statement which must be consistent with the function name defined by `udfNormalFunc` and is also the name of the compiled DLL (.so file). -- path：the absolute path of the DLL file including the name of the shared object file (.so). The path needs to be quoted by single or double quotes. -- OUTPUTTYPE：the output data type, the value is the literal string of the type -- B：the size of intermediate buffer, in bytes; it's an optional parameter and the range is [0,512] - -For details about how to use intermediate result, please refer to example program [demo.c](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/demo.c). - -For example, below SQL statement can be used to create a UDF from `demo.so`. - -```sql -CREATE AGGREGATE FUNCTION demo AS "/home/taos/udf_example/demo.so" OUTPUTTYPE DOUBLE bufsize 14; -``` - -### Manage UDF - -- Delete UDF - -``` -DROP FUNCTION ids(X); -``` - -- ids(X)：same as that in `CREATE FUNCTION` statement - -```sql -DROP FUNCTION add_one; -``` - -- Show Available UDF +The generated DLL file `add_one.so` can now be used to implement your function. Note: GCC 7.5 or later is required. -```sql -SHOW FUNCTIONS; -``` - -### Use UDF - -The function name specified when creating UDF can be used directly in SQL statements, just like builtin functions. - -```sql -SELECT X(c) FROM table/STable; -``` - -The above SQL statement invokes function X for column c. - -## Restrictions for UDF - -In current version there are some restrictions for UDF +## Manage and Use User-Defined Functions +After compiling your function into a DLL, you add it to TDengine. For more information, see [User-Defined Functions](../12-taos-sql/26-udf.md). -1. Only Linux is supported when creating and invoking UDF for both client side and server side -2. UDF can't be mixed with builtin functions -3. Only one UDF can be used in a SQL statement -4. Only a single column is supported as input for UDF -5. Once created successfully, UDF is persisted in MNode of TDengineUDF -6. UDF can't be created through REST interface -7. The function name used when creating UDF in SQL must be consistent with the function name defined in the DLL, i.e. the name defined by `udfNormalFunc` -8. The name of a UDF should not conflict with any of TDengine's built-in functions +## Sample Code -## Examples +### Sample scalar function: [bit_and](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/bit_and.c) -### Scalar function example [add_one](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/add_one.c) +The bit_add function implements bitwise addition for multiple columns. If there is only one column, the column is returned. The bit_add function ignores null values.
-add_one.c +bit_and.c ```c -{{#include tests/script/sh/add_one.c}} +{{#include tests/script/sh/bit_and.c}} ```
-### Aggregate function example [abs_max](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/abs_max.c) - -
-abs_max.c - -```c -{{#include tests/script/sh/abs_max.c}} -``` - -
+### Sample aggregate function: [l2norm](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/l2norm.c) -### Example for using intermediate result [demo](https://github.com/taosdata/TDengine/blob/develop/tests/script/sh/demo.c) +The l2norm function finds the second-order norm for all data in the input column. This squares the values, takes a cumulative sum, and finds the square root.
-demo.c +l2norm.c ```c -{{#include tests/script/sh/demo.c}} +{{#include tests/script/sh/l2norm.c}} ```
diff --git a/docs/en/07-develop/_sub_c.mdx b/docs/en/07-develop/_sub_c.mdx index da492a0269f064d8cdf9dfb80969894131d94015..b0667268e9978533e84e68ea3fe5f285538df762 100644 --- a/docs/en/07-develop/_sub_c.mdx +++ b/docs/en/07-develop/_sub_c.mdx @@ -1,3 +1,3 @@ ```c -{{#include docs/examples/c/subscribe_demo.c}} -``` \ No newline at end of file +{{#include docs/examples/c/tmq_example.c}} +``` diff --git a/docs/en/07-develop/_sub_python.mdx b/docs/en/07-develop/_sub_python.mdx index 490b76fca6deb61e61dc59c2096b30742a7d25f7..1309da5b416799492a6b85aae4b775e227c0ad6e 100644 --- a/docs/en/07-develop/_sub_python.mdx +++ b/docs/en/07-develop/_sub_python.mdx @@ -1,3 +1,3 @@ ```py -{{#include docs/examples/python/subscribe_demo.py}} -``` \ No newline at end of file +{{#include docs/examples/python/tmq_example.py}} +``` diff --git a/docs/en/07-develop/index.md b/docs/en/07-develop/index.md index e3f55f290753f79ac1708337082ce90bb050b21f..1ef5e23f72f707f7a9decce6ea0bfed8fd642c0c 100644 --- a/docs/en/07-develop/index.md +++ b/docs/en/07-develop/index.md @@ -2,13 +2,12 @@ title: Developer Guide --- -To develop an application to process time-series data using TDengine, we recommend taking the following steps: - -1. Choose the method to connect to TDengine. No matter what programming language you use, you can always use the REST interface to access TDengine, but you can also use connectors unique to each programming language. -2. Design the data model based on your own use cases. Learn the [concepts](/concept/) of TDengine including "one table for one data collection point" and the "super table" (STable) concept; learn about static labels, collected metrics, and subtables. Depending on the characteristics of your data and your requirements, you may decide to create one or more databases, and you should design the STable schema to fit your data. +Before creating an application to process time-series data with TDengine, consider the following: +1. Choose the method to connect to TDengine. TDengine offers a REST API that can be used with any programming language. It also has connectors for a variety of languages. +2. Design the data model based on your own use cases. Consider the main [concepts](/concept/) of TDengine, including "one table per data collection point" and the supertable. Learn about static labels, collected metrics, and subtables. Depending on the characteristics of your data and your requirements, you decide to create one or more databases and design a supertable schema that fit your data. 3. Decide how you will insert data. TDengine supports writing using standard SQL, but also supports schemaless writing, so that data can be written directly without creating tables manually. 4. Based on business requirements, find out what SQL query statements need to be written. You may be able to repurpose any existing SQL. -5. If you want to run real-time analysis based on time series data, including various dashboards, it is recommended that you use the TDengine continuous query feature instead of deploying complex streaming processing systems such as Spark or Flink. +5. If you want to run real-time analysis based on time series data, including various dashboards, use the TDengine stream processing component instead of deploying complex systems such as Spark or Flink. 6. If your application has modules that need to consume inserted data, and they need to be notified when new data is inserted, it is recommended that you use the data subscription function provided by TDengine without the need to deploy Kafka. 7. In many use cases (such as fleet management), the application needs to obtain the latest status of each data collection point. It is recommended that you use the cache function of TDengine instead of deploying Redis separately. 8. If you find that the SQL functions of TDengine cannot meet your requirements, then you can use user-defined functions to solve the problem. diff --git a/docs/en/10-deployment/01-deploy.md b/docs/en/10-deployment/01-deploy.md new file mode 100644 index 0000000000000000000000000000000000000000..bfbb547bd4177cba369ec9d3d2541bceed853ef0 --- /dev/null +++ b/docs/en/10-deployment/01-deploy.md @@ -0,0 +1,191 @@ +--- +sidebar_label: Manual Deployment +title: Manual Deployment and Management +--- + +## Prerequisites + +### Step 0 + +The FQDN of all hosts must be setup properly. For e.g. FQDNs may have to be configured in the /etc/hosts file on each host. You must confirm that each FQDN can be accessed from any other host. For e.g. you can do this by using the `ping` command. If you have a DNS server on your network, contact your network administrator for assistance. + +### Step 1 + +If any previous version of TDengine has been installed and configured on any host, the installation needs to be removed and the data needs to be cleaned up. For details about uninstalling please refer to [Install and Uninstall](/operation/pkg-install). To clean up the data, please use `rm -rf /var/lib/taos/\*` assuming the `dataDir` is configured as `/var/lib/taos`. + +:::note +FQDN information is written to file. If you have started TDengine without configuring or changing the FQDN, ensure that data is backed up or no longer needed before running the `rm -rf /var/lib\taos/\*` command. +::: + +:::note +- The host where the client program runs also needs to be configured properly for FQDN, to make sure all hosts for client or server can be accessed from any other. In other words, the hosts where the client is running are also considered as a part of the cluster. +::: + +### Step 2 + +- Please ensure that your firewall rules do not block TCP/UDP on ports 6030-6042 on all hosts in the cluster. + +### Step 3 + +Now it's time to install TDengine on all hosts but without starting `taosd`. Note that the versions on all hosts should be same. If you are prompted to input the existing TDengine cluster, simply press carriage return to ignore the prompt. + +### Step 4 + +Now each physical node (referred to, hereinafter, as `dnode` which is an abbreviation for "data node") of TDengine needs to be configured properly. + +To get the hostname on any host, the command `hostname -f` can be executed. + +`ping ` command can be executed on each host to check whether any other host is accessible from it. If any host is not accessible, the network configuration, like /etc/hosts or DNS configuration, needs to be checked and revised, to make any two hosts accessible to each other. Hosts that are not accessible to each other cannot form a cluster. + +On the physical machine running the application, ping the dnode that is running taosd. If the dnode is not accessible, the application cannot connect to taosd. In this case, verify the DNS and hosts settings on the physical node running the application. + +The end point of each dnode is the output hostname and port, such as h1.taosdata.com:6030. + +### Step 5 + +Modify the TDengine configuration file `/etc/taos/taos.cfg` on each node. Assuming the first dnode of TDengine cluster is "h1.taosdata.com:6030", its `taos.cfg` is configured as following. + +```c +// firstEp is the end point to connect to when any dnode starts +firstEp h1.taosdata.com:6030 + +// must be configured to the FQDN of the host where the dnode is launched +fqdn h1.taosdata.com + +// the port used by the dnode, default is 6030 +serverPort 6030 + +``` + +`firstEp` and `fqdn` must be configured properly. In `taos.cfg` of all dnodes in TDengine cluster, `firstEp` must be configured to point to same address, i.e. the first dnode of the cluster. `fqdn` and `serverPort` compose the address of each node itself. Retain the default values for other parameters. + +For all the dnodes in a TDengine cluster, the below parameters must be configured exactly the same, any node whose configuration is different from dnodes already in the cluster can't join the cluster. + +| **#** | **Parameter** | **Definition** | +| ----- | ------------------ | ------------------------------------------- | +| 1 | statusInterval | The interval by which dnode reports its status to mnode | +| 2 | timezone | Timezone | +| 3 | locale | System region and encoding | +| 4 | charset | Character set | + +## Start Cluster + +The first dnode can be started following the instructions in [Get Started](/get-started/). Then TDengine CLI `taos` can be launched to execute command `show dnodes`, the output is as following for example: + +``` +taos> show dnodes; +id | endpoint | vnodes | support_vnodes | status | create_time | note | +============================================================================================================================================ +1 | h1.taosdata.com:6030 | 0 | 1024 | ready | 2022-07-16 10:50:42.673 | | +Query OK, 1 rows affected (0.007984s) + + +``` + +From the above output, it is shown that the end point of the started dnode is "h1.taosdata.com:6030", which is the `firstEp` of the cluster. + +## Add DNODE + +There are a few steps necessary to add other dnodes in the cluster. + +Second, we can start `taosd` as instructed in [Get Started](/get-started/). + +Then, on the first dnode i.e. h1.taosdata.com in our example, use TDengine CLI `taos` to execute the following command: + +```sql +CREATE DNODE "h2.taos.com:6030"; +```` + +This adds the end point of the new dnode (from Step 4) into the end point list of the cluster. In the command "fqdn:port" should be quoted using double quotes. Change `"h2.taos.com:6030"` to the end point of your new dnode. + +Then on the first dnode h1.taosdata.com, execute `show dnodes` in `taos` + +```sql +SHOW DNODES; +``` + +to show whether the second dnode has been added in the cluster successfully or not. If the status of the newly added dnode is offline, please check: + +- Whether the `taosd` process is running properly or not +- In the log file `taosdlog.0` to see whether the fqdn and port are correct and add the correct end point if not. +The above process can be repeated to add more dnodes in the cluster. + +:::tip + +Any node that is in the cluster and online can be the firstEp of new nodes. +Nodes use the firstEp parameter only when joining a cluster for the first time. After a node has joined the cluster, it stores the latest mnode in its end point list and no longer makes use of firstEp. +However, firstEp is used by clients that connect to the cluster. For example, if you run `taos shell` without arguments, it connects to the firstEp by default. +Two dnodes that are launched without a firstEp value operate independently of each other. It is not possible to add one dnode to the other dnode and form a cluster. It is also not possible to form two independent clusters into a new cluster. + +::: + +## Show DNODEs + +The below command can be executed in TDengine CLI `taos` + +```sql +SHOW DNODES; +``` + +to list all dnodes in the cluster, including ID, end point (fqdn:port), status (ready, offline), number of vnodes, number of free vnodes and so on. We recommend executing this command after adding or removing a dnode. + +Below is the example output of this command. + +``` +taos> show dnodes; + id | endpoint | vnodes | support_vnodes | status | create_time | note | +============================================================================================================================================ + 1 | trd01:6030 | 100 | 1024 | ready | 2022-07-15 16:47:47.726 | | +Query OK, 1 rows affected (0.006684s) +``` + +## Show VGROUPs + +To utilize system resources efficiently and provide scalability, data sharding is required. The data of each database is divided into multiple shards and stored in multiple vnodes. These vnodes may be located on different dnodes. One way of scaling out is to add more vnodes on dnodes. Each vnode can only be used for a single DB, but one DB can have multiple vnodes. The allocation of vnode is scheduled automatically by mnode based on system resources of the dnodes. + +Launch TDengine CLI `taos` and execute below command: + +```sql +USE SOME_DATABASE; +SHOW VGROUPS; +``` + +The example output is below: + +``` +taos> use db; +Database changed. + +taos> show vgroups; + vgroup_id | db_name | tables | v1_dnode | v1_status | v2_dnode | v2_status | v3_dnode | v3_status | status | nfiles | file_size | tsma | +================================================================================================================================================================================================ + 2 | db | 0 | 1 | leader | NULL | NULL | NULL | NULL | NULL | NULL | NULL | 0 | + 3 | db | 0 | 1 | leader | NULL | NULL | NULL | NULL | NULL | NULL | NULL | 0 | + 4 | db | 0 | 1 | leader | NULL | NULL | NULL | NULL | NULL | NULL | NULL | 0 | +Query OK, 8 row(s) in set (0.001154s) +``` + +## Drop DNODE + +Before running the TDengine CLI, ensure that the taosd process has been stopped on the dnode that you want to delete. + +```sql +DROP DNODE "fqdn:port"; +``` + +or + +```sql +DROP DNODE dnodeId; +``` + +to drop or remove a dnode from the cluster. In the command, you can get `dnodeId` from `show dnodes`. + +:::warning + +- Once a dnode is dropped, it can't rejoin the cluster. To rejoin, the dnode needs to deployed again after cleaning up the data directory. Before dropping a dnode, the data belonging to the dnode MUST be migrated/backed up according to your data retention, data security or other SOPs. +- Please note that `drop dnode` is different from stopping `taosd` process. `drop dnode` just removes the dnode out of TDengine cluster. Only after a dnode is dropped, can the corresponding `taosd` process be stopped. +- Once a dnode is dropped, other dnodes in the cluster will be notified of the drop and will not accept the request from the dropped dnode. +- dnodeID is allocated automatically and can't be manually modified. dnodeID is generated in ascending order without duplication. + +::: diff --git a/docs/en/10-deployment/03-k8s.md b/docs/en/10-deployment/03-k8s.md new file mode 100644 index 0000000000000000000000000000000000000000..b3f71ed5bd0e0dbaf3108cc40be6b18bdf5fb7e8 --- /dev/null +++ b/docs/en/10-deployment/03-k8s.md @@ -0,0 +1,392 @@ +--- +sidebar_label: Kubernetes +title: Deploying a TDengine Cluster in Kubernetes +--- + +TDengine is a cloud-native time-series database that can be deployed on Kubernetes. This document gives a step-by-step description of how you can use YAML files to create a TDengine cluster and introduces common operations for TDengine in a Kubernetes environment. + +## Prerequisites + +Before deploying TDengine on Kubernetes, perform the following: + +* Install and configure minikube, kubectl, and helm. +* Install and deploy Kubernetes and ensure that it can be accessed and used normally. Update any container registries or other services as necessary. + +You can download the configuration files in this document from [GitHub](https://github.com/taosdata/TDengine-Operator/tree/3.0/src/tdengine). + +## Configure the service + +Create a service configuration file named `taosd-service.yaml`. Record the value of `metadata.name` (in this example, `taos`) for use in the next step. Add the ports required by TDengine: + +```yaml +--- +apiVersion: v1 +kind: Service +metadata: + name: "taosd" + labels: + app: "tdengine" +spec: + ports: + - name: tcp6030 + - protocol: "TCP" + port: 6030 + - name: tcp6041 + - protocol: "TCP" + port: 6041 + selector: + app: "tdengine" +``` + +## Configure the service as StatefulSet + +Configure the TDengine service as a StatefulSet. +Create the `tdengine.yaml` file and set `replicas` to 3. In this example, the region is set to Asia/Shanghai and 10 GB of standard storage are allocated per node. You can change the configuration based on your environment and business requirements. + +```yaml +--- +apiVersion: apps/v1 +kind: StatefulSet +metadata: + name: "tdengine" + labels: + app: "tdengine" +spec: + serviceName: "taosd" + replicas: 3 + updateStrategy: + type: RollingUpdate + selector: + matchLabels: + app: "tdengine" + template: + metadata: + name: "tdengine" + labels: + app: "tdengine" + spec: + containers: + - name: "tdengine" + image: "tdengine/tdengine:3.0.0.0" + imagePullPolicy: "IfNotPresent" + ports: + - name: tcp6030 + - protocol: "TCP" + containerPort: 6030 + - name: tcp6041 + - protocol: "TCP" + containerPort: 6041 + env: + # POD_NAME for FQDN config + - name: POD_NAME + valueFrom: + fieldRef: + fieldPath: metadata.name + # SERVICE_NAME and NAMESPACE for fqdn resolve + - name: SERVICE_NAME + value: "taosd" + - name: STS_NAME + value: "tdengine" + - name: STS_NAMESPACE + valueFrom: + fieldRef: + fieldPath: metadata.namespace + # TZ for timezone settings, we recommend to always set it. + - name: TZ + value: "Asia/Shanghai" + # TAOS_ prefix will configured in taos.cfg, strip prefix and camelCase. + - name: TAOS_SERVER_PORT + value: "6030" + # Must set if you want a cluster. + - name: TAOS_FIRST_EP + value: "$(STS_NAME)-0.$(SERVICE_NAME).$(STS_NAMESPACE).svc.cluster.local:$(TAOS_SERVER_PORT)" + # TAOS_FQND should always be setted in k8s env. + - name: TAOS_FQDN + value: "$(POD_NAME).$(SERVICE_NAME).$(STS_NAMESPACE).svc.cluster.local" + volumeMounts: + - name: taosdata + mountPath: /var/lib/taos + readinessProbe: + exec: + command: + - taos-check + initialDelaySeconds: 5 + timeoutSeconds: 5000 + livenessProbe: + exec: + command: + - taos-check + initialDelaySeconds: 15 + periodSeconds: 20 + volumeClaimTemplates: + - metadata: + name: taosdata + spec: + accessModes: + - "ReadWriteOnce" + storageClassName: "standard" + resources: + requests: + storage: "10Gi" +``` + +## Use kubectl to deploy TDengine + +Run the following commands: + +```bash +kubectl apply -f taosd-service.yaml +kubectl apply -f tdengine.yaml +``` + +The preceding configuration generates a TDengine cluster with three nodes in which dnodes are automatically configured. You can run the `show dnodes` command to query the nodes in the cluster: + +```bash +kubectl exec -i -t tdengine-0 -- taos -s "show dnodes" +kubectl exec -i -t tdengine-1 -- taos -s "show dnodes" +kubectl exec -i -t tdengine-2 -- taos -s "show dnodes" +``` + +The output is as follows: + +``` +taos> show dnodes + id | endpoint | vnodes | support_vnodes | status | create_time | note | +============================================================================================================================================ + 1 | tdengine-0.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:14:57.285 | | + 2 | tdengine-1.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:11.302 | | + 3 | tdengine-2.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:23.290 | | +Query OK, 3 rows in database (0.003655s) +``` + +## Enable port forwarding + +The kubectl port forwarding feature allows applications to access the TDengine cluster running on Kubernetes. + +``` +kubectl port-forward tdengine-0 6041:6041 & +``` + +Use curl to verify that the TDengine REST API is working on port 6041: + +``` +$ curl -u root:taosdata -d "show databases" 127.0.0.1:6041/rest/sql +Handling connection for 6041 +{"code":0,"column_meta":[["name","VARCHAR",64],["create_time","TIMESTAMP",8],["vgroups","SMALLINT",2],["ntables","BIGINT",8],["replica","TINYINT",1],["strict","VARCHAR",4],["duration","VARCHAR",10],["keep","VARCHAR",32],["buffer","INT",4],["pagesize","INT",4],["pages","INT",4],["minrows","INT",4],["maxrows","INT",4],["comp","TINYINT",1],["precision","VARCHAR",2],["status","VARCHAR",10],["retention","VARCHAR",60],["single_stable","BOOL",1],["cachemodel","VARCHAR",11],["cachesize","INT",4],["wal_level","TINYINT",1],["wal_fsync_period","INT",4],["wal_retention_period","INT",4],["wal_retention_size","BIGINT",8],["wal_roll_period","INT",4],["wal_segment_size","BIGINT",8]],"data":[["information_schema",null,null,16,null,null,null,null,null,null,null,null,null,null,null,"ready",null,null,null,null,null,null,null,null,null,null],["performance_schema",null,null,10,null,null,null,null,null,null,null,null,null,null,null,"ready",null,null,null,null,null,null,null,null,null,null]],"rows":2} +``` + +## Enable the dashboard for visualization + + The minikube dashboard command enables visualized cluster management. + +``` +$ minikube dashboard +* Verifying dashboard health ... +* Launching proxy ... +* Verifying proxy health ... +* Opening http://127.0.0.1:46617/api/v1/namespaces/kubernetes-dashboard/services/http:kubernetes-dashboard:/proxy/ in your default browser... +http://127.0.0.1:46617/api/v1/namespaces/kubernetes-dashboard/services/http:kubernetes-dashboard:/proxy/ +``` + +In some public clouds, minikube cannot be remotely accessed if it is bound to 127.0.0.1. In this case, use the kubectl proxy command to map the port to 0.0.0.0. Then, you can access the dashboard by using a web browser to open the dashboard URL above on the public IP address and port of the virtual machine. + +``` +$ kubectl proxy --accept-hosts='^.*$' --address='0.0.0.0' +``` + +## Scaling Out Your Cluster + +TDengine clusters can scale automatically: + +```bash +kubectl scale statefulsets tdengine --replicas=4 +``` + +The preceding command increases the number of replicas to 4. After running this command, query the pod status: + +```bash +kubectl get pods -l app=tdengine +``` + +The output is as follows: + +``` +NAME READY STATUS RESTARTS AGE +tdengine-0 1/1 Running 0 161m +tdengine-1 1/1 Running 0 161m +tdengine-2 1/1 Running 0 32m +tdengine-3 1/1 Running 0 32m +``` + +The status of all pods is Running. Once the pod status changes to Ready, you can check the dnode status: + +```bash +kubectl exec -i -t tdengine-3 -- taos -s "show dnodes" +``` + +The following output shows that the TDengine cluster has been expanded to 4 replicas: + +``` +taos> show dnodes + id | endpoint | vnodes | support_vnodes | status | create_time | note | +============================================================================================================================================ + 1 | tdengine-0.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:14:57.285 | | + 2 | tdengine-1.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:11.302 | | + 3 | tdengine-2.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:23.290 | | + 4 | tdengine-3.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:33:16.039 | | +Query OK, 4 rows in database (0.008377s) +``` + +## Scaling In Your Cluster + +When you scale in a TDengine cluster, your data is migrated to different nodes. You must run the drop dnodes command in TDengine to remove dnodes before scaling in your Kubernetes environment. + +Note: In a Kubernetes StatefulSet service, the newest pods are always removed first. For this reason, when you scale in your TDengine cluster, ensure that you drop the newest dnodes. + +``` +$ kubectl exec -i -t tdengine-0 -- taos -s "drop dnode 4" +``` + +```bash +$ kubectl exec -it tdengine-0 -- taos -s "show dnodes" + +taos> show dnodes + id | endpoint | vnodes | support_vnodes | status | create_time | note | +============================================================================================================================================ + 1 | tdengine-0.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:14:57.285 | | + 2 | tdengine-1.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:11.302 | | + 3 | tdengine-2.taosd.default.sv... | 0 | 256 | ready | 2022-08-10 13:15:23.290 | | +Query OK, 3 rows in database (0.004861s) +``` + +Verify that the dnode have been successfully removed by running the `kubectl exec -i -t tdengine-0 -- taos -s "show dnodes"` command. Then run the following command to remove the pod: + +``` +kubectl scale statefulsets tdengine --replicas=3 +``` + +The newest pod in the deployment is removed. Run the `kubectl get pods -l app=tdengine` command to query the pod status: + +``` +$ kubectl get pods -l app=tdengine +NAME READY STATUS RESTARTS AGE +tdengine-0 1/1 Running 0 4m7s +tdengine-1 1/1 Running 0 3m55s +tdengine-2 1/1 Running 0 2m28s +``` + +After the pod has been removed, manually delete the PersistentVolumeClaim (PVC). Otherwise, future scale-outs will attempt to use existing data. + +```bash +$ kubectl delete pvc taosdata-tdengine-3 +``` + +Your cluster has now been safely scaled in, and you can scale it out again as necessary. + +```bash +$ kubectl scale statefulsets tdengine --replicas=4 +statefulset.apps/tdengine scaled +it@k8s-2:~/TDengine-Operator/src/tdengine$ kubectl get pods -l app=tdengine +NAME READY STATUS RESTARTS AGE +tdengine-0 1/1 Running 0 35m +tdengine-1 1/1 Running 0 34m +tdengine-2 1/1 Running 0 12m +tdengine-3 0/1 ContainerCreating 0 4s +it@k8s-2:~/TDengine-Operator/src/tdengine$ kubectl get pods -l app=tdengine +NAME READY STATUS RESTARTS AGE +tdengine-0 1/1 Running 0 35m +tdengine-1 1/1 Running 0 34m +tdengine-2 1/1 Running 0 12m +tdengine-3 0/1 Running 0 7s +it@k8s-2:~/TDengine-Operator/src/tdengine$ kubectl exec -it tdengine-0 -- taos -s "show dnodes" + +taos> show dnodes +id | endpoint | vnodes | support_vnodes | status | create_time | offline reason | +====================================================================================================================================== +1 | tdengine-0.taosd.default.sv... | 0 | 4 | ready | 2022-07-25 17:38:49.012 | | +2 | tdengine-1.taosd.default.sv... | 1 | 4 | ready | 2022-07-25 17:39:01.517 | | +5 | tdengine-2.taosd.default.sv... | 0 | 4 | ready | 2022-07-25 18:01:36.479 | | +6 | tdengine-3.taosd.default.sv... | 0 | 4 | ready | 2022-07-25 18:13:54.411 | | +Query OK, 4 row(s) in set (0.001348s) +``` + +## Remove a TDengine Cluster + +To fully remove a TDengine cluster, you must delete its statefulset, svc, configmap, and pvc entries: + +```bash +kubectl delete statefulset -l app=tdengine +kubectl delete svc -l app=tdengine +kubectl delete pvc -l app=tdengine +kubectl delete configmap taoscfg + +``` + +## Troubleshooting + +### Error 1 + +If you remove a pod without first running `drop dnode`, some TDengine nodes will go offline. + +``` +$ kubectl exec -it tdengine-0 -- taos -s "show dnodes" + +taos> show dnodes +id | endpoint | vnodes | support_vnodes | status | create_time | offline reason | +====================================================================================================================================== +1 | tdengine-0.taosd.default.sv... | 0 | 4 | ready | 2022-07-25 17:38:49.012 | | +2 | tdengine-1.taosd.default.sv... | 1 | 4 | ready | 2022-07-25 17:39:01.517 | | +5 | tdengine-2.taosd.default.sv... | 0 | 4 | offline | 2022-07-25 18:01:36.479 | status msg timeout | +6 | tdengine-3.taosd.default.sv... | 0 | 4 | offline | 2022-07-25 18:13:54.411 | status msg timeout | +Query OK, 4 row(s) in set (0.001323s) +``` + +### Error 2 + +If the number of nodes after a scale-in is less than the value of the replica parameter, the cluster will go down: + +Create a database with replica set to 2 and add data. + +```bash +kubectl exec -i -t tdengine-0 -- \ + taos -s \ + "create database if not exists test replica 2; + use test; + create table if not exists t1(ts timestamp, n int); + insert into t1 values(now, 1)(now+1s, 2);" + + +``` + +Scale in to one node: + +```bash +kubectl scale statefulsets tdengine --replicas=1 + +``` + +In the TDengine CLI, you can see that no database operations succeed: + +``` +taos> show dnodes; + id | end_point | vnodes | cores | status | role | create_time | offline reason | +====================================================================================================================================== + 1 | tdengine-0.taosd.default.sv... | 2 | 40 | ready | any | 2021-06-01 15:55:52.562 | | + 2 | tdengine-1.taosd.default.sv... | 1 | 40 | offline | any | 2021-06-01 15:56:07.212 | status msg timeout | +Query OK, 2 row(s) in set (0.000845s) + +taos> show dnodes; + id | end_point | vnodes | cores | status | role | create_time | offline reason | +====================================================================================================================================== + 1 | tdengine-0.taosd.default.sv... | 2 | 40 | ready | any | 2021-06-01 15:55:52.562 | | + 2 | tdengine-1.taosd.default.sv... | 1 | 40 | offline | any | 2021-06-01 15:56:07.212 | status msg timeout | +Query OK, 2 row(s) in set (0.000837s) + +taos> use test; +Database changed. + +taos> insert into t1 values(now, 3); + +DB error: Unable to resolve FQDN (0.013874s) + +``` diff --git a/docs/en/10-deployment/05-helm.md b/docs/en/10-deployment/05-helm.md new file mode 100644 index 0000000000000000000000000000000000000000..48cd9df32c16d346ceece01f01ee3880231427e7 --- /dev/null +++ b/docs/en/10-deployment/05-helm.md @@ -0,0 +1,417 @@ +--- +sidebar_label: Helm +title: Use Helm to deploy TDengine +--- + +Helm is a package manager for Kubernetes that can provide more capabilities in deploying on Kubernetes. + +## Install Helm + +```bash +curl -fsSL -o get_helm.sh \ + https://raw.githubusercontent.com/helm/helm/master/scripts/get-helm-3 +chmod +x get_helm.sh +./get_helm.sh + +``` + +Helm uses the kubectl and kubeconfig configurations to perform Kubernetes operations. For more information, see the Rancher configuration for Kubernetes installation. + +## Install TDengine Chart + +To use TDengine Chart, download it from GitHub: + +```bash +wget https://github.com/taosdata/TDengine-Operator/raw/3.0/helm/tdengine-3.0.0.tgz + +``` + +Query the storageclass of your Kubernetes deployment: + +```bash +kubectl get storageclass + +``` + +With minikube, the default value is standard. + +Use Helm commands to install TDengine: + +```bash +helm install tdengine tdengine-3.0.0.tgz \ + --set storage.className= + +``` + +You can configure a small storage size in minikube to ensure that your deployment does not exceed your available disk space. + +```bash +helm install tdengine tdengine-3.0.0.tgz \ + --set storage.className=standard \ + --set storage.dataSize=2Gi \ + --set storage.logSize=10Mi + +``` + +After TDengine is deployed, TDengine Chart outputs information about how to use TDengine: + +```bash +export POD_NAME=$(kubectl get pods --namespace default \ + -l "app.kubernetes.io/name=tdengine,app.kubernetes.io/instance=tdengine" \ + -o jsonpath="{.items[0].metadata.name}") +kubectl --namespace default exec $POD_NAME -- taos -s "show dnodes; show mnodes" +kubectl --namespace default exec -it $POD_NAME -- taos + +``` + +You can test the deployment by creating a table: + +```bash +kubectl --namespace default exec $POD_NAME -- \ + taos -s "create database test; + use test; + create table t1 (ts timestamp, n int); + insert into t1 values(now, 1)(now + 1s, 2); + select * from t1;" + +``` + +## Configuring Values + +You can configure custom parameters in TDengine with the `values.yaml` file. + +Run the `helm show values` command to see all parameters supported by TDengine Chart. + +```bash +helm show values tdengine-3.0.0.tgz + +``` + +Save the output of this command as `values.yaml`. Then you can modify this file with your desired values and use it to deploy a TDengine cluster: + +```bash +helm install tdengine tdengine-3.0.0.tgz -f values.yaml + +``` + +The parameters are described as follows: + +```yaml +# Default values for tdengine. +# This is a YAML-formatted file. +# Declare variables to be passed into helm templates. + +replicaCount: 1 + +image: + prefix: tdengine/tdengine + #pullPolicy: Always + # Overrides the image tag whose default is the chart appVersion. +# tag: "3.0.0.0" + +service: + # ClusterIP is the default service type, use NodeIP only if you know what you are doing. + type: ClusterIP + ports: + # TCP range required + tcp: [6030, 6041, 6042, 6043, 6044, 6046, 6047, 6048, 6049, 6060] + # UDP range + udp: [6044, 6045] + + +# Set timezone here, not in taoscfg +timezone: "Asia/Shanghai" + +resources: + # We usually recommend not to specify default resources and to leave this as a conscious + # choice for the user. This also increases chances charts run on environments with little + # resources, such as Minikube. If you do want to specify resources, uncomment the following + # lines, adjust them as necessary, and remove the curly braces after 'resources:'. + # limits: + # cpu: 100m + # memory: 128Mi + # requests: + # cpu: 100m + # memory: 128Mi + +storage: + # Set storageClassName for pvc. K8s use default storage class if not set. + # + className: "" + dataSize: "100Gi" + logSize: "10Gi" + +nodeSelectors: + taosd: + # node selectors + +clusterDomainSuffix: "" +# Config settings in taos.cfg file. +# +# The helm/k8s support will use environment variables for taos.cfg, +# converting an upper-snake-cased variable like `TAOS_DEBUG_FLAG`, +# to a camelCase taos config variable `debugFlag`. +# +# See the variable list at https://www.taosdata.com/cn/documentation/administrator . +# +# Note: +# 1. firstEp/secondEp: should not be setted here, it's auto generated at scale-up. +# 2. serverPort: should not be setted, we'll use the default 6030 in many places. +# 3. fqdn: will be auto generated in kubenetes, user should not care about it. +# 4. role: currently role is not supported - every node is able to be mnode and vnode. +# +# Btw, keep quotes "" around the value like below, even the value will be number or not. +taoscfg: + # Starts as cluster or not, must be 0 or 1. + # 0: all pods will start as a seperate TDengine server + # 1: pods will start as TDengine server cluster. [default] + CLUSTER: "1" + + # number of replications, for cluster only + TAOS_REPLICA: "1" + + + # number of days per DB file + # TAOS_DAYS: "10" + + # number of days to keep DB file, default is 10 years. + #TAOS_KEEP: "3650" + + # cache block size (Mbyte) + #TAOS_CACHE: "16" + + # number of cache blocks per vnode + #TAOS_BLOCKS: "6" + + # minimum rows of records in file block + #TAOS_MIN_ROWS: "100" + + # maximum rows of records in file block + #TAOS_MAX_ROWS: "4096" + + # + # TAOS_NUM_OF_THREADS_PER_CORE: number of threads per CPU core + #TAOS_NUM_OF_THREADS_PER_CORE: "1.0" + + # + # TAOS_NUM_OF_COMMIT_THREADS: number of threads to commit cache data + #TAOS_NUM_OF_COMMIT_THREADS: "4" + + # + # TAOS_RATIO_OF_QUERY_CORES: + # the proportion of total CPU cores available for query processing + # 2.0: the query threads will be set to double of the CPU cores. + # 1.0: all CPU cores are available for query processing [default]. + # 0.5: only half of the CPU cores are available for query. + # 0.0: only one core available. + #TAOS_RATIO_OF_QUERY_CORES: "1.0" + + # + # TAOS_KEEP_COLUMN_NAME: + # the last_row/first/last aggregator will not change the original column name in the result fields + #TAOS_KEEP_COLUMN_NAME: "0" + + # enable/disable backuping vnode directory when removing vnode + #TAOS_VNODE_BAK: "1" + + # enable/disable installation / usage report + #TAOS_TELEMETRY_REPORTING: "1" + + # enable/disable load balancing + #TAOS_BALANCE: "1" + + # max timer control blocks + #TAOS_MAX_TMR_CTRL: "512" + + # time interval of system monitor, seconds + #TAOS_MONITOR_INTERVAL: "30" + + # number of seconds allowed for a dnode to be offline, for cluster only + #TAOS_OFFLINE_THRESHOLD: "8640000" + + # RPC re-try timer, millisecond + #TAOS_RPC_TIMER: "1000" + + # RPC maximum time for ack, seconds. + #TAOS_RPC_MAX_TIME: "600" + + # time interval of dnode status reporting to mnode, seconds, for cluster only + #TAOS_STATUS_INTERVAL: "1" + + # time interval of heart beat from shell to dnode, seconds + #TAOS_SHELL_ACTIVITY_TIMER: "3" + + # minimum sliding window time, milli-second + #TAOS_MIN_SLIDING_TIME: "10" + + # minimum time window, milli-second + #TAOS_MIN_INTERVAL_TIME: "10" + + # maximum delay before launching a stream computation, milli-second + #TAOS_MAX_STREAM_COMP_DELAY: "20000" + + # maximum delay before launching a stream computation for the first time, milli-second + #TAOS_MAX_FIRST_STREAM_COMP_DELAY: "10000" + + # retry delay when a stream computation fails, milli-second + #TAOS_RETRY_STREAM_COMP_DELAY: "10" + + # the delayed time for launching a stream computation, from 0.1(default, 10% of whole computing time window) to 0.9 + #TAOS_STREAM_COMP_DELAY_RATIO: "0.1" + + # max number of vgroups per db, 0 means configured automatically + #TAOS_MAX_VGROUPS_PER_DB: "0" + + # max number of tables per vnode + #TAOS_MAX_TABLES_PER_VNODE: "1000000" + + # the number of acknowledgments required for successful data writing + #TAOS_QUORUM: "1" + + # enable/disable compression + #TAOS_COMP: "2" + + # write ahead log (WAL) level, 0: no wal; 1: write wal, but no fysnc; 2: write wal, and call fsync + #TAOS_WAL_LEVEL: "1" + + # if walLevel is set to 2, the cycle of fsync being executed, if set to 0, fsync is called right away + #TAOS_FSYNC: "3000" + + # the compressed rpc message, option: + # -1 (no compression) + # 0 (all message compressed), + # > 0 (rpc message body which larger than this value will be compressed) + #TAOS_COMPRESS_MSG_SIZE: "-1" + + # max length of an SQL + #TAOS_MAX_SQL_LENGTH: "1048576" + + # the maximum number of records allowed for super table time sorting + #TAOS_MAX_NUM_OF_ORDERED_RES: "100000" + + # max number of connections allowed in dnode + #TAOS_MAX_SHELL_CONNS: "5000" + + # max number of connections allowed in client + #TAOS_MAX_CONNECTIONS: "5000" + + # stop writing logs when the disk size of the log folder is less than this value + #TAOS_MINIMAL_LOG_DIR_G_B: "0.1" + + # stop writing temporary files when the disk size of the tmp folder is less than this value + #TAOS_MINIMAL_TMP_DIR_G_B: "0.1" + + # if disk free space is less than this value, taosd service exit directly within startup process + #TAOS_MINIMAL_DATA_DIR_G_B: "0.1" + + # One mnode is equal to the number of vnode consumed + #TAOS_MNODE_EQUAL_VNODE_NUM: "4" + + # enbale/disable http service + #TAOS_HTTP: "1" + + # enable/disable system monitor + #TAOS_MONITOR: "1" + + # enable/disable recording the SQL statements via restful interface + #TAOS_HTTP_ENABLE_RECORD_SQL: "0" + + # number of threads used to process http requests + #TAOS_HTTP_MAX_THREADS: "2" + + # maximum number of rows returned by the restful interface + #TAOS_RESTFUL_ROW_LIMIT: "10240" + + # The following parameter is used to limit the maximum number of lines in log files. + # max number of lines per log filters + # numOfLogLines 10000000 + + # enable/disable async log + #TAOS_ASYNC_LOG: "0" + + # + # time of keeping log files, days + #TAOS_LOG_KEEP_DAYS: "0" + + # The following parameters are used for debug purpose only. + # debugFlag 8 bits mask: FILE-SCREEN-UNUSED-HeartBeat-DUMP-TRACE_WARN-ERROR + # 131: output warning and error + # 135: output debug, warning and error + # 143: output trace, debug, warning and error to log + # 199: output debug, warning and error to both screen and file + # 207: output trace, debug, warning and error to both screen and file + # + # debug flag for all log type, take effect when non-zero value\ + #TAOS_DEBUG_FLAG: "143" + + # enable/disable recording the SQL in taos client + #TAOS_ENABLE_RECORD_SQL: "0" + + # generate core file when service crash + #TAOS_ENABLE_CORE_FILE: "1" + + # maximum display width of binary and nchar fields in the shell. The parts exceeding this limit will be hidden + #TAOS_MAX_BINARY_DISPLAY_WIDTH: "30" + + # enable/disable stream (continuous query) + #TAOS_STREAM: "1" + + # in retrieve blocking model, only in 50% query threads will be used in query processing in dnode + #TAOS_RETRIEVE_BLOCKING_MODEL: "0" + + # the maximum allowed query buffer size in MB during query processing for each data node + # -1 no limit (default) + # 0 no query allowed, queries are disabled + #TAOS_QUERY_BUFFER_SIZE: "-1" +``` + +## Scaling Out + +For information about scaling out your deployment, see Kubernetes. Additional Helm-specific is described as follows. + +First, obtain the name of the StatefulSet service for your deployment. + +```bash +export STS_NAME=$(kubectl get statefulset \ + -l "app.kubernetes.io/name=tdengine" \ + -o jsonpath="{.items[0].metadata.name}") + +``` + +You can scale out your deployment by adding replicas. The following command scales a deployment to three nodes: + +```bash +kubectl scale --replicas 3 statefulset/$STS_NAME + +``` + +Run the `show dnodes` and `show mnodes` commands to check whether the scale-out was successful. + +## Scaling In + +:::warning +Exercise caution when scaling in a cluster. + +::: + +Determine which dnodes you want to remove and drop them manually. + +```bash +kubectl --namespace default exec $POD_NAME -- \ + cat /var/lib/taos/dnode/dnodeEps.json \ + | jq '.dnodeInfos[1:] |map(.dnodeFqdn + ":" + (.dnodePort|tostring)) | .[]' -r +kubectl --namespace default exec $POD_NAME -- taos -s "show dnodes" +kubectl --namespace default exec $POD_NAME -- taos -s 'drop dnode ""' + +``` + +## Remove a TDengine Cluster + +You can use Helm to remove your cluster: + +```bash +helm uninstall tdengine + +``` + +However, Helm does not remove PVCs automatically. After you remove your cluster, manually remove all PVCs. diff --git a/docs/en/10-deployment/_category_.yml b/docs/en/10-deployment/_category_.yml new file mode 100644 index 0000000000000000000000000000000000000000..0bb1ba461bd4c0e350c60fa3a8bc7723429a3f9f --- /dev/null +++ b/docs/en/10-deployment/_category_.yml @@ -0,0 +1 @@ +label: Deployment diff --git a/docs/en/10-deployment/index.md b/docs/en/10-deployment/index.md new file mode 100644 index 0000000000000000000000000000000000000000..7054a33e4a40222ed5eb9a15837990e3e7a81cff --- /dev/null +++ b/docs/en/10-deployment/index.md @@ -0,0 +1,14 @@ +--- +title: Deployment +--- + +TDengine has a native distributed design and provides the ability to scale out. A few nodes can form a TDengine cluster. If you need higher processing power, you just need to add more nodes into the cluster. TDengine uses virtual node technology to virtualize a node into multiple virtual nodes to achieve load balancing. At the same time, TDengine can group virtual nodes on different nodes into virtual node groups, and use the replication mechanism to ensure the high availability of the system. The cluster feature of TDengine is completely open source. + +This document describes how to manually deploy a cluster on a host as well as how to deploy on Kubernetes and by using Helm. + +```mdx-code-block +import DocCardList from '@theme/DocCardList'; +import {useCurrentSidebarCategory} from '@docusaurus/theme-common'; + + +``` diff --git a/docs/en/12-taos-sql/01-data-type.md b/docs/en/12-taos-sql/01-data-type.md index d038219c8ac66db52416001f7a79c71018e2ca33..b830994ac9323f85d9ca68a40366edd9f2da1432 100644 --- a/docs/en/12-taos-sql/01-data-type.md +++ b/docs/en/12-taos-sql/01-data-type.md @@ -1,9 +1,10 @@ --- +sidebar_label: Data Types title: Data Types description: "TDengine supports a variety of data types including timestamp, float, JSON and many others." --- -## TIMESTAMP +## Timestamp When using TDengine to store and query data, the most important part of the data is timestamp. Timestamp must be specified when creating and inserting data rows. Timestamp must follow the rules below: @@ -18,52 +19,54 @@ Time precision in TDengine can be set by the `PRECISION` parameter when executin ```sql CREATE DATABASE db_name PRECISION 'ns'; ``` - ## Data Types In TDengine, the data types below can be used when specifying a column or tag. | # | **type** | **Bytes** | **Description** | -| --- | :-------: | --------- | ------------------------- | -| 1 | TIMESTAMP | 8 | Default precision is millisecond, microsecond and nanosecond are also supported | +| --- | :-------: | --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | +| 1 | TIMESTAMP | 8 | Default precision is millisecond, microsecond and nanosecond are also supported | | 2 | INT | 4 | Integer, the value range is [-2^31, 2^31-1] | -| 3 |INT UNSIGNED|4 | Unsigned integer, the value range is [0, 2^31-1] | +| 3 | INT UNSIGNED| 4| unsigned integer, the value range is [0, 2^32-1] | 4 | BIGINT | 8 | Long integer, the value range is [-2^63, 2^63-1] | -| 5 | BIGINT UNSIGNED | 8 | Unsigned long integer, the value range is [0, 2^63-1] | +| 5 | BIGINT UNSIGNED | 8 | unsigned long integer, the value range is [0, 2^64-1] | | 6 | FLOAT | 4 | Floating point number, the effective number of digits is 6-7, the value range is [-3.4E38, 3.4E38] | | 7 | DOUBLE | 8 | Double precision floating point number, the effective number of digits is 15-16, the value range is [-1.7E308, 1.7E308] | -| 8 | BINARY | User Defined | Single-byte string for ASCII visible characters. Length must be specified when defining a column or tag of binary type. The string length can be up to 16374 bytes. The string value must be quoted with single quotes. The literal single quote inside the string must be preceded with back slash like `\'` | -| 9 | SMALLINT | 2 | Short integer, the value range is [-32768, 32767] | -| 10 | SMALLINT UNSIGNED | 2 | Unsigned short integer, the value range is [0, 32767] | -| 11 | TINYINT | 1 | Single-byte integer, the value range is [-128, 127] | -| 12 | TINYINT UNSIGNED | 1 | Unsigned single-byte integer, the value range is [0, 127] | -| 13 | BOOL | 1 | Bool, the value range is {true, false} | -| 14 | NCHAR | User Defined| Multi-Byte string that can include multi byte characters like Chinese characters. Each character of NCHAR type consumes 4 bytes storage. The string value should be quoted with single quotes. Literal single quote inside the string must be preceded with backslash, like `\’`. The length must be specified when defining a column or tag of NCHAR type, for example nchar(10) means it can store at most 10 characters of nchar type and will consume fixed storage of 40 bytes. An error will be reported if the string value exceeds the length defined. | +| 8 | BINARY | User Defined | Single-byte string for ASCII visible characters. Length must be specified when defining a column or tag of binary type. | +| 9 | SMALLINT | 2 | Short integer, the value range is [-32768, 32767] | +| 10 | INT UNSIGNED| 2| unsigned integer, the value range is [0, 65535]| +| 11 | TINYINT | 1 | Single-byte integer, the value range is [-128, 127] | +| 12 | TINYINT UNSIGNED | 1 | unsigned single-byte integer, the value range is [0, 255] | +| 13 | BOOL | 1 | Bool, the value range is {true, false} | +| 14 | NCHAR | User Defined| Multi-Byte string that can include multi byte characters like Chinese characters. Each character of NCHAR type consumes 4 bytes storage. The string value should be quoted with single quotes. Literal single quote inside the string must be preceded with backslash, like `\’`. The length must be specified when defining a column or tag of NCHAR type, for example nchar(10) means it can store at most 10 characters of nchar type and will consume fixed storage of 40 bytes. An error will be reported if the string value exceeds the length defined. | | 15 | JSON | | JSON type can only be used on tags. A tag of json type is excluded with any other tags of any other type | -| 16 | VARCHAR | User Defined| Alias of BINARY type | +| 16 | VARCHAR | User-defined | Alias of BINARY | + :::note - TDengine is case insensitive and treats any characters in the sql command as lower case by default, case sensitive strings must be quoted with single quotes. -- Only ASCII visible characters are suggested to be used in a column or tag of BINARY type. Multi-byte characters must be stored in NCHAR type. +- Only ASCII visible characters are suggested to be used in a column or tag of BINARY type. Multi-byte characters must be stored in NCHAR type. +- The length of BINARY can be up to 16374 bytes. The string value must be quoted with single quotes. You must specify a length in bytes for a BINARY value, for example binary(20) for up to twenty single-byte characters. If the data exceeds the specified length, an error will occur. The literal single quote inside the string must be preceded with back slash like `\'` - Numeric values in SQL statements will be determined as integer or float type according to whether there is decimal point or whether scientific notation is used, so attention must be paid to avoid overflow. For example, 9999999999999999999 will be considered as overflow because it exceeds the upper limit of long integer, but 9999999999999999999.0 will be considered as a legal float number. ::: + ## Constants -TDengine supports constants of multiple data type. +TDengine supports a variety of constants: | # | **Syntax** | **Type** | **Description** | | --- | :-------: | --------- | -------------------------------------- | -| 1 | [{+ \| -}]123 | BIGINT | Numeric constants are treated as BIGINT type. The value will be truncated if it exceeds the range of BIGINT type. | -| 2 | 123.45 | DOUBLE | Floating number constants are treated as DOUBLE type. TDengine determines whether it's a floating number based on if decimal point or scientific notation is used. | -| 3 | 1.2E3 | DOUBLE | Constants in scientific notation are treated ad DOUBLE type. | -| 4 | 'abc' | BINARY | String constants enclosed by single quotes are treated as BINARY type. Its size is determined as the acutal length. Single quote itself can be included by preceding backslash, i.e. `\'`, in a string constant. | -| 5 | "abc" | BINARY | String constants enclosed by double quotes are treated as BINARY type. Its size is determined as the acutal length. Double quote itself can be included by preceding backslash, i.e. `\"`, in a string constant. | -| 6 | TIMESTAMP {'literal' \| "literal"} | TIMESTAMP | A string constant following `TIMESTAMP` keyword is treated as TIMESTAMP type. The string should be in the format of "YYYY-MM-DD HH:mm:ss.MS". Its time precision is same as that of the current database being used. | -| 7 | {TRUE \| FALSE} | BOOL | BOOL type contant. | -| 8 | {'' \| "" \| '\t' \| "\t" \| ' ' \| " " \| NULL } | -- | NULL constant, it can be used for any type.| +| 1 | [{+ \| -}]123 | BIGINT | Integer literals are of type BIGINT. Data that exceeds the length of the BIGINT type is truncated. | +| 2 | 123.45 | DOUBLE | Floating-point literals are of type DOUBLE. Numeric values will be determined as integer or float type according to whether there is decimal point or whether scientific notation is used. | +| 3 | 1.2E3 | DOUBLE | Literals in scientific notation are of type DOUBLE. | +| 4 | 'abc' | BINARY | Content enclosed in single quotation marks is of type BINARY. The size of a BINARY is the size of the string in bytes. A literal single quote inside the string must be escaped with a backslash (\'). | +| 5 | 'abc' | BINARY | Content enclosed in double quotation marks is of type BINARY. The size of a BINARY is the size of the string in bytes. A literal double quote inside the string must be escaped with a backslash (\"). | +| 6 | TIMESTAMP {'literal' \| "literal"} | TIMESTAMP | The TIMESTAMP keyword indicates that the following string literal is interpreted as a timestamp. The string must be in YYYY-MM-DD HH:mm:ss.MS format. The precision is inherited from the database configuration. | +| 7 | {TRUE \| FALSE} | BOOL | Boolean literals are of type BOOL. | +| 8 | {'' \| "" \| '\t' \| "\t" \| ' ' \| " " \| NULL } | -- | The preceding characters indicate null literals. These can be used with any data type. | :::note -- TDengine determines whether it's a floating number based on if decimal point or scientific notation is used. So whether the value is determined as overflow depends on both the value and the determined type. For example, 9999999999999999999 is determined as overflow because it exceeds the upper limit of BIGINT type, while 9999999999999999999.0 is considered as a valid floating number because it is within the range of DOUBLE type. +Numeric values will be determined as integer or float type according to whether there is decimal point or whether scientific notation is used, so attention must be paid to avoid overflow. For example, 9999999999999999999 will be considered as overflow because it exceeds the upper limit of long integer, but 9999999999999999999.0 will be considered as a legal float number. ::: diff --git a/docs/en/12-taos-sql/02-database.md b/docs/en/12-taos-sql/02-database.md index c2961d62415cd7d23b031777082801426b221190..d9dadae976bf07bbf6cfb49401d55bb0bf18da49 100644 --- a/docs/en/12-taos-sql/02-database.md +++ b/docs/en/12-taos-sql/02-database.md @@ -4,123 +4,153 @@ title: Database description: "create and drop database, show or change database parameters" --- -## Create Database +## Create a Database + +```sql +CREATE DATABASE [IF NOT EXISTS] db_name [database_options] + +database_options: + database_option ... + +database_option: { + BUFFER value + | CACHEMODEL {'none' | 'last_row' | 'last_value' | 'both'} + | CACHESIZE value + | COMP {0 | 1 | 2} + | DURATION value + | WAL_FSYNC_PERIOD value + | MAXROWS value + | MINROWS value + | KEEP value + | PAGES value + | PAGESIZE value + | PRECISION {'ms' | 'us' | 'ns'} + | REPLICA value + | RETENTIONS ingestion_duration:keep_duration ... + | STRICT {'off' | 'on'} + | WAL_LEVEL {1 | 2} + | VGROUPS value + | SINGLE_STABLE {0 | 1} + | WAL_RETENTION_PERIOD value + | WAL_ROLL_PERIOD value + | WAL_RETENTION_SIZE value + | WAL_SEGMENT_SIZE value +} +``` + +## Parameters + +- BUFFER: specifies the size (in MB) of the write buffer for each vnode. Enter a value between 3 and 16384. The default value is 96. +- CACHEMODEL: specifies how the latest data in subtables is stored in the cache. The default value is none. + - none: The latest data is not cached. + - last_row: The last row of each subtable is cached. This option significantly improves the performance of the LAST_ROW function. + - last_value: The last non-null value of each column in each subtable is cached. This option significantly improves the performance of the LAST function under normal circumstances, such as statements including the WHERE, ORDER BY, GROUP BY, and INTERVAL keywords. + - both: The last row of each subtable and the last non-null value of each column in each subtable are cached. +- CACHESIZE: specifies the amount (in MB) of memory used for subtable caching on each vnode. Enter a value between 1 and 65536. The default value is 1. +- COMP: specifies how databases are compressed. The default value is 2. + - 0: Compression is disabled. + - 1: One-pass compression is enabled. + - 2: Two-pass compression is enabled. +- DURATION: specifies the time period contained in each data file. After the time specified by this parameter has elapsed, TDengine creates a new data file to store incoming data. You can use m (minutes), h (hours), and d (days) as the unit, for example DURATION 100h or DURATION 10d. If you do not include a unit, d is used by default. +- WAL_FSYNC_PERIOD: specifies the interval (in milliseconds) at which data is written from the WAL to disk. This parameter takes effect only when the WAL parameter is set to 2. The default value is 3000. Enter a value between 0 and 180000. The value 0 indicates that incoming data is immediately written to disk. +- MAXROWS: specifies the maximum number of rows recorded in a block. The default value is 4096. +- MINROWS: specifies the minimum number of rows recorded in a block. The default value is 100. +- KEEP: specifies the time for which data is retained. Enter a value between 1 and 365000. The default value is 3650. The value of the KEEP parameter must be greater than or equal to the value of the DURATION parameter. TDengine automatically deletes data that is older than the value of the KEEP parameter. You can use m (minutes), h (hours), and d (days) as the unit, for example KEEP 100h or KEEP 10d. If you do not include a unit, d is used by default. +- PAGES: specifies the number of pages in the metadata storage engine cache on each vnode. Enter a value greater than or equal to 64. The default value is 256. The space occupied by metadata storage on each vnode is equal to the product of the values of the PAGESIZE and PAGES parameters. The space occupied by default is 1 MB. +- PAGESIZE: specifies the size (in KB) of each page in the metadata storage engine cache on each vnode. The default value is 4. Enter a value between 1 and 16384. +- PRECISION: specifies the precision at which a database records timestamps. Enter ms for milliseconds, us for microseconds, or ns for nanoseconds. The default value is ms. +- REPLICA: specifies the number of replicas that are made of the database. Enter 1 or 3. The default value is 1. The value of the REPLICA parameter cannot exceed the number of dnodes in the cluster. +- RETENTIONS: specifies the retention period for data aggregated at various intervals. For example, RETENTIONS 15s:7d,1m:21d,15m:50d indicates that data aggregated every 15 seconds is retained for 7 days, data aggregated every 1 minute is retained for 21 days, and data aggregated every 15 minutes is retained for 50 days. You must enter three aggregation intervals and corresponding retention periods. +- STRICT: specifies whether strong data consistency is enabled. The default value is off. + - on: Strong consistency is enabled and implemented through the Raft consensus algorithm. In this mode, an operation is considered successful once it is confirmed by half of the nodes in the cluster. + - off: Strong consistency is disabled. In this mode, an operation is considered successful when it is initiated by the local node. +- WAL_LEVEL: specifies whether fsync is enabled. The default value is 1. + - 1: WAL is enabled but fsync is disabled. + - 2: WAL and fsync are both enabled. +- VGROUPS: specifies the initial number of vgroups when a database is created. +- SINGLE_STABLE: specifies whether the database can contain more than one supertable. + - 0: The database can contain multiple supertables. + - 1: The database can contain only one supertable. +- WAL_RETENTION_PERIOD: specifies the time after which WAL files are deleted. This parameter is used for data subscription. Enter a time in seconds. The default value is 0. A value of 0 indicates that each WAL file is deleted immediately after its contents are written to disk. -1: WAL files are never deleted. +- WAL_RETENTION_SIZE: specifies the size at which WAL files are deleted. This parameter is used for data subscription. Enter a size in KB. The default value is 0. A value of 0 indicates that each WAL file is deleted immediately after its contents are written to disk. -1: WAL files are never deleted. +- WAL_ROLL_PERIOD: specifies the time after which WAL files are rotated. After this period elapses, a new WAL file is created. The default value is 0. A value of 0 indicates that a new WAL file is created only after the previous WAL file was written to disk. +- WAL_SEGMENT_SIZE: specifies the maximum size of a WAL file. After the current WAL file reaches this size, a new WAL file is created. The default value is 0. A value of 0 indicates that a new WAL file is created only after the previous WAL file was written to disk. + +### Example Statement + +```sql +create database if not exists db vgroups 10 buffer 10 + +``` + +The preceding SQL statement creates a database named db that has 10 vgroups and whose vnodes have a 10 MB cache. + +### Specify the Database in Use ``` -CREATE DATABASE [IF NOT EXISTS] db_name [KEEP keep] [DAYS days] [UPDATE 1]; +USE db_name; ``` -:::info - -1. KEEP specifies the number of days for which the data in the database will be retained. The default value is 3650 days, i.e. 10 years. The data will be deleted automatically once its age exceeds this threshold. -2. UPDATE specifies whether the data can be updated and how the data can be updated. - 1. UPDATE set to 0 means update operation is not allowed. The update for data with an existing timestamp will be discarded silently and the original record in the database will be preserved as is. - 2. UPDATE set to 1 means the whole row will be updated. The columns for which no value is specified will be set to NULL. - 3. UPDATE set to 2 means updating a subset of columns for a row is allowed. The columns for which no value is specified will be kept unchanged. -3. The maximum length of database name is 33 bytes. -4. The maximum length of a SQL statement is 65,480 bytes. -5. Below are the parameters that can be used when creating a database - - cache: [Description](/reference/config/#cache) - - blocks: [Description](/reference/config/#blocks) - - days: [Description](/reference/config/#days) - - keep: [Description](/reference/config/#keep) - - minRows: [Description](/reference/config/#minrows) - - maxRows: [Description](/reference/config/#maxrows) - - wal: [Description](/reference/config/#wallevel) - - fsync: [Description](/reference/config/#fsync) - - update: [Description](/reference/config/#update) - - cacheLast: [Description](/reference/config/#cachelast) - - replica: [Description](/reference/config/#replica) - - quorum: [Description](/reference/config/#quorum) - - comp: [Description](/reference/config/#comp) - - precision: [Description](/reference/config/#precision) -6. Please note that all of the parameters mentioned in this section are configured in configuration file `taos.cfg` on the TDengine server. If not specified in the `create database` statement, the values from taos.cfg are used by default. To override default parameters, they must be specified in the `create database` statement. - -::: +The preceding SQL statement switches to the specified database. (If you connect to TDengine over the REST API, this statement does not take effect.) -## Show Current Configuration +## Drop a Database ``` -SHOW VARIABLES; +DROP DATABASE [IF EXISTS] db_name ``` -## Specify The Database In Use +The preceding SQL statement deletes the specified database. This statement will delete all tables in the database and destroy all vgroups associated with it. Exercise caution when using this statement. -``` -USE db_name; -``` - -:::note -This way is not applicable when using a REST connection. In a REST connection the database name must be specified before a table or stable name. For e.g. to query the stable "meters" in database "test" the query would be "SELECT count(*) from test.meters" +## Change Database Configuration -::: +```sql +ALTER DATABASE db_name [alter_database_options] -## Drop Database +alter_database_options: + alter_database_option ... -``` -DROP DATABASE [IF EXISTS] db_name; +alter_database_option: { + CACHEMODEL {'none' | 'last_row' | 'last_value' | 'both'} + | CACHESIZE value + | WAL_LEVEL value + | WAL_FSYNC_PERIOD value + | KEEP value +} ``` :::note -All data in the database will be deleted too. This command must be used with extreme caution. Please follow your organization's data integrity, data backup, data security or any other applicable SOPs before using this command. +Other parameters cannot be modified after the database has been created. ::: -## Change Database Configuration +## View Databases -Some examples are shown below to demonstrate how to change the configuration of a database. Please note that some configuration parameters can be changed after the database is created, but some cannot. For details of the configuration parameters of database please refer to [Configuration Parameters](/reference/config/). +### View All Databases ``` -ALTER DATABASE db_name COMP 2; -``` - -COMP parameter specifies whether the data is compressed and how the data is compressed. - -``` -ALTER DATABASE db_name REPLICA 2; -``` - -REPLICA parameter specifies the number of replicas of the database. - -``` -ALTER DATABASE db_name KEEP 365; +SHOW DATABASES; ``` -KEEP parameter specifies the number of days for which the data will be kept. +### View the CREATE Statement for a Database ``` -ALTER DATABASE db_name QUORUM 2; +SHOW CREATE DATABASE db_name; ``` -QUORUM parameter specifies the necessary number of confirmations to determine whether the data is written successfully. +The preceding SQL statement can be used in migration scenarios. This command can be used to get the CREATE statement, which can be used in another TDengine instance to create the exact same database. -``` -ALTER DATABASE db_name BLOCKS 100; -``` +### View Database Configuration -BLOCKS parameter specifies the number of memory blocks used by each VNODE. - -``` -ALTER DATABASE db_name CACHELAST 0; +```sql +SHOW DATABASES \G; ``` -CACHELAST parameter specifies whether and how the latest data of a sub table is cached. +The preceding SQL statement shows the value of each parameter for the specified database. One value is displayed per line. -:::tip -The above parameters can be changed using `ALTER DATABASE` command without restarting. For more details of all configuration parameters please refer to [Configuration Parameters](/reference/config/). - -::: +## Delete Expired Data -## Show All Databases - -``` -SHOW DATABASES; -``` - -## Show The Create Statement of A Database - -``` -SHOW CREATE DATABASE db_name; +```sql +TRIM DATABASE db_name; ``` -This command is useful when migrating the data from one TDengine cluster to another. This command can be used to get the CREATE statement, which can be used in another TDengine instance to create the exact same database. +The preceding SQL statement deletes data that has expired and orders the remaining data in accordance with the storage configuration. diff --git a/docs/en/12-taos-sql/03-table.md b/docs/en/12-taos-sql/03-table.md index f065a8e2396583bb7a512446b513ed60056ad55e..bf32cf171bbeea23ada946d5011a73dd70ddd6ca 100644 --- a/docs/en/12-taos-sql/03-table.md +++ b/docs/en/12-taos-sql/03-table.md @@ -1,127 +1,198 @@ --- -sidebar_label: Table title: Table -description: create super table, normal table and sub table, drop tables and change tables --- ## Create Table -``` -CREATE TABLE [IF NOT EXISTS] tb_name (timestamp_field_name TIMESTAMP, field1_name data_type1 [, field2_name data_type2 ...]); -``` - -:::info +You create standard tables and subtables with the `CREATE TABLE` statement. + +```sql +CREATE TABLE [IF NOT EXISTS] [db_name.]tb_name (create_definition [, create_definitionn] ...) [table_options] + +CREATE TABLE create_subtable_clause + +CREATE TABLE [IF NOT EXISTS] [db_name.]tb_name (create_definition [, create_definitionn] ...) + [TAGS (create_definition [, create_definitionn] ...)] + [table_options] + +create_subtable_clause: { + create_subtable_clause [create_subtable_clause] ... + | [IF NOT EXISTS] [db_name.]tb_name USING [db_name.]stb_name [(tag_name [, tag_name] ...)] TAGS (tag_value [, tag_value] ...) +} + +create_definition: + col_name column_definition + +column_definition: + type_name [comment 'string_value'] + +table_options: + table_option ... + +table_option: { + COMMENT 'string_value' + | WATERMARK duration[,duration] + | MAX_DELAY duration[,duration] + | ROLLUP(func_name [, func_name] ...) + | SMA(col_name [, col_name] ...) + | TTL value +} + +``` + +**More explanations** 1. The first column of a table MUST be of type TIMESTAMP. It is automatically set as the primary key. 2. The maximum length of the table name is 192 bytes. 3. The maximum length of each row is 48k bytes, please note that the extra 2 bytes used by each BINARY/NCHAR column are also counted. 4. The name of the subtable can only consist of characters from the English alphabet, digits and underscore. Table names can't start with a digit. Table names are case insensitive. 5. The maximum length in bytes must be specified when using BINARY or NCHAR types. -6. Escape character "\`" can be used to avoid the conflict between table names and reserved keywords, above rules will be bypassed when using escape character on table names, but the upper limit for the name length is still valid. The table names specified using escape character are case sensitive. Only ASCII visible characters can be used with escape character. +6. Escape character "\`" can be used to avoid the conflict between table names and reserved keywords, above rules will be bypassed when using escape character on table names, but the upper limit for the name length is still valid. The table names specified using escape character are case sensitive. For example \`aBc\` and \`abc\` are different table names but `abc` and `aBc` are same table names because they are both converted to `abc` internally. + Only ASCII visible characters can be used with escape character. -::: +**Parameter description** +1. COMMENT: specifies comments for the table. This parameter can be used with supertables, standard tables, and subtables. +2. WATERMARK: specifies the time after which the window is closed. The default value is 5 seconds. Enter a value between 0 and 15 minutes in milliseconds, seconds, or minutes. You can enter multiple values separated by commas (,). This parameter applies only to supertables and takes effect only when the RETENTIONS parameter has been specified for the database. +3. MAX_DELAY: specifies the maximum latency for pushing computation results. The default value is 15 minutes or the value of the INTERVAL parameter, whichever is smaller. Enter a value between 0 and 15 minutes in milliseconds, seconds, or minutes. You can enter multiple values separated by commas (,). Note: Retain the default value if possible. Configuring a small MAX_DELAY may cause results to be frequently pushed, affecting storage and query performance. This parameter applies only to supertables and takes effect only when the RETENTIONS parameter has been specified for the database. +4. ROLLUP: specifies aggregate functions to roll up. Rolling up a function provides downsampled results based on multiple axes. This parameter applies only to supertables and takes effect only when the RETENTIONS parameter has been specified for the database. You can specify only one function to roll up. The rollup takes effect on all columns except TS. Enter one of the following values: avg, sum, min, max, last, or first. +5. SMA: specifies functions on which to enable small materialized aggregates (SMA). SMA is user-defined precomputation of aggregates based on data blocks. Enter one of the following values: max, min, or sum This parameter can be used with supertables and standard tables. +6. TTL: specifies the time to live (TTL) for the table. If the period specified by the TTL parameter elapses without any data being written to the table, TDengine will automatically delete the table. Note: The system may not delete the table at the exact moment that the TTL expires. Enter a value in days. The default value is 0. Note: The TTL parameter has a higher priority than the KEEP parameter. If a table is marked for deletion because the TTL has expired, it will be deleted even if the time specified by the KEEP parameter has not elapsed. This parameter can be used with standard tables and subtables. -### Create Subtable Using STable As Template +## Create Subtables -``` +### Create a Subtable + +```sql CREATE TABLE [IF NOT EXISTS] tb_name USING stb_name TAGS (tag_value1, ...); ``` -The above command creates a subtable using the specified super table as a template and the specified tag values. +### Create a Subtable with Specified Tags -### Create Subtable Using STable As Template With A Subset of Tags - -``` +```sql CREATE TABLE [IF NOT EXISTS] tb_name USING stb_name (tag_name1, ...) TAGS (tag_value1, ...); ``` -The tags for which no value is specified will be set to NULL. +The preceding SQL statement creates a subtable based on a supertable but specifies a subset of tags to use. Tags that are not included in this subset are assigned a null value. -### Create Tables in Batch +### Create Multiple Subtables -``` +```sql CREATE TABLE [IF NOT EXISTS] tb_name1 USING stb_name TAGS (tag_value1, ...) [IF NOT EXISTS] tb_name2 USING stb_name TAGS (tag_value2, ...) ...; ``` -This can be used to create a lot of tables in a single SQL statement while making table creation much faster. +You can create multiple subtables in a single SQL statement provided that all subtables use the same supertable. For performance reasons, do not create more than 3000 tables per statement. + +## Modify a Table -:::info +```sql +ALTER TABLE [db_name.]tb_name alter_table_clause + +alter_table_clause: { + alter_table_options + | ADD COLUMN col_name column_type + | DROP COLUMN col_name + | MODIFY COLUMN col_name column_type + | RENAME COLUMN old_col_name new_col_name +} + +alter_table_options: + alter_table_option ... + +alter_table_option: { + TTL value + | COMMENT 'string_value' +} -- Creating tables in batch must use a super table as a template. -- The length of single statement is suggested to be between 1,000 and 3,000 bytes for best performance. +``` -::: +**More explanations** +You can perform the following modifications on existing tables: +1. ADD COLUMN: adds a column to the supertable. +2. DROP COLUMN: deletes a column from the supertable. +3. MODIFY COLUMN: changes the length of the data type specified for the column. Note that you can only specify a length greater than the current length. +4. RENAME COLUMN: renames a specified column in the table. -## Drop Tables +### Add a Column -``` -DROP TABLE [IF EXISTS] tb_name; +```sql +ALTER TABLE tb_name ADD COLUMN field_name data_type; ``` -## Show All Tables In Current Database +### Delete a Column -``` -SHOW TABLES [LIKE tb_name_wildcard]; +```sql +ALTER TABLE tb_name DROP COLUMN field_name; ``` -## Show Create Statement of A Table +### Modify the Data Length +```sql +ALTER TABLE tb_name MODIFY COLUMN field_name data_type(length); ``` -SHOW CREATE TABLE tb_name; -``` - -This is useful when migrating the data in one TDengine cluster to another one because it can be used to create the exact same tables in the target database. -## Show Table Definition +### Rename a Column +```sql +ALTER TABLE tb_name RENAME COLUMN old_col_name new_col_name ``` -DESCRIBE tb_name; + +## Modify a Subtable + +```sql +ALTER TABLE [db_name.]tb_name alter_table_clause + +alter_table_clause: { + alter_table_options + | SET TAG tag_name = new_tag_value +} + +alter_table_options: + alter_table_option ... + +alter_table_option: { + TTL value + | COMMENT 'string_value' +} ``` -## Change Table Definition +**More explanations** +1. Only the value of a tag can be modified directly. For all other modifications, you must modify the supertable from which the subtable was created. -### Add A Column +### Change Tag Value Of Sub Table ``` -ALTER TABLE tb_name ADD COLUMN field_name data_type; +ALTER TABLE tb_name SET TAG tag_name=new_tag_value; ``` -:::info +## Delete a Table -1. The maximum number of columns is 4096, the minimum number of columns is 2. -2. The maximum length of a column name is 64 bytes. +The following SQL statement deletes one or more tables. -::: - -### Remove A Column - -``` -ALTER TABLE tb_name DROP COLUMN field_name; +```sql +DROP TABLE [IF EXISTS] [db_name.]tb_name [, [IF EXISTS] [db_name.]tb_name] ... ``` -:::note -If a table is created using a super table as template, the table definition can only be changed on the corresponding super table, and the change will be automatically applied to all the subtables created using this super table as template. For tables created in the normal way, the table definition can be changed directly on the table. +## View Tables -::: +### View All Tables -### Change Column Length +The following SQL statement shows all tables in the current database. -``` -ALTER TABLE tb_name MODIFY COLUMN field_name data_type(length); +```sql +SHOW TABLES [LIKE tb_name_wildchar]; ``` -If the type of a column is variable length, like BINARY or NCHAR, this command can be used to change the length of the column. +### View the CREATE Statement for a Table -:::note -If a table is created using a super table as template, the table definition can only be changed on the corresponding super table, and the change will be automatically applied to all the subtables created using this super table as template. For tables created in the normal way, the table definition can be changed directly on the table. +``` +SHOW CREATE TABLE tb_name; +``` -::: +This command is useful in migrating data from one TDengine cluster to another because it can be used to create the exact same tables in the target database. -### Change Tag Value Of Sub Table +## View the Table Schema ``` -ALTER TABLE tb_name SET TAG tag_name=new_tag_value; -``` - -This command can be used to change the tag value if the table is created using a super table as template. +DESCRIBE [db_name.]tb_name; +``` \ No newline at end of file diff --git a/docs/en/12-taos-sql/04-stable.md b/docs/en/12-taos-sql/04-stable.md index b8a608792ab327a81129d29ddd0ff44d7af6e6c5..6a0a0922cce7d9f831f333e4999789798be8d867 100644 --- a/docs/en/12-taos-sql/04-stable.md +++ b/docs/en/12-taos-sql/04-stable.md @@ -1,118 +1,159 @@ --- -sidebar_label: STable -title: Super Table +sidebar_label: Supertable +title: Supertable --- -:::note +## Create a Supertable -Keyword `STable`, abbreviated for super table, is supported since version 2.0.15. +```sql +CREATE STABLE [IF NOT EXISTS] stb_name (create_definition [, create_definitionn] ...) TAGS (create_definition [, create_definition] ...) [table_options] + +create_definition: + col_name column_definition + +column_definition: + type_name [COMMENT 'string_value'] +``` -::: +**More explanations** +- Each supertable can have a maximum of 4096 columns, including tags. The minimum number of columns is 3: a timestamp column used as the key, one tag column, and one data column. +- When you create a supertable, you can add comments to columns and tags. +- The TAGS keyword defines the tag columns for the supertable. The following restrictions apply to tag columns: + - A tag column can use the TIMESTAMP data type, but the values in the column must be fixed numbers. Timestamps including formulae, such as "now + 10s", cannot be stored in a tag column. + - The name of a tag column cannot be the same as the name of any other column. + - The name of a tag column cannot be a reserved keyword. + - Each supertable must contain between 1 and 128 tags. The total length of the TAGS keyword cannot exceed 16 KB. +- For more information about table parameters, see Create a Table. -## Create STable +## View a Supertable + +### View All Supertables ``` -CREATE STable [IF NOT EXISTS] stb_name (timestamp_field_name TIMESTAMP, field1_name data_type1 [, field2_name data_type2 ...]) TAGS (tag1_name tag_type1, tag2_name tag_type2 [, tag3_name tag_type3]); +SHOW STABLES [LIKE tb_name_wildcard]; ``` -The SQL statement of creating a STable is similar to that of creating a table, but a special column set named `TAGS` must be specified with the names and types of the tags. - -:::info +The preceding SQL statement shows all supertables in the current TDengine database, including the name, creation time, number of columns, number of tags, and number of subtabels for each supertable. -1. A tag can be of type timestamp, since version 2.1.3.0, but its value must be fixed and arithmetic operations cannot be performed on it. Prior to version 2.1.3.0, tag types specified in TAGS could not be of type timestamp. -2. The tag names specified in TAGS should NOT be the same as other columns. -3. The tag names specified in TAGS should NOT be the same as any reserved keywords.(Please refer to [keywords](/taos-sql/keywords/) -4. The maximum number of tags specified in TAGS is 128, there must be at least one tag, and the total length of all tag columns should NOT exceed 16KB. - -::: - -## Drop STable +### View the CREATE Statement for a Supertable ``` -DROP STable [IF EXISTS] stb_name; +SHOW CREATE STABLE stb_name; ``` -All the subtables created using the deleted STable will be deleted automatically. +The preceding SQL statement can be used in migration scenarios. It returns the CREATE statement that was used to create the specified supertable. You can then use the returned statement to create an identical supertable on another TDengine database. -## Show All STables +## View the Supertable Schema ``` -SHOW STableS [LIKE tb_name_wildcard]; +DESCRIBE [db_name.]stb_name; ``` -This command can be used to display the information of all STables in the current database, including name, creation time, number of columns, number of tags, and number of tables created using this STable. - -## Show The Create Statement of A STable +## Drop STable ``` -SHOW CREATE STable stb_name; +DROP STABLE [IF EXISTS] [db_name.]stb_name ``` -This command is useful in migrating data from one TDengine cluster to another because it can be used to create the exact same STable in the target database. +Note: Deleting a supertable will delete all subtables created from the supertable, including all data within those subtables. -## Get STable Definition +## Modify a Supertable + +```sql +ALTER STABLE [db_name.]tb_name alter_table_clause + +alter_table_clause: { + alter_table_options + | ADD COLUMN col_name column_type + | DROP COLUMN col_name + | MODIFY COLUMN col_name column_type + | ADD TAG tag_name tag_type + | DROP TAG tag_name + | MODIFY TAG tag_name tag_type + | RENAME TAG old_tag_name new_tag_name +} + +alter_table_options: + alter_table_option ... + +alter_table_option: { + COMMENT 'string_value' +} ``` -DESCRIBE stb_name; -``` -## Change Columns Of STable +**More explanations** + +Modifications to the table schema of a supertable take effect on all subtables within the supertable. You cannot modify the table schema of subtables individually. When you modify the tag schema of a supertable, the modifications automatically take effect on all of its subtables. + +- ADD COLUMN: adds a column to the supertable. +- DROP COLUMN: deletes a column from the supertable. +- MODIFY COLUMN: changes the length of a BINARY or NCHAR column. Note that you can only specify a length greater than the current length. +- ADD TAG: adds a tag to the supertable. +- DROP TAG: deletes a tag from the supertable. When you delete a tag from a supertable, it is automatically deleted from all subtables within the supertable. +- MODIFY TAG: modifies the definition of a tag in the supertable. You can use this keyword to change the length of a BINARY or NCHAR tag column. Note that you can only specify a length greater than the current length. +- RENAME TAG: renames a specified tag in the supertable. When you rename a tag in a supertable, it is automatically renamed in all subtables within the supertable. -### Add A Column +### Add a Column ``` -ALTER STable stb_name ADD COLUMN field_name data_type; +ALTER STABLE stb_name ADD COLUMN col_name column_type; ``` -### Remove A Column +### Delete a Column ``` -ALTER STable stb_name DROP COLUMN field_name; +ALTER STABLE stb_name DROP COLUMN col_name; ``` -### Change Column Length +### Modify the Data Length ``` -ALTER STable stb_name MODIFY COLUMN field_name data_type(length); +ALTER STABLE stb_name MODIFY COLUMN col_name data_type(length); ``` -This command can be used to change (or more specifically, increase) the length of a column of variable length types, like BINARY or NCHAR. - -## Change Tags of A STable +The preceding SQL statement changes the length of a BINARY or NCHAR data column. Note that you can only specify a length greater than the current length. ### Add A Tag ``` -ALTER STable stb_name ADD TAG new_tag_name tag_type; +ALTER STABLE stb_name ADD TAG tag_name tag_type; ``` -This command is used to add a new tag for a STable and specify the tag type. +The preceding SQL statement adds a tag of the specified type to the supertable. A supertable cannot contain more than 128 tags. The total length of all tags in a supertable cannot exceed 16 KB. ### Remove A Tag ``` -ALTER STable stb_name DROP TAG tag_name; +ALTER STABLE stb_name DROP TAG tag_name; ``` -The tag will be removed automatically from all the subtables, created using the super table as template, once a tag is removed from a super table. +The preceding SQL statement deletes a tag from the supertable. When you delete a tag from a supertable, it is automatically deleted from all subtables within the supertable. ### Change A Tag ``` -ALTER STable stb_name CHANGE TAG old_tag_name new_tag_name; +ALTER STABLE stb_name RENAME TAG old_tag_name new_tag_name; ``` -The tag name will be changed automatically for all the subtables, created using the super table as template, once a tag name is changed for a super table. +The preceding SQL statement renames a tag in the supertable. When you rename a tag in a supertable, it is automatically renamed in all subtables within the supertable. ### Change Tag Length ``` -ALTER STable stb_name MODIFY TAG tag_name data_type(length); +ALTER STABLE stb_name MODIFY TAG tag_name data_type(length); ``` -This command can be used to change (or more specifically, increase) the length of a tag of variable length types, like BINARY or NCHAR. +The preceding SQL statement changes the length of a BINARY or NCHAR tag column. Note that you can only specify a length greater than the current length. (Available in 2.1.3.0 and later versions) + +### View a Supertable +You can run projection and aggregate SELECT queries on supertables, and you can filter by tag or column by using the WHERE keyword. + +If you do not include an ORDER BY clause, results are returned by subtable. These results are not ordered. You can include an ORDER BY clause in your query to strictly order the results. + + :::note -Changing tag values can be applied to only subtables. All other tag operations, like add tag, remove tag, however, can be applied to only STable. If a new tag is added for a STable, the tag will be added with NULL value for all its subtables. +All tag operations except for updating the value of a tag must be performed on the supertable and not on individual subtables. If you add a tag to an existing supertable, the tag is automatically added with a null value to all subtables within the supertable. ::: diff --git a/docs/en/12-taos-sql/05-insert.md b/docs/en/12-taos-sql/05-insert.md index 1336cd7238a19190583ea9d268a64df242ffd3c9..e7d56fb3c734affa92c8c71c190b1132cd89e335 100644 --- a/docs/en/12-taos-sql/05-insert.md +++ b/docs/en/12-taos-sql/05-insert.md @@ -1,4 +1,5 @@ --- +sidebar_label: Insert title: Insert --- @@ -17,47 +18,62 @@ INSERT INTO ...]; ``` -## Insert Single or Multiple Rows +**Timestamps** -Single row or multiple rows specified with VALUES can be inserted into a specific table. For example: +1. All data writes must include a timestamp. With regard to timestamps, note the following: -A single row is inserted using the below statement. +2. The precision of a timestamp depends on its format. The precision configured for the database affects only timestamps that are inserted as long integers (UNIX time). Timestamps inserted as date and time strings are not affected. As an example, the timestamp 2021-07-13 16:16:48 is equivalent to 1626164208 in UNIX time. This UNIX time is modified to 1626164208000 for databases with millisecond precision, 1626164208000000 for databases with microsecond precision, and 1626164208000000000 for databases with nanosecond precision. -```sq; +3. If you want to insert multiple rows simultaneously, do not use the NOW function in the timestamp. Using the NOW function in this situation will cause multiple rows to have the same timestamp and prevent them from being stored correctly. This is because the NOW function obtains the current time on the client, and multiple instances of NOW in a single statement will return the same time. + The earliest timestamp that you can use when inserting data is equal to the current time on the server minus the value of the KEEP parameter. The latest timestamp that you can use when inserting data is equal to the current time on the server plus the value of the DURATION parameter. You can configure the KEEP and DURATION parameters when you create a database. The default values are 3650 days for the KEEP parameter and 10 days for the DURATION parameter. + +**Syntax** + +1. The USING clause automatically creates the specified subtable if it does not exist. If it's unknown whether the table already exists, the table can be created automatically while inserting using the SQL statement below. To use this functionality, a STable must be used as template and tag values must be provided. Any tags that you do not specify will be assigned a null value. + +2. You can insert data into specified columns. Any columns in which you do not insert data will be assigned a null value. + +3. The VALUES clause inserts one or more rows of data into a table. + +4. The FILE clause inserts tags or data from a comma-separates values (CSV) file. Do not include headers in your CSV files. + +5. A single INSERT statement can write data to multiple tables. + +6. The INSERT statement is fully parsed before being executed, so that if any element of the statement fails, the entire statement will fail. For example, the following statement will not create a table because the latter part of the statement is invalid: + + ```sql + INSERT INTO d1001 USING meters TAGS('Beijing.Chaoyang', 2) VALUES('a'); + ``` + +7. However, an INSERT statement that writes data to multiple subtables can succeed for some tables and fail for others. This situation is caused because vnodes perform write operations independently of each other. One vnode failing to write data does not affect the ability of other vnodes to write successfully. + +## Insert a Record + +Single row or multiple rows specified with VALUES can be inserted into a specific table. A single row is inserted using the below statement. + +```sql INSERT INTO d1001 VALUES (NOW, 10.2, 219, 0.32); ``` +## Insert Multiple Records + Double rows are inserted using the below statement. ```sql INSERT INTO d1001 VALUES ('2021-07-13 14:06:32.272', 10.2, 219, 0.32) (1626164208000, 10.15, 217, 0.33); ``` -:::note - -1. In the second example above, different formats are used in the two rows to be inserted. In the first row, the timestamp format is a date and time string, which is interpreted from the string value only. In the second row, the timestamp format is a long integer, which will be interpreted based on the database time precision. -2. When trying to insert multiple rows in a single statement, only the timestamp of one row can be set as NOW, otherwise there will be duplicate timestamps among the rows and the result may be out of expectation because NOW will be interpreted as the time when the statement is executed. -3. The oldest timestamp that is allowed is subtracting the KEEP parameter from current time. -4. The newest timestamp that is allowed is adding the DAYS parameter to current time. - -::: - -## Insert Into Specific Columns +## Write to a Specified Column -Data can be inserted into specific columns, either single row or multiple row, while other columns will be inserted as NULL value. +Data can be inserted into specific columns, either single row or multiple row, while other columns will be inserted as NULL value. The key (timestamp) cannot be null. For example: -``` +```sql INSERT INTO d1001 (ts, current, phase) VALUES ('2021-07-13 14:06:33.196', 10.27, 0.31); ``` -:::info -If no columns are explicitly specified, all the columns must be provided with values, this is called "all column mode". The insert performance of all column mode is much better than specifying a subset of columns, so it's encouraged to use "all column mode" while providing NULL value explicitly for the columns for which no actual value can be provided. - -::: - ## Insert Into Multiple Tables -One or multiple rows can be inserted into multiple tables in a single SQL statement, with or without specifying specific columns. +One or multiple rows can be inserted into multiple tables in a single SQL statement, with or without specifying specific columns. For example: ```sql INSERT INTO d1001 VALUES ('2021-07-13 14:06:34.630', 10.2, 219, 0.32) ('2021-07-13 14:06:35.779', 10.15, 217, 0.33) @@ -66,19 +82,19 @@ INSERT INTO d1001 VALUES ('2021-07-13 14:06:34.630', 10.2, 219, 0.32) ('2021-07- ## Automatically Create Table When Inserting -If it's unknown whether the table already exists, the table can be created automatically while inserting using the SQL statement below. To use this functionality, a STable must be used as template and tag values must be provided. +If it's unknown whether the table already exists, the table can be created automatically while inserting using the SQL statement below. To use this functionality, a STable must be used as template and tag values must be provided. For example: ```sql INSERT INTO d21001 USING meters TAGS ('California.SanFrancisco', 2) VALUES ('2021-07-13 14:06:32.272', 10.2, 219, 0.32); ``` -It's not necessary to provide values for all tags when creating tables automatically, the tags without values provided will be set to NULL. +It's not necessary to provide values for all tags when creating tables automatically, the tags without values provided will be set to NULL. For example: ```sql INSERT INTO d21001 USING meters (groupId) TAGS (2) VALUES ('2021-07-13 14:06:33.196', 10.15, 217, 0.33); ``` -Multiple rows can also be inserted into the same table in a single SQL statement. +Multiple rows can also be inserted into the same table in a single SQL statement. For example: ```sql INSERT INTO d21001 USING meters TAGS ('California.SanFrancisco', 2) VALUES ('2021-07-13 14:06:34.630', 10.2, 219, 0.32) ('2021-07-13 14:06:35.779', 10.15, 217, 0.33) @@ -86,10 +102,6 @@ INSERT INTO d21001 USING meters TAGS ('California.SanFrancisco', 2) VALUES ('202 d21003 USING meters (groupId) TAGS (2) (ts, current, phase) VALUES ('2021-07-13 14:06:34.255', 10.27, 0.31); ``` -:::info -Prior to version 2.0.20.5, when using `INSERT` to create tables automatically and specifying the columns, the column names must follow the table name immediately. From version 2.0.20.5, the column names can follow the table name immediately, also can be put between `TAGS` and `VALUES`. In the same SQL statement, however, these two ways of specifying column names can't be mixed. -::: - ## Insert Rows From A File Besides using `VALUES` to insert one or multiple rows, the data to be inserted can also be prepared in a CSV file with comma as separator and each field value quoted by single quotes. Table definition is not required in the CSV file. For example, if file "/tmp/csvfile.csv" contains the below data: @@ -107,58 +119,13 @@ INSERT INTO d1001 FILE '/tmp/csvfile.csv'; ## Create Tables Automatically and Insert Rows From File -From version 2.1.5.0, tables can be automatically created using a super table as template when inserting data from a CSV file, like below: - ```sql INSERT INTO d21001 USING meters TAGS ('California.SanFrancisco', 2) FILE '/tmp/csvfile.csv'; ``` -Multiple tables can be automatically created and inserted in a single SQL statement, like below: +When writing data from a file, you can automatically create the specified subtable if it does not exist. For example: ```sql INSERT INTO d21001 USING meters TAGS ('California.SanFrancisco', 2) FILE '/tmp/csvfile_21001.csv' d21002 USING meters (groupId) TAGS (2) FILE '/tmp/csvfile_21002.csv'; ``` - -## More About Insert - -For SQL statement like `insert`, a stream parsing strategy is applied. That means before an error is found and the execution is aborted, the part prior to the error point has already been executed. Below is an experiment to help understand the behavior. - -First, a super table is created. - -```sql -CREATE TABLE meters(ts TIMESTAMP, current FLOAT, voltage INT, phase FLOAT) TAGS(location BINARY(30), groupId INT); -``` - -It can be proven that the super table has been created by `SHOW STableS`, but no table exists using `SHOW TABLES`. - -``` -taos> SHOW STableS; - name | created_time | columns | tags | tables | -============================================================================================ - meters | 2020-08-06 17:50:27.831 | 4 | 2 | 0 | -Query OK, 1 row(s) in set (0.001029s) - -taos> SHOW TABLES; -Query OK, 0 row(s) in set (0.000946s) -``` - -Then, try to create table d1001 automatically when inserting data into it. - -```sql -INSERT INTO d1001 USING meters TAGS('California.SanFrancisco', 2) VALUES('a'); -``` - -The output shows the value to be inserted is invalid. But `SHOW TABLES` proves that the table has been created automatically by the `INSERT` statement. - -``` -DB error: invalid SQL: 'a' (invalid timestamp) (0.039494s) - -taos> SHOW TABLES; - table_name | created_time | columns | STable_name | -====================================================================================================== - d1001 | 2020-08-06 17:52:02.097 | 4 | meters | -Query OK, 1 row(s) in set (0.001091s) -``` - -From the above experiment, we can see that while the value to be inserted is invalid the table is still created. diff --git a/docs/en/12-taos-sql/06-select.md b/docs/en/12-taos-sql/06-select.md index 8a017cf92e40aa4a854dcd531b7df291a9243515..439205696b3012c17865898d50635ee0fed580ab 100644 --- a/docs/en/12-taos-sql/06-select.md +++ b/docs/en/12-taos-sql/06-select.md @@ -1,118 +1,124 @@ --- +sidebar_label: Select title: Select --- ## Syntax -```SQL -SELECT select_expr [, select_expr ...] - FROM {tb_name_list} - [WHERE where_condition] - [SESSION(ts_col, tol_val)] - [STATE_WINDOW(col)] - [INTERVAL(interval_val [, interval_offset]) [SLIDING sliding_val]] - [FILL(fill_mod_and_val)] - [GROUP BY col_list] - [ORDER BY col_list { DESC | ASC }] +```sql +SELECT {DATABASE() | CLIENT_VERSION() | SERVER_VERSION() | SERVER_STATUS() | NOW() | TODAY() | TIMEZONE()} + +SELECT [DISTINCT] select_list + from_clause + [WHERE condition] + [PARTITION BY tag_list] + [window_clause] + [group_by_clause] + [order_by_clasue] [SLIMIT limit_val [SOFFSET offset_val]] [LIMIT limit_val [OFFSET offset_val]] - [>> export_file]; -``` + [>> export_file] -## Wildcard +select_list: + select_expr [, select_expr] ... -Wildcard \* can be used to specify all columns. The result includes only data columns for normal tables. +select_expr: { + * + | query_name.* + | [schema_name.] {table_name | view_name} .* + | t_alias.* + | expr [[AS] c_alias] +} -``` -taos> SELECT * FROM d1001; - ts | current | voltage | phase | -====================================================================================== - 2018-10-03 14:38:05.000 | 10.30000 | 219 | 0.31000 | - 2018-10-03 14:38:15.000 | 12.60000 | 218 | 0.33000 | - 2018-10-03 14:38:16.800 | 12.30000 | 221 | 0.31000 | -Query OK, 3 row(s) in set (0.001165s) -``` +from_clause: { + table_reference [, table_reference] ... + | join_clause [, join_clause] ... +} + +table_reference: + table_expr t_alias + +table_expr: { + table_name + | view_name + | ( subquery ) +} -The result includes both data columns and tag columns for super table. +join_clause: + table_reference [INNER] JOIN table_reference ON condition +window_clause: { + SESSION(ts_col, tol_val) + | STATE_WINDOW(col) + | INTERVAL(interval_val [, interval_offset]) [SLIDING (sliding_val)] [WATERMARK(watermark_val)] [FILL(fill_mod_and_val)] + +changes_option: { + DURATION duration_val + | ROWS rows_val +} + +group_by_clause: + GROUP BY expr [, expr] ... HAVING condition + +order_by_clasue: + ORDER BY order_expr [, order_expr] ... + +order_expr: + {expr | position | c_alias} [DESC | ASC] [NULLS FIRST | NULLS LAST] ``` -taos> SELECT * FROM meters; - ts | current | voltage | phase | location | groupid | -===================================================================================================================================== - 2018-10-03 14:38:05.500 | 11.80000 | 221 | 0.28000 | California.LoSangeles | 2 | - 2018-10-03 14:38:16.600 | 13.40000 | 223 | 0.29000 | California.LoSangeles | 2 | - 2018-10-03 14:38:05.000 | 10.80000 | 223 | 0.29000 | California.LoSangeles | 3 | - 2018-10-03 14:38:06.500 | 11.50000 | 221 | 0.35000 | California.LoSangeles | 3 | - 2018-10-03 14:38:04.000 | 10.20000 | 220 | 0.23000 | California.SanFrancisco | 3 | - 2018-10-03 14:38:16.650 | 10.30000 | 218 | 0.25000 | California.SanFrancisco | 3 | - 2018-10-03 14:38:05.000 | 10.30000 | 219 | 0.31000 | California.SanFrancisco | 2 | - 2018-10-03 14:38:15.000 | 12.60000 | 218 | 0.33000 | California.SanFrancisco | 2 | - 2018-10-03 14:38:16.800 | 12.30000 | 221 | 0.31000 | California.SanFrancisco | 2 | -Query OK, 9 row(s) in set (0.002022s) + +## Lists + +A query can be performed on some or all columns. Data and tag columns can all be included in the SELECT list. + +## Wildcards + +You can use an asterisk (\*) as a wildcard character to indicate all columns. For standard tables, the asterisk indicates only data columns. For supertables and subtables, tag columns are also included. + +```sql +SELECT * FROM d1001; ``` -Wildcard can be used with table name as prefix. Both SQL statements below have the same effect and return all columns. +You can use a table name as a prefix before an asterisk. For example, the following SQL statements both return all columns from the d1001 table: -```SQL +```sql SELECT * FROM d1001; SELECT d1001.* FROM d1001; ``` -In a JOIN query, however, the results are different with or without a table name prefix. \* without table prefix will return all the columns of both tables, but \* with table name as prefix will return only the columns of that table. +However, in a JOIN query, using a table name prefix with an asterisk returns different results. In this case, querying * returns all data in all columns in all tables (not including tags), whereas using a table name prefix returns all data in all columns in the specified table only. -``` -taos> SELECT * FROM d1001, d1003 WHERE d1001.ts=d1003.ts; - ts | current | voltage | phase | ts | current | voltage | phase | -================================================================================================================================== - 2018-10-03 14:38:05.000 | 10.30000| 219 | 0.31000 | 2018-10-03 14:38:05.000 | 10.80000| 223 | 0.29000 | -Query OK, 1 row(s) in set (0.017385s) +```sql +SELECT * FROM d1001, d1003 WHERE d1001.ts=d1003.ts; +SELECT d1001.* FROM d1001,d1003 WHERE d1001.ts = d1003.ts; ``` -``` -taos> SELECT d1001.* FROM d1001,d1003 WHERE d1001.ts = d1003.ts; - ts | current | voltage | phase | -====================================================================================== - 2018-10-03 14:38:05.000 | 10.30000 | 219 | 0.31000 | -Query OK, 1 row(s) in set (0.020443s) -``` +The first of the preceding SQL statements returns all columns from the d1001 and d1003 tables, but the second of the preceding SQL statements returns all columns from the d1001 table only. -Wildcard \* can be used with some functions, but the result may be different depending on the function being used. For example, `count(*)` returns only one column, i.e. the number of rows; `first`, `last` and `last_row` return all columns of the selected row. +With regard to the other SQL functions that support wildcards, the differences are as follows: +`count(*)` only returns one column. `first`, `last`, and `last_row` return all columns. -``` -taos> SELECT COUNT(*) FROM d1001; - count(*) | -======================== - 3 | -Query OK, 1 row(s) in set (0.001035s) -``` +### Tag Columns -``` -taos> SELECT FIRST(*) FROM d1001; - first(ts) | first(current) | first(voltage) | first(phase) | -========================================================================================= - 2018-10-03 14:38:05.000 | 10.30000 | 219 | 0.31000 | -Query OK, 1 row(s) in set (0.000849s) +You can query tag columns in supertables and subtables and receive results in the same way as querying data columns. + +```sql +SELECT location, groupid, current FROM d1001 LIMIT 2; ``` -## Tags +### Distinct Values -Starting from version 2.0.14, tag columns can be selected together with data columns when querying sub tables. Please note however, that, wildcard \* cannot be used to represent any tag column. This means that tag columns must be specified explicitly like the example below. +The DISTINCT keyword returns only values that are different over one or more columns. You can use the DISTINCT keyword with tag columns and data columns. -``` -taos> SELECT location, groupid, current FROM d1001 LIMIT 2; - location | groupid | current | -====================================================================== - California.SanFrancisco | 2 | 10.30000 | - California.SanFrancisco | 2 | 12.60000 | -Query OK, 2 row(s) in set (0.003112s) -``` +The following SQL statement returns distinct values from a tag column: -## Get distinct values +```sql +SELECT DISTINCT tag_name [, tag_name ...] FROM stb_name; +``` -`DISTINCT` keyword can be used to get all the unique values of tag columns from a super table. It can also be used to get all the unique values of data columns from a table or subtable. +The following SQL statement returns distinct values from a data column: ```sql -SELECT DISTINCT tag_name [, tag_name ...] FROM stb_name; SELECT DISTINCT col_name [, col_name ...] FROM tb_name; ``` @@ -124,231 +130,188 @@ SELECT DISTINCT col_name [, col_name ...] FROM tb_name; ::: -## Columns Names of Result Set +### Column Names -When using `SELECT`, the column names in the result set will be the same as that in the select clause if `AS` is not used. `AS` can be used to rename the column names in the result set. For example +When using `SELECT`, the column names in the result set will be the same as that in the select clause if `AS` is not used. `AS` can be used to rename the column names in the result set. For example: -``` +```sql taos> SELECT ts, ts AS primary_key_ts FROM d1001; - ts | primary_key_ts | -==================================================== - 2018-10-03 14:38:05.000 | 2018-10-03 14:38:05.000 | - 2018-10-03 14:38:15.000 | 2018-10-03 14:38:15.000 | - 2018-10-03 14:38:16.800 | 2018-10-03 14:38:16.800 | -Query OK, 3 row(s) in set (0.001191s) ``` `AS` can't be used together with `first(*)`, `last(*)`, or `last_row(*)`. -## Implicit Columns +### Pseudocolumns + +**TBNAME** +The TBNAME pseudocolumn in a supertable contains the names of subtables within the supertable. -`Select_exprs` can be column names of a table, or function expression or arithmetic expression on columns. The maximum number of allowed column names and expressions is 256. Timestamp and the corresponding tag names will be returned in the result set if `interval` or `group by tags` are used, and timestamp will always be the first column in the result set. +The following SQL statement returns all unique subtable names and locations within the meters supertable: -## Table List +```mysql +SELECT DISTINCT TBNAME, location FROM meters; +``` -`FROM` can be followed by a number of tables or super tables, or can be followed by a sub-query. If no database is specified as current database in use, table names must be preceded with database name, like `power.d1001`. +Use the `INS_TAGS` system table in `INFORMATION_SCHEMA` to query the information for subtables in a supertable. For example, the following statement returns the name and tag values for each subtable in the `meters` supertable. -```SQL -SELECT * FROM power.d1001; +```mysql +SELECT table_name, tag_name, tag_type, tag_value FROM information_schema.ins_tags WHERE stable_name='meters'; ``` -has same effect as +The following SQL statement returns the number of subtables within the meters supertable. -```SQL -USE power; -SELECT * FROM d1001; +```mysql +SELECT COUNT(*) FROM (SELECT DISTINCT TBNAME FROM meters); ``` -## Special Query +In the preceding two statements, only tags can be used as filtering conditions in the WHERE clause. For example: -Some special query functions can be invoked without `FROM` sub-clause. For example, the statement below can be used to get the current database in use. +**\_QSTART and \_QEND** -``` -taos> SELECT DATABASE(); - database() | -================================= - power | -Query OK, 1 row(s) in set (0.000079s) -``` +The \_QSTART and \_QEND pseudocolumns contain the beginning and end of the time range of a query. If the WHERE clause in a statement does not contain valid timestamps, the time range is equal to [-2^63, 2^63 - 1]. -If no database is specified upon logging in and no database is specified with `USE` after login, NULL will be returned by `select database()`. +The \_QSTART and \_QEND pseudocolumns cannot be used in a WHERE clause. -``` -taos> SELECT DATABASE(); - database() | -================================= - NULL | -Query OK, 1 row(s) in set (0.000184s) -``` +**\_WSTART, \_WEND, and \_DURATION** +\_WSTART, \_WEND, and \_WDURATION pseudocolumns +The \_WSTART, \_WEND, and \_WDURATION pseudocolumns indicate the beginning, end, and duration of a window. -The statement below can be used to get the version of client or server. +These pseudocolumns can be used only in time window-based aggregations and must occur after the aggregation clause. -``` -taos> SELECT CLIENT_VERSION(); - client_version() | -=================== - 2.0.0.0 | -Query OK, 1 row(s) in set (0.000070s) +**\_c0 and \_ROWTS** -taos> SELECT SERVER_VERSION(); - server_version() | -=================== - 2.0.0.0 | -Query OK, 1 row(s) in set (0.000077s) +In TDengine, the first column of all tables must be a timestamp. This column is the primary key of the table. The \_c0 and \_ROWTS pseudocolumns both represent the values of this column. These pseudocolumns enable greater flexibility and standardization. For example, you can use functions such as MAX and MIN with these pseudocolumns. + +```sql +select _rowts, max(current) from meters; ``` -The statement below is used to check the server status. An integer, like `1`, is returned if the server status is OK, otherwise an error code is returned. This is compatible with the status check for TDengine from connection pool or 3rd party tools, and can avoid the problem of losing the connection from a connection pool when using the wrong heartbeat checking SQL statement. +## Query Objects -``` -taos> SELECT SERVER_STATUS(); - server_status() | -================== - 1 | -Query OK, 1 row(s) in set (0.000074s) +`FROM` can be followed by a number of tables or super tables, or can be followed by a sub-query. +If no database is specified as current database in use, table names must be preceded with database name, for example, `power.d1001`. -taos> SELECT SERVER_STATUS() AS status; - status | -============== - 1 | -Query OK, 1 row(s) in set (0.000081s) -``` +You can perform INNER JOIN statements based on the primary key. The following conditions apply: -## \_block_dist +1. You can use FROM table list or an explicit JOIN clause. +2. For standard tables and subtables, you must specify an ON condition and the condition must be equivalent to the primary key. +3. For supertables, the ON condition must be equivalent to the primary key. In addition, the tag columns of the tables on which the INNER JOIN is performed must have a one-to-one relationship. You cannot specify an OR condition. +4. The tables that are included in a JOIN clause must be of the same type (supertable, standard table, or subtable). +5. You can include subqueries before and after the JOIN keyword. +6. You cannot include more than ten tables in a JOIN clause. +7. You cannot include a FILL clause and a JOIN clause in the same statement. -**Description**: Get the data block distribution of a table or STable. +## GROUP BY -```SQL title="Syntax" -SELECT _block_dist() FROM { tb_name | stb_name } -``` +If you use a GROUP BY clause, the SELECT list can only include the following items: -**Restrictions**：No argument is allowed, where clause is not allowed +1. Constants +2. Aggregate functions +3. Expressions that are consistent with the expression following the GROUP BY clause +4. Expressions that include the preceding expression -**Sub Query**：Sub query or nested query are not supported +The GROUP BY clause groups each row of data by the value of the expression following the clause and returns a combined result for each group. -**Return value**: A string which includes the data block distribution of the specified table or STable, i.e. the histogram of rows stored in the data blocks of the table or STable. +The expressions in a GROUP BY clause can include any column in any table or view. It is not necessary that the expressions appear in the SELECT list. -```text title="Result" -summary: -5th=[392], 10th=[392], 20th=[392], 30th=[392], 40th=[792], 50th=[792] 60th=[792], 70th=[792], 80th=[792], 90th=[792], 95th=[792], 99th=[792] Min=[392(Rows)] Max=[800(Rows)] Avg=[666(Rows)] Stddev=[2.17] Rows=[2000], Blocks=[3], Size=[5.440(Kb)] Comp=[0.23] RowsInMem=[0] SeekHeaderTime=[1(us)] -``` +The GROUP BY clause does not guarantee that the results are ordered. If you want to ensure that grouped data is ordered, use the ORDER BY clause. -**More explanation about above example**: -- Histogram about the rows stored in the data blocks of the table or STable: the value of rows for 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90%, 95%, and 99% -- Minimum number of rows stored in a data block, i.e. Min=[392(Rows)] -- Maximum number of rows stored in a data block, i.e. Max=[800(Rows)] -- Average number of rows stored in a data block, i.e. Avg=[666(Rows)] -- stddev of number of rows, i.e. Stddev=[2.17] -- Total number of rows, i.e. Rows[2000] -- Total number of data blocks, i.e. Blocks=[3] -- Total disk size consumed, i.e. Size=[5.440(Kb)] -- Compression ratio, which means the compressed size divided by original size, i.e. Comp=[0.23] -- Total number of rows in memory, i.e. RowsInMem=[0], which means no rows in memory -- The time spent on reading head file (to retrieve data block information), i.e. SeekHeaderTime=[1(us)], which means 1 microsecond. +## PARTITION BY -## Special Keywords in TAOS SQL +The PARTITION BY clause is a TDengine-specific extension to standard SQL. This clause partitions data based on the part_list and performs computations per partition. -- `TBNAME`： it is treated as a special tag when selecting on a super table, representing the name of subtables in that super table. -- `_c0`: represents the first column of a table or super table. +For more information, see TDengine Extensions. -## Tips +## ORDER BY -To get all the subtables and corresponding tag values from a super table: +The ORDER BY keyword orders query results. If you do not include an ORDER BY clause in a query, the order of the results can be inconsistent. -```SQL -SELECT TBNAME, location FROM meters; -``` +You can specify integers after ORDER BY to indicate the order in which you want the items in the SELECT list to be displayed. For example, 1 indicates the first item in the select list. + +You can specify ASC for ascending order or DESC for descending order. + +You can also use the NULLS keyword to specify the position of null values. Ascending order uses NULLS LAST by default. Descending order uses NULLS FIRST by default. + +## LIMIT + +The LIMIT keyword controls the number of results that are displayed. You can also use the OFFSET keyword to specify the result to display first. `LIMIT` and `OFFSET` are executed after `ORDER BY` in the query execution. You can include an offset in a LIMIT clause. For example, LIMIT 5 OFFSET 2 can also be written LIMIT 2, 5. Both of these clauses display the third through the seventh results. + +In a statement that includes a PARTITON BY clause, the LIMIT keyword is performed on each partition, not on the entire set of results. + +## SLIMIT + +The SLIMIT keyword is used with a PARTITION BY clause to control the number of partitions that are displayed. You can include an offset in a SLIMIT clause. For example, SLIMIT 5 OFFSET 2 can also be written LIMIT 2, 5. Both of these clauses display the third through the seventh partitions. + +Note: If you include an ORDER BY clause, only one partition can be displayed. + +## Special Query -To get the number of sub tables in a super table： +Some special query functions can be invoked without `FROM` sub-clause. -```SQL -SELECT COUNT(TBNAME) FROM meters; +## Obtain Current Database + +The following SQL statement returns the current database. If a database has not been specified on login or with the `USE` command, a null value is returned. + +```sql +SELECT DATABASE(); ``` -Only filter on `TAGS` are allowed in the `where` clause for above two query statements. For example: +### Obtain Current Version +```sql +SELECT CLIENT_VERSION(); +SELECT SERVER_VERSION(); ``` -taos> SELECT TBNAME, location FROM meters; - tbname | location | -================================================================== - d1004 | California.LosAngeles | - d1003 | California.LosAngeles | - d1002 | California.SanFrancisco | - d1001 | California.SanFrancisco | -Query OK, 4 row(s) in set (0.000881s) -taos> SELECT COUNT(tbname) FROM meters WHERE groupId > 2; - count(tbname) | -======================== - 2 | -Query OK, 1 row(s) in set (0.001091s) +## Obtain Server Status + +The following SQL statement returns the status of the TDengine server. An integer indicates that the server is running normally. An error code indicates that an error has occurred. This statement can also detect whether a connection pool or third-party tool is connected to TDengine properly. By using this statement, you can ensure that connections in a pool are not lost due to an incorrect heartbeat detection statement. + +```sql +SELECT SERVER_STATUS(); ``` -- Wildcard \* can be used to get all columns, or specific column names can be specified. Arithmetic operation can be performed on columns of numerical types, columns can be renamed in the result set. -- Arithmetic operation on columns can't be used in where clause. For example, `where a*2>6;` is not allowed but `where a>6/2;` can be used instead for the same purpose. -- Arithmetic operation on columns can't be used as the objectives of select statement. For example, `select min(2*a) from t;` is not allowed but `select 2*min(a) from t;` can be used instead. -- Logical operation can be used in `WHERE` clause to filter numeric values, wildcard can be used to filter string values. -- Result sets are arranged in ascending order of the first column, i.e. timestamp, but it can be controlled to output as descending order of timestamp. If `order by` is used on other columns, the result may not be as expected. By the way, \_c0 is used to represent the first column, i.e. timestamp. -- `LIMIT` parameter is used to control the number of rows to output. `OFFSET` parameter is used to specify from which row to output. `LIMIT` and `OFFSET` are executed after `ORDER BY` in the query execution. A simple tip is that `LIMIT 5 OFFSET 2` can be abbreviated as `LIMIT 2, 5`. -- What is controlled by `LIMIT` is the number of rows in each group when `GROUP BY` is used. -- `SLIMIT` parameter is used to control the number of groups when `GROUP BY` is used. Similar to `LIMIT`, `SLIMIT 5 OFFSET 2` can be abbreviated as `SLIMIT 2, 5`. -- ">>" can be used to output the result set of `select` statement to the specified file. +### Obtain Current Time -## Where +```sql +SELECT NOW(); +``` -Logical operations in below table can be used in the `where` clause to filter the resulting rows. +### Obtain Current Date -| **Operation** | **Note** | **Applicable Data Types** | -| ------------- | ------------------------ | ----------------------------------------- | -| > | larger than | all types except bool | -| < | smaller than | all types except bool | -| >= | larger than or equal to | all types except bool | -| <= | smaller than or equal to | all types except bool | -| = | equal to | all types | -| <\> | not equal to | all types | -| is [not] null | is null or is not null | all types | -| between and | within a certain range | all types except bool | -| in | match any value in a set | all types except first column `timestamp` | -| like | match a wildcard string | **`binary`** **`nchar`** | -| match/nmatch | filter regex | **`binary`** **`nchar`** | +```sql +SELECT TODAY(); +``` -**Explanations**: +### Obtain Current Time Zone -- Operator `<\>` is equal to `!=`, please note that this operator can't be used on the first column of any table, i.e.timestamp column. -- Operator `like` is used together with wildcards to match strings - - '%' matches 0 or any number of characters, '\_' matches any single ASCII character. - - `\_` is used to match the \_ in the string. - - The maximum length of wildcard string is 100 bytes from version 2.1.6.1 (before that the maximum length is 20 bytes). `maxWildCardsLength` in `taos.cfg` can be used to control this threshold. A very long wildcard string may slowdown the execution performance of `LIKE` operator. -- `AND` keyword can be used to filter multiple columns simultaneously. AND/OR operation can be performed on single or multiple columns from version 2.3.0.0. However, before 2.3.0.0 `OR` can't be used on multiple columns. -- For timestamp column, only one condition can be used; for other columns or tags, `OR` keyword can be used to combine multiple logical operators. For example, `((value > 20 AND value < 30) OR (value < 12))`. - - From version 2.3.0.0, multiple conditions can be used on timestamp column, but the result set can only contain single time range. -- From version 2.0.17.0, operator `BETWEEN AND` can be used in where clause, for example `WHERE col2 BETWEEN 1.5 AND 3.25` means the filter condition is equal to "1.5 ≤ col2 ≤ 3.25". -- From version 2.1.4.0, operator `IN` can be used in the where clause. For example, `WHERE city IN ('California.SanFrancisco', 'California.SanDiego')`. For bool type, both `{true, false}` and `{0, 1}` are allowed, but integers other than 0 or 1 are not allowed. FLOAT and DOUBLE types are impacted by floating point precision errors. Only values that match the condition within the tolerance will be selected. Non-primary key column of timestamp type can be used with `IN`. -- From version 2.3.0.0, regular expression is supported in the where clause with keyword `match` or `nmatch`. The regular expression is case insensitive. +```sql +SELECT TIMEZONE(); +``` ## Regular Expression ### Syntax -```SQL +```txt WHERE (column|tbname) **match/MATCH/nmatch/NMATCH** _regex_ ``` ### Specification -The regular expression being used must be compliant with POSIX specification, please refer to [Regular Expressions](https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html). +TDengine supports POSIX regular expression syntax. For more information, see [Regular Expressions](https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap09.html). ### Restrictions -Regular expression can be used against only table names, i.e. `tbname`, and tags of binary/nchar types, but can't be used against data columns. +Regular expression filtering is supported only on table names (TBNAME), BINARY tags, and NCHAR tags. Regular expression filtering cannot be performed on data columns. -The maximum length of regular expression string is 128 bytes. Configuration parameter `maxRegexStringLen` can be used to set the maximum allowed regular expression. It's a configuration parameter on the client side, and will take effect after restarting the client. +A regular expression string cannot exceed 128 bytes. You can configure this value by modifying the maxRegexStringLen parameter on the TDengine Client. The modified value takes effect when the client is restarted. ## JOIN -From version 2.2.0.0, inner join is fully supported in TDengine. More specifically, the inner join between table and table, between STable and STable, and between sub query and sub query are supported. +TDengine supports natural joins between supertables, between standard tables, and between subqueries. The difference between natural joins and inner joins is that natural joins require that the fields being joined in the supertables or standard tables must have the same name. Data or tag columns must be joined with the equivalent column in another table. -Only primary key, i.e. timestamp, can be used in the join operation between table and table. For example: +For standard tables, only the timestamp (primary key) can be used in join operations. For example: ```sql SELECT * @@ -356,25 +319,26 @@ FROM temp_tb_1 t1, pressure_tb_1 t2 WHERE t1.ts = t2.ts ``` -In the join operation between STable and STable, besides the primary key, i.e. timestamp, tags can also be used. For example: +For supertables, tags as well as timestamps can be used in join operations. For example: ```sql SELECT * -FROM temp_STable t1, temp_STable t2 +FROM temp_stable t1, temp_stable t2 WHERE t1.ts = t2.ts AND t1.deviceid = t2.deviceid AND t1.status=0; ``` -Similarly, join operations can be performed on the result set of multiple sub queries. +Similarly, join operations can be performed on the result sets of multiple subqueries. :::note -Restrictions on join operation: -- The number of tables or STables in a single join operation can't exceed 10. -- `FILL` is not allowed in the query statement that includes JOIN operation. -- Arithmetic operation is not allowed on the result set of join operation. -- `GROUP BY` is not allowed on a part of tables that participate in join operation. -- `OR` can't be used in the conditions for join operation -- join operation can't be performed on data columns, i.e. can only be performed on tags or primary key, i.e. timestamp +The following restriction apply to JOIN statements: + +- The number of tables or supertables in a single join operation cannot exceed 10. +- `FILL` cannot be used in a JOIN statement. +- Arithmetic operations cannot be performed on the result sets of join operation. +- `GROUP BY` is not allowed on a segment of the tables that participate in a join operation. +- `OR` cannot be used in the conditions for join operation +- Join operation can be performed only on tags or timestamps. You cannot perform a join operation on data columns. ::: @@ -384,7 +348,7 @@ Nested query is also called sub query. This means that in a single SQL statement From 2.2.0.0, unassociated sub query can be used in the `FROM` clause. Unassociated means the sub query doesn't use the parameters in the parent query. More specifically, in the `tb_name_list` of `SELECT` statement, an independent SELECT statement can be used. So a complete nested query looks like: -```SQL +``` SELECT ... FROM (SELECT ... FROM ...) ...; ``` @@ -408,42 +372,42 @@ SELECT ... FROM (SELECT ... FROM ...) ...; ## UNION ALL -```SQL title=Syntax +```txt title=Syntax SELECT ... UNION ALL SELECT ... [UNION ALL SELECT ...] ``` -`UNION ALL` operator can be used to combine the result set from multiple select statements as long as the result set of these select statements have exactly the same columns. `UNION ALL` doesn't remove redundant rows from multiple result sets. In a single SQL statement, at most 100 `UNION ALL` can be supported. +TDengine supports the `UNION ALL` operation. `UNION ALL` operator can be used to combine the result set from multiple select statements as long as the result set of these select statements have exactly the same columns. `UNION ALL` doesn't remove redundant rows from multiple result sets. In a single SQL statement, at most 100 `UNION ALL` can be supported. ### Examples table `tb1` is created using below SQL statement: -```SQL +``` CREATE TABLE tb1 (ts TIMESTAMP, col1 INT, col2 FLOAT, col3 BINARY(50)); ``` The rows in the past one hour in `tb1` can be selected using below SQL statement: -```SQL +``` SELECT * FROM tb1 WHERE ts >= NOW - 1h; ``` The rows between 2018-06-01 08:00:00.000 and 2018-06-02 08:00:00.000 and col3 ends with 'nny' can be selected in the descending order of timestamp using below SQL statement: -```SQL +``` SELECT * FROM tb1 WHERE ts > '2018-06-01 08:00:00.000' AND ts <= '2018-06-02 08:00:00.000' AND col3 LIKE '%nny' ORDER BY ts DESC; ``` The sum of col1 and col2 for rows later than 2018-06-01 08:00:00.000 and whose col2 is bigger than 1.2 can be selected and renamed as "complex", while only 10 rows are output from the 5th row, by below SQL statement: -```SQL +``` SELECT (col1 + col2) AS 'complex' FROM tb1 WHERE ts > '2018-06-01 08:00:00.000' AND col2 > 1.2 LIMIT 10 OFFSET 5; ``` The rows in the past 10 minutes and whose col2 is bigger than 3.14 are selected and output to the result file `/home/testoutput.csv` with below SQL statement: -```SQL +``` SELECT COUNT(*) FROM tb1 WHERE ts >= NOW - 10m AND col2 > 3.14 >> /home/testoutput.csv; ``` diff --git a/docs/en/12-taos-sql/08-delete-data.mdx b/docs/en/12-taos-sql/08-delete-data.mdx index 86443dca53b08f5f5c489d40f4a2ea8afc8081fb..999c467ad05b6d3e349e322141acfb02a49de5ff 100644 --- a/docs/en/12-taos-sql/08-delete-data.mdx +++ b/docs/en/12-taos-sql/08-delete-data.mdx @@ -4,8 +4,7 @@ description: "Delete data from table or Stable" title: Delete Data --- -TDengine provides the functionality of deleting data from a table or STable according to specified time range, it can be used to cleanup abnormal data generated due to device failure. Please be noted that this functionality is only available in Enterprise version, please refer to [TDengine Enterprise Edition](https://tdengine.com/products#enterprise-edition-link) - +TDengine provides the functionality of deleting data from a table or STable according to specified time range, it can be used to cleanup abnormal data generated due to device failure. **Syntax:** @@ -16,21 +15,21 @@ DELETE FROM [ db_name. ] tb_name [WHERE condition]; **Description:** Delete data from a table or STable **Parameters:** - + - `db_name`: Optional parameter, specifies the database in which the table exists; if not specified, the current database will be used. - `tb_name`: Mandatory parameter, specifies the table name from which data will be deleted, it can be normal table, subtable or STable. -- `condition`: Optional parameter, specifies the data filter condition. If no condition is specified all data will be deleted, so please be cautions to delete data without any condition. The condition used here is only applicable to the first column, i.e. the timestamp column. If the table is a STable, the condition is also applicable to tag columns. +- `condition`: Optional parameter, specifies the data filter condition. If no condition is specified all data will be deleted, so please be cautions to delete data without any condition. The condition used here is only applicable to the first column, i.e. the timestamp column. **More Explanations:** - -The data can't be recovered once deleted, so please be cautious to use the functionality of deleting data. It's better to firstly make sure the data to be deleted using `select` then execute `delete`. + +The data can't be recovered once deleted, so please be cautious to use the functionality of deleting data. It's better to firstly make sure the data to be deleted using `select` then execute `delete`. **Example:** - -`meters` is a STable, in which `groupid` is a tag column of int type. Now we want to delete the data older than 2021-10-01 10:40:00.100 and `groupid` is 1. The SQL for this purpose is like below: + +`meters` is a STable, in which `groupid` is a tag column of int type. Now we want to delete the data older than 2021-10-01 10:40:00.100. You can perform this action by running the following SQL statement: ```sql -delete from meters where ts < '2021-10-01 10:40:00.100' and groupid=1 ; +delete from meters where ts < '2021-10-01 10:40:00.100' ; ``` The output is: diff --git a/docs/en/12-taos-sql/10-function.md b/docs/en/12-taos-sql/10-function.md index 6375422b07a2ee7d5c9b6a0074060a39888da773..36043e745664f6871fae7a4bb4a58d0f3fb010a4 100644 --- a/docs/en/12-taos-sql/10-function.md +++ b/docs/en/12-taos-sql/10-function.md @@ -1,13 +1,14 @@ --- +sidebar_label: Functions title: Functions toc_max_heading_level: 4 --- -## Single-Row Functions +## Single Row Functions -Single-Row functions return a result row for each row in the query result. +Single row functions return a result for each row. -### Numeric Functions +### Mathematical Functions #### ABS @@ -17,16 +18,15 @@ SELECT ABS(field_name) FROM { tb_name | stb_name } [WHERE clause] **Description**: The absolute value of a specific field. -**Return value type**: Same as input type. +**Return value type**: Same as the field being used -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Cannot be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### ACOS @@ -34,18 +34,17 @@ SELECT ABS(field_name) FROM { tb_name | stb_name } [WHERE clause] SELECT ACOS(field_name) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The anti-cosine of a specific field. +**Description**: The arc cosine of a specific field. -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Cannot be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### ASIN @@ -53,18 +52,18 @@ SELECT ACOS(field_name) FROM { tb_name | stb_name } [WHERE clause] SELECT ASIN(field_name) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The anti-sine of a specific field. +**Description**: The arc sine of a specific field. -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables + +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. -**More explanations**: -- Cannot be used with aggregate functions. #### ATAN @@ -72,37 +71,36 @@ SELECT ASIN(field_name) FROM { tb_name | stb_name } [WHERE clause] SELECT ATAN(field_name) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: anti-tangent of a specific field. +**Description**: The arc tangent of a specific field. -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables + +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. -**More explanations**: -- Cannot be used with aggregate functions. #### CEIL -``` +```sql SELECT CEIL(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The rounded up value of a specific field. +**Description**: The rounded up value of a specific field -**Return value type**: Same as input type. +**Return value type**: Same as the field being used -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable +**Applicable table types**: standard tables and supertables -**Applicable nested query**: Inner query and outer query. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**More explanations**: -- Can't be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### COS @@ -112,45 +110,43 @@ SELECT COS(field_name) FROM { tb_name | stb_name } [WHERE clause] **Description**: The cosine of a specific field. -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Can't be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### FLOOR -``` +```sql SELECT FLOOR(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The rounded down value of a specific field. - -**More explanations**: Refer to `CEIL` function for usage restrictions. +**Description**: The rounded down value of a specific field + **More explanations**: The restrictions are same as those of the `CEIL` function. #### LOG ```sql -SELECT LOG(field_name, base) FROM { tb_name | stb_name } [WHERE clause] +SELECT LOG(field_name[, base]) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The logarithm of a specific field with `base` as the radix. If `base` parameter is ignored, natural logarithm of the field is returned. +**Description**: The logarithm of a specific field with `base` as the radix. If you do not enter a base, the natural logarithm of the field is returned. + +**Return value type**: Double -**Return value type**: DOUBLE. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**Applicable nested query**: Inner query and Outer query. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. -**More explanations**: -- Can't be used with aggregate functions #### POW @@ -158,28 +154,28 @@ SELECT LOG(field_name, base) FROM { tb_name | stb_name } [WHERE clause] SELECT POW(field_name, power) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The power of a specific field with `power` as the index. +**Description**: The power of a specific field with `power` as the exponent. + +**Return value type**: Double -**Return value type**: DOUBLE. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**Applicable nested query**: Inner query and Outer query. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. -**More explanations**: -- Can't be used with aggregate functions. #### ROUND -``` +```sql SELECT ROUND(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The rounded value of a specific field. +**Description**: The rounded value of a specific field. + **More explanations**: The restrictions are same as those of the `CEIL` function. -**More explanations**: Refer to `CEIL` function for usage restrictions. #### SIN @@ -189,18 +185,15 @@ SELECT SIN(field_name) FROM { tb_name | stb_name } [WHERE clause] **Description**: The sine of a specific field. -**Description**: The anti-cosine of a specific field. +**Return value type**: Double -**Return value type**: DOUBLE. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. - -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Can't be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### SQRT @@ -210,16 +203,15 @@ SELECT SQRT(field_name) FROM { tb_name | stb_name } [WHERE clause] **Description**: The square root of a specific field. -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Can't be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. #### TAN @@ -229,39 +221,35 @@ SELECT TAN(field_name) FROM { tb_name | stb_name } [WHERE clause] **Description**: The tangent of a specific field. -**Description**: The anti-cosine of a specific field. - -**Return value type**: DOUBLE. +**Return value type**: Double -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: -- Can't be used with aggregate functions. +**Usage**: This function can only be used on data columns. It can be used with selection and projection functions but not with aggregation functions. -### String Functions +### Concatenation Functions -String functiosn take strings as input and output numbers or strings. +Concatenation functions take strings as input and produce string or numeric values as output. #### CHAR_LENGTH -``` +```sql SELECT CHAR_LENGTH(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The mumber of characters of a string. - -**Return value type**: INTEGER. +**Description**: The length in number of characters of a string -**Applicable data types**: VARCHAR, NCHAR. +**Return value type**: Bigint -**Applicable table types**: table, STable. +**Applicable data types**: VARCHAR and NCHAR -**Applicable nested query**: Inner query and Outer query. +**Nested query**: It can be used in both the outer query and inner query in a nested query. +**Applicable table types**: standard tables and supertables #### CONCAT @@ -269,144 +257,139 @@ SELECT CHAR_LENGTH(str|column) FROM { tb_name | stb_name } [WHERE clause] SELECT CONCAT(str1|column1, str2|column2, ...) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The concatenation result of two or more strings. +**Description**: The concatenation result of two or more strings -**Return value type**: If all input strings are VARCHAR type, the result is VARCHAR type too. If any one of input strings is NCHAR type, then the result is NCHAR. If input strings contain NULL value, the result is NULL. +**Return value type**: If the concatenated strings are VARCHARs, the result is a VARCHAR. If the concatenated strings are NCHARs, the result is an NCHAR. If an input value is null, the result is null. -**Applicable data types**: VARCHAR, NCHAR. At least 2 input strings are required, and at most 8 input strings are allowed. +**Applicable data types**: VARCHAR and NCHAR You can concatenate between 2 and 8 strings. -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### CONCAT_WS -``` +```sql SELECT CONCAT_WS(separator, str1|column1, str2|column2, ...) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The concatenation result of two or more strings with separator. +**Description**: The concatenation result of two or more strings with separator -**Return value type**: If all input strings are VARCHAR type, the result is VARCHAR type too. If any one of input strings is NCHAR type, then the result is NCHAR. If input strings contain NULL value, the result is NULL. +**Return value type**: If the concatenated strings are VARCHARs, the result is a VARCHAR. If the concatenated strings are NCHARs, the result is an NCHAR. If an input value is null, the result is null. -**Applicable data types**: VARCHAR, NCHAR. At least 3 input strings are required, and at most 9 input strings are allowed. +**Applicable data types**: VARCHAR and NCHAR You can concatenate between 3 and 9 strings. -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### LENGTH -``` +```sql SELECT LENGTH(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The length in bytes of a string. +**Description**: The length in bytes of a string -**Return value type**: INTEGER. +**Return value type**: Bigint -**Applicable data types**: VARCHAR, NCHAR. +**Applicable data types**: VARCHAR and NCHAR fields or columns -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### LOWER -``` +```sql SELECT LOWER(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: Convert the input string to lower case. +**Description**: Convert the input string to lower case -**Return value type**: Same as input type. +**Return value type**: Same as input -**Applicable data types**: VARCHAR, NCHAR. +**Applicable data types**: VARCHAR and NCHAR -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### LTRIM -``` +```sql SELECT LTRIM(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: Remove the left leading blanks of a string. +**Description**: Remove the left leading blanks of a string -**Return value type**: Same as input type. +**Return value type**: Same as input -**Applicable data types**: VARCHAR, NCHAR. +**Applicable data types**: VARCHAR and NCHAR -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### RTRIM -``` -SELECT RTRIM(str|column) FROM { tb_name | stb_name } [WHERE clause] +```sql +SELECT LTRIM(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: Remove the right tailing blanks of a string. +**Description**: Remove the right tailing blanks of a string -**Return value type**: Same as input type. +**Return value type**: Same as input -**Applicable data types**: VARCHAR, NCHAR. +**Applicable data types**: VARCHAR and NCHAR -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables #### SUBSTR -``` +```sql SELECT SUBSTR(str,pos[,len]) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: The sub-string starting from `pos` with length of `len` from the original string `str`. - -**Return value type**: Same as input type. +**Description**: The sub-string starting from `pos` with length of `len` from the original string `str` - If `len` is not specified, it means from `pos` to the end. -**Applicable data types**: VARCHAR, NCHAR. +**Return value type**: Same as input -**Applicable table types**: table, STable. +**Applicable data types**: VARCHAR and NCHAR Parameter `pos` can be an positive or negative integer; If it's positive, the starting position will be counted from the beginning of the string; if it's negative, the starting position will be counted from the end of the string. -**Applicable nested query**: Inner query and Outer query. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**More explanations**: +**Applicable table types**: table, STable -- If the input is NULL, the output is NULL -- Parameter `pos` can be an positive or negative integer; If it's positive, the starting position will be counted from the beginning of the string; if it's negative, the starting position will be counted from the end of the string. -- If `len` is not specified, it means from `pos` to the end of string. #### UPPER -``` +```sql SELECT UPPER(str|column) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: Convert the input string to upper case. +**Description**: Convert the input string to upper case -**Return value type**: Same as input type. +**Return value type**: Same as input -**Applicable data types**: VARCHAR, NCHAR. +**Applicable data types**: VARCHAR and NCHAR -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: table, STable ### Conversion Functions -Conversion functions convert from one data type to another. +Conversion functions change the data type of a value. #### CAST @@ -414,19 +397,23 @@ Conversion functions convert from one data type to another. SELECT CAST(expression AS type_name) FROM { tb_name | stb_name } [WHERE clause] ``` -**Description**: Used for type casting. Convert `expression` to the type specified by `type_name`. +**Description**: Convert the input data `expression` into the type specified by `type_name`. This function can be used only in SELECT statements. -**Return value type**: The type specified by parameter `type_name`. +**Return value type**: The type specified by parameter `type_name` + +**Applicable data types**: All data types except JSON + +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable data types**: `expression` can be any data type except for JSON. +**Applicable table types**: standard tables and supertables **More explanations**: -- Error will be reported for unsupported type casting. +- Error will be reported for unsupported type casting - Some values of some supported data types may not be casted, below are known issues: - 1）When casting VARCHAR/NCHAR to BIGINT/BIGINT UNSIGNED, some characters may be treated as illegal, for example "a" may be converted to 0. - 2）When casting to numeric type, if converted result is out of range the destination data type can hold, overflow may occur and casting behavior is undefined. - 3) When casting to VARCHAR/NCHAR type, if converted string length exceeds the length specified in `type_name`, the result will be truncated. (e.g. CAST("abcd" as BINARY(2)) will return string "ab"). + 1. Some strings cannot be converted to numeric values. For example, the string `a` may be converted to `0`. However, this does not produce an error. + 2. If a converted numeric value is larger than the maximum size for the specified type, an overflow will occur. However, this does not produce an error. + 3. If a converted string value is larger than the maximum size for the specified type, the output value will be truncated. However, this does not produce an error. #### TO_ISO8601 @@ -434,18 +421,22 @@ SELECT CAST(expression AS type_name) FROM { tb_name | stb_name } [WHERE clause] SELECT TO_ISO8601(ts[, timezone]) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The ISO8601 date/time format converted from a UNIX timestamp, with timezone attached. `timezone` parameter allows attaching any customized timezone string to the output format. If `timezone` parameter is not specified, the timezone information of client side system will be attached. +**Description**: The ISO8601 date/time format converted from a UNIX timestamp, plus the timezone. You can specify any time zone with the timezone parameter. If you do not enter this parameter, the time zone on the client is used. -**Return value type**: VARCHAR. +**Return value type**: VARCHAR -**Applicable data types**: INTEGER, TIMESTAMP. +**Applicable data types**: Integers and timestamps -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. + +**Applicable table types**: standard tables and supertables **More explanations**: -- If the input is INTEGER represents UNIX timestamp, the precision of the returned value is determined by the digits of the input integer. -- If the input is of TIMESTAMP type, The precision of the returned value is same as the precision set for the current database in use. +- You can specify a time zone in the following format: [z/Z, +/-hhmm, +/-hh, +/-hh:mm]。 For example, TO_ISO8601(1, "+00:00"). +- If the input is a UNIX timestamp, the precision of the returned value is determined by the digits of the input timestamp +- If the input is a column of TIMESTAMP type, the precision of the returned value is same as the precision set for the current data base in use + #### TO_JSON @@ -453,38 +444,44 @@ SELECT TO_ISO8601(ts[, timezone]) FROM { tb_name | stb_name } [WHERE clause]; SELECT TO_JSON(str_literal) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: Convert a JSON string to a JSON body. +**Description**: Converts a string into JSON. -**Return value type**: JSON. +**Return value type**: JSON -**Applicable data types**: JSON string, in the format like '{ "literal" : literal }'. '{}' is NULL value. keys in the string must be string constants, values can be constants of numeric types, bool, string or NULL. Escaping characters are not allowed in the JSON string. +**Applicable data types**: JSON strings in the form `{"literal": literal}`. `{}` indicates a null value. The key must be a string literal. The value can be a numeric literal, string literal, Boolean literal, or null literal. str_literal cannot include escape characters. -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. + +**Applicable table types**: table, STable -**Applicable nested query**: Inner query and Outer query. #### TO_UNIXTIMESTAMP ```sql -SELECT TO_UNIXTIMESTAMP(datetime_string | ts_col) FROM { tb_name | stb_name } [WHERE clause]; +SELECT TO_UNIXTIMESTAMP(datetime_string) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: UNIX timestamp converted from a string of date/time format. +**Description**: UNIX timestamp converted from a string of date/time format -**Return value type**: BIGINT. +**Return value type**: BIGINT -**Applicable data types**: VARCHAR, NCHAR. +**Applicable column types**: VARCHAR and NCHAR -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. + +**Applicable table types**: standard tables and supertables **More explanations**: -- The input string must be compatible with ISO8601/RFC3339 standard, NULL will be returned if the string cannot be converted. -- The precision of the returned timestamp is same as the precision set for the current database in use. +- The input string must be compatible with ISO8601/RFC3339 standard, NULL will be returned if the string can't be converted +- The precision of the returned timestamp is same as the precision set for the current data base in use + + +### Time and Date Functions -### DateTime Functions +These functions perform operations on times and dates. -DateTime functions applied to timestamp data. NOW(), TODAY() and TIMEZONE() are executed only once even though they may occur multiple times in a single SQL statement. +All functions that return the current time, such as `NOW`, `TODAY`, and `TIMEZONE`, are calculated only once per statement even if they appear multiple times. #### NOW @@ -494,61 +491,66 @@ SELECT select_expr FROM { tb_name | stb_name } WHERE ts_col cond_operatior NOW() INSERT INTO tb_name VALUES (NOW(), ...); ``` -**Description**: The current time of the client side system. +**Description**: The current time of the client side system -**Return value type**: TIMESTAMP. +**Return value type**: TIMESTAMP -**Applicable data types**: TIMESTAMP only if used in WHERE/INSERT clause. +**Applicable column types**: TIMESTAMP only -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables + +**Nested query**: It can be used in both the outer query and inner query in a nested query. **More explanations**: -- Addition and Subtraction operation with time duration can be performed, for example NOW() + 1s, the time unit can be one of the followings: - b(nanosecond), u(microsecond), a(millisecond)), s(second), m(minute), h(hour), d(day), w(week). -- The precision of the returned timestamp is same as the precision set for the current database in use. +- Add and Subtract operation can be performed, for example NOW() + 1s, the time unit can be: + b(nanosecond), u(microsecond), a(millisecond)), s(second), m(minute), h(hour), d(day), w(week) +- The precision of the returned timestamp is same as the precision set for the current data base in use + #### TIMEDIFF ```sql -SELECT TIMEDIFF(ts1 | datetime_string1, ts2 | datetime_string2 [, time_unit]) FROM { tb_name | stb_name } [WHERE clause]; +SELECT TIMEDIFF(ts | datetime_string1, ts | datetime_string2 [, time_unit]) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The difference(duration) between two timestamps, and rounded to the time unit specified by `time_unit`. +**Description**: The difference between two timestamps, and rounded to the time unit specified by `time_unit` -**Return value type**: BIGINT. +**Return value type**: BIGINT -**Applicable data types**: INTEGER/TIMESTAMP represents UNIX timestamp, or VARCHAR/NCHAR string in date/time format. +**Applicable column types**: UNIX-style timestamps in BIGINT and TIMESTAMP format and other timestamps in VARCHAR and NCHAR format -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**Nested query**: It can be used in both the outer query and inner query in a nested query. +**More explanations**: - Time unit specified by `time_unit` can be: - 1b(nanosecond), 1u(microsecond),1a(millisecond),1s(second),1m(minute),1h(hour),1d(day),1w(week). -- If `time_unit` parameter is not specified, the precision of the returned time duration is same as the precision set for the current database in use. -- If input date-time string cannot be converted to UNIX timestamp, NULL value is returned. + 1b (nanoseconds), 1u (microseconds), 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), or 1w (weeks) +- The precision of the returned timestamp is same as the precision set for the current data base in use +- If the input data is not formatted as a timestamp, the returned value is null. + #### TIMETRUNCATE ```sql -SELECT TIMETRUNCATE(ts_val | datetime_string | ts_col, time_unit) FROM { tb_name | stb_name } [WHERE clause]; +SELECT TIMETRUNCATE(ts | datetime_string , time_unit) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: Truncate the input timestamp with unit specified by `time_unit`. +**Description**: Truncate the input timestamp with unit specified by `time_unit` -**Return value type**: TIMESTAMP. +**Return value type**: TIMESTAMP -**Applicable data types**: INTEGER/TIMESTAMP represents UNIX timestamp, or VARCHAR/NCHAR string in date/time format. +**Applicable column types**: UNIX-style timestamps in BIGINT and TIMESTAMP format and other timestamps in VARCHAR and NCHAR format -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables **More explanations**: - - Time unit specified by `time_unit` can be: - 1b(nanosecond),1u(microsecond),1a(millisecond),1s(second),1m(minute),1h(hour),1d(day),1w(week). -- The precision of the returned timestamp is same as the precision set for the current database in use. -- If input date-time string cannot be converted to UNIX timestamp, NULL value is returned. + 1b (nanoseconds), 1u (microseconds), 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), or 1w (weeks) +- The precision of the returned timestamp is same as the precision set for the current data base in use +- If the input data is not formatted as a timestamp, the returned value is null. + #### TIMEZONE @@ -556,13 +558,14 @@ SELECT TIMETRUNCATE(ts_val | datetime_string | ts_col, time_unit) FROM { tb_name SELECT TIMEZONE() FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The timezone of the client side system. +**Description**: The timezone of the client side system -**Return value type**: VARCHAR. +**Applicable data types**: VARCHAR -**Applicable data types**: None. +**Applicable column types**: None + +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. #### TODAY @@ -572,269 +575,269 @@ SELECT select_expr FROM { tb_name | stb_name } WHERE ts_col cond_operatior TODAY INSERT INTO tb_name VALUES (TODAY(), ...); ``` -**Description**: The timestamp of 00:00:00 of the client side system. +**Description**: The timestamp of 00:00:00 of the client side system -**Return value type**: TIMESTAMP. +**Return value type**: TIMESTAMP -**Applicable data types**: TIMESTAMP only if used in WHERE/INSERT clause. +**Applicable column types**: TIMESTAMP only -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables **More explanations**: -- Addition and Subtraction operation can be performed with time durations, for example NOW() + 1s, the time unit can be: - b(nanosecond), u(microsecond), a(millisecond)), s(second), m(minute), h(hour), d(day), w(week). -- The precision of the returned timestamp is same as the precision set for the current database in use. +- Add and Subtract operation can be performed, for example TODAY() + 1s, the time unit can be: + b(nanosecond), u(microsecond), a(millisecond)), s(second), m(minute), h(hour), d(day), w(week) +- The precision of the returned timestamp is same as the precision set for the current data base in use + ## Aggregate Functions -Aggregate functions return single result row for each group in the query result set. Groups are determined by `GROUP BY` clause or time window clause if they are used; or the whole result is considered a group if neither of them is used. +Aggregate functions return one row per group. You can use windows or GROUP BY to group data. Otherwise, the entire query is considered a single group. + +TDengine supports the following aggregate functions: ### APERCENTILE +```sql +SELECT APERCENTILE(field_name, P[, algo_type]) FROM { tb_name | stb_name } [WHERE clause] ``` -SELECT APERCENTILE(field_name, P[, algo_type]) -FROM { tb_name | stb_name } [WHERE clause] -``` - -**Description**: Similar to `PERCENTILE`, but a approximated result is returned. -**Return value type**: DOUBLE. +**Description**: Similar to `PERCENTILE`, but a simulated result is returned -**Applicable data types**: Numeric types. +**Return value type**: DOUBLE -**Applicable table types**: table, STable. +**Applicable data types**: Numeric -**More explanations** +**Applicable table types**: standard tables and supertables +**Explanations**： - _P_ is in range [0,100], when _P_ is 0, the result is same as using function MIN; when _P_ is 100, the result is same as function MAX. -- **algo_type** can only be input as `default` or `t-digest`, if it's not specified `default` will be used, i.e. `apercentile(column_name, 50)` is same as `apercentile(column_name, 50, "default")`. -- If `default` is used, histogram based algorithm is used for calculation. If `t-digest` is used, `t-digest` sampling algorithm is used to calculate the result. - -**Nested query**: It can be used in both the outer query and inner query in a nested query. +- `algo_type` can only be input as `default` or `t-digest` Enter `default` to use a histogram-based algorithm. Enter `t-digest` to use the t-digest algorithm to calculate the approximation of the quantile. `default` is used by default. ### AVG -``` +```sql SELECT AVG(field_name) FROM tb_name [WHERE clause]; ``` -**Description**: Get the average value of a column in a table or STable. +**Description**: The average value of the specified fields. + +**Return value type**: DOUBLE -**Return value type**: DOUBLE. +**Applicable data types**: Numeric -**Applicable data types**: Numeric type. +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. ### COUNT -``` +```sql SELECT COUNT([*|field_name]) FROM tb_name [WHERE clause]; ``` -**Description**: Get the number of rows in a table or a super table. +**Description**: The number of records in the specified fields. -**Return value type**: BIGINT. +**Return value type**: BIGINT -**Applicable data types**: All data types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables **More explanation**: -- Wildcard (\*) is used to represent all columns. If \* used `COUNT` function will get the total number of all rows. -- The number of non-NULL values will be returned if this function is used on a specific column. +- Wildcard (\*) is used to represent all columns. +If you input a specific column, the number of non-null values in the column is returned. + ### ELAPSED -```mysql -SELECT ELAPSED(field_name[, time_unit]) FROM { tb_name | stb_name } [WHERE clause] [INTERVAL(interval [, offset]) [SLIDING sliding]]; +```sql +SELECT ELAPSED(ts_primary_key [, time_unit]) FROM { tb_name | stb_name } [WHERE clause] [INTERVAL(interval [, offset]) [SLIDING sliding]]; ``` -**Description**：`elapsed` function can be used to calculate the continuous time length in which there is valid data. If it's used with `INTERVAL` clause, the returned result is the calcualted time length within each time window. If it's used without `INTERVAL` caluse, the returned result is the calculated time duration within the specified time range. Please be noted that the calculated time duration is in the specified `time_unit`. +**Description**：`elapsed` function can be used to calculate the continuous time length in which there is valid data. If it's used with `INTERVAL` clause, the returned result is the calcualted time length within each time window. If it's used without `INTERVAL` caluse, the returned result is the calculated time length within the specified time range. Please be noted that the return value of `elapsed` is the number of `time_unit` in the calculated time length. -**Return value type**：DOUBLE. +**Return value type**: Double if the input value is not NULL; -**Applicable data type**：TIMESTAMP. +**Return value type**: TIMESTAMP -**Applicable tables**: table, STable, outter in nested query. +**Applicable tables**: table, STable, outter in nested query **Explanations**： - - `field_name` parameter can only be the first column of a table, i.e. timestamp primary key. -- The minimum value of `time_unit` is the time precision of the database. If `time_unit` is not specified, the time precision of the database is used as the default ime unit. +- The minimum value of `time_unit` is the time precision of the database. If `time_unit` is not specified, the time precision of the database is used as the default time unit. Time unit specified by `time_unit` can be: + 1b (nanoseconds), 1u (microseconds), 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), or 1w (weeks) - It can be used with `INTERVAL` to get the time valid time length of each time window. Please be noted that the return value is same as the time window for all time windows except for the first and the last time window. - `order by asc/desc` has no effect on the result. - `group by tbname` must be used together when `elapsed` is used against a STable. - `group by` must NOT be used together when `elapsed` is used against a table or sub table. -- When used in nested query, it's only applicable when the inner query outputs an implicit timestamp column as the primary key. For example, `select elapsed(ts) from (select diff(value) from sub1)` is legal usage while `select elapsed(ts) from (select * from sub1)` is not. -- It cannot be used with `leastsquares`, `diff`, `derivative`, `top`, `bottom`, `last_row`, `interp`. +- When used in nested query, it's only applicable when the inner query outputs an implicit timestamp column as the primary key. For example, `select elapsed(ts) from (select diff(value) from sub1)` is legal usage while `select elapsed(ts) from (select * from sub1)` is not. In addition, because elapsed has a strict dependency on the timeline, a statement like `select elapsed(ts) from (select diff(value) from st group by tbname) will return a meaningless result. +- It can't be used with `leastsquares`, `diff`, `derivative`, `top`, `bottom`, `last_row`, `interp`. + ### LEASTSQUARES -``` +```sql SELECT LEASTSQUARES(field_name, start_val, step_val) FROM tb_name [WHERE clause]; ``` **Description**: The linear regression function of the specified column and the timestamp column (primary key), `start_val` is the initial value and `step_val` is the step value. -**Return value type**: VARCHAR string in the format of "(slope, intercept)". +**Return value type**: A string in the format of "(slope, intercept)" + +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Applicable table types**: table only -**Applicable table types**: table only. ### SPREAD -``` +```sql SELECT SPREAD(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The difference between the max and the min value of a specific column. +**Description**: The difference between the max and the min of a specific column -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable data types**: Integers and timestamps -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: Can be used on a column of TIMESTAMP type, the result time unit precision is same as the current database in use. ### STDDEV -``` +```sql SELECT STDDEV(field_name) FROM tb_name [WHERE clause]; ``` -**Description**: Standard deviation of a specific column in a table or STable. +**Description**: Standard deviation of a specific column in a table or STable -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric + +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. ### SUM -``` +```sql SELECT SUM(field_name) FROM tb_name [WHERE clause]; ``` -**Description**: The summation of values of a specific column in a table or STable. +**Description**: The sum of a specific column in a table or STable + +**Return value type**: DOUBLE or BIGINT -**Return value type**: DOUBLE. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. ### HYPERLOGLOG -``` +```sql SELECT HYPERLOGLOG(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**:The cardinal number of a specific column is returned by using hyperloglog algorithm. +**Description**： + The cardinal number of a specific column is returned by using hyperloglog algorithm. The benefit of using hyperloglog algorithm is that the memory usage is under control when the data volume is huge. + However, when the data volume is very small, the result may be not accurate, it's recommented to use `select count(data) from (select unique(col) as data from table)` in this case. + +**Return value type**: Integer -**Return value type**: INTEGER. +**Applicable data types**: Numeric -**Applicable data types**: All data types. +**Applicable table types**: standard tables and supertables -**More explanations**: The benefit of using hyperloglog algorithm is that the memory usage is under control when the data volume is huge. However, when the data volume is very small, the result may be not accurate, it's recommented to use `select count(data) from (select unique(col) as data from table)` in this case. ### HISTOGRAM -``` +```sql SELECT HISTOGRAM(field_name，bin_type, bin_description, normalized) FROM tb_name [WHERE clause]; ``` **Description**：Returns count of data points in user-specified ranges. -**Return value type**：DOUBLE or BIGINT, depends on normalized parameter settings. +**Return value type** If normalized is set to 1, a DOUBLE is returned; otherwise a BIGINT is returned -**Applicable data type**：Numerical types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: table, STable **Explanations**： - - bin_type: parameter to indicate the bucket type, valid inputs are: "user_input", "linear_bin", "log_bin"。 -- bin_description: parameter to describe the rule to generate buckets，can be in the following JSON formats for each bin_type respectively: - - - "user_input": "[1, 3, 5, 7]": User specified bin values. - - - "linear_bin": "{"start": 0.0, "width": 5.0, "count": 5, "infinity": true}" - "start" - bin starting point. - "width" - bin offset. - "count" - number of bins generated. - "infinity" - whether to add（-inf, inf）as start/end point in generated set of bins. - The above "linear_bin" descriptor generates a set of bins: [-inf, 0.0, 5.0, 10.0, 15.0, 20.0, +inf]. +- bin_description: parameter to describe how to generate buckets，can be in the following JSON formats for each bin_type respectively: + - "user_input": "[1, 3, 5, 7]": + User specified bin values. + + - "linear_bin": "{"start": 0.0, "width": 5.0, "count": 5, "infinity": true}" + "start" - bin starting point. "width" - bin offset. "count" - number of bins generated. "infinity" - whether to add（-inf, inf）as start/end point in generated set of bins. + The above "linear_bin" descriptor generates a set of bins: [-inf, 0.0, 5.0, 10.0, 15.0, 20.0, +inf]. + + - "log_bin": "{"start":1.0, "factor": 2.0, "count": 5, "infinity": true}" + "start" - bin starting point. "factor" - exponential factor of bin offset. "count" - number of bins generated. "infinity" - whether to add（-inf, inf）as start/end point in generated range of bins. + The above "linear_bin" descriptor generates a set of bins: [-inf, 1.0, 2.0, 4.0, 8.0, 16.0, +inf]. +- normalized: setting to 1/0 to turn on/off result normalization. Valid values are 0 or 1. - - "log_bin": "{"start":1.0, "factor": 2.0, "count": 5, "infinity": true}" - "start" - bin starting point. - "factor" - exponential factor of bin offset. - "count" - number of bins generated. - "infinity" - whether to add（-inf, inf）as start/end point in generated range of bins. - The above "log_bin" descriptor generates a set of bins:[-inf, 1.0, 2.0, 4.0, 8.0, 16.0, +inf]. - -- normalized: setting to 1/0 to turn on/off result normalization. ### PERCENTILE -``` +```sql SELECT PERCENTILE(field_name, P) FROM { tb_name } [WHERE clause]; ``` **Description**: The value whose rank in a specific column matches the specified percentage. If such a value matching the specified percentage doesn't exist in the column, an interpolation value will be returned. -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable column types**: Numeric -**Applicable table types**: table. +**Applicable table types**: table only **More explanations**: _P_ is in range [0,100], when _P_ is 0, the result is same as using function MIN; when _P_ is 100, the result is same as function MAX. -## Selector Functions -Selector functiosn choose one or more rows in the query result according to the semantics. You can specify to output primary timestamp column and other columns including tbname and tags so that you can easily know which rows the selected values belong to. +## Selection Functions + +Selection functions return one or more results depending. You can specify the timestamp column, tbname pseudocolumn, or tag columns to show which rows contain the selected value. ### BOTTOM -``` +```sql SELECT BOTTOM(field_name, K) FROM { tb_name | stb_name } [WHERE clause]; ``` **Description**: The least _k_ values of a specific column in a table or STable. If a value has multiple occurrences in the column but counting all of them in will exceed the upper limit _k_, then a part of them will be returned randomly. -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanation**: -- _k_ must be in range [1,100]. -- The timestamp associated with the selected values are returned too. -- Can't be used with `FILL`. +- _k_ must be in range [1,100] +- The timestamp associated with the selected values are returned too +- Can't be used with `FILL` ### FIRST -``` +```sql SELECT FIRST(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The first non-null value of a specific column in a table or STable. +**Description**: The first non-null value of a specific column in a table or STable -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: All data types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanation**: - FIRST(\*) can be used to get the first non-null value of all columns - NULL will be returned if all the values of the specified column are all NULL @@ -842,17 +845,17 @@ SELECT FIRST(field_name) FROM { tb_name | stb_name } [WHERE clause]; ### INTERP -``` +```sql SELECT INTERP(field_name) FROM { tb_name | stb_name } [WHERE where_condition] [ RANGE(timestamp1,timestamp2) ] [EVERY(interval)] [FILL ({ VALUE | PREV | NULL | LINEAR | NEXT})]; ``` **Description**: The value that matches the specified timestamp range is returned, if existing; or an interpolation value is returned. -**Return value type**: Same as the column being operated upon. +**Return value type**: Same as the column being operated upon -**Applicable data types**: Numeric data types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable, nested query. +**Applicable table types**: standard tables and supertables **More explanations** @@ -866,201 +869,208 @@ SELECT INTERP(field_name) FROM { tb_name | stb_name } [WHERE where_condition] [ ### LAST -``` +```sql SELECT LAST(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The last non-NULL value of a specific column in a table or STable. +**Description**: The last non-NULL value of a specific column in a table or STable -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: All data types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanation**: - LAST(\*) can be used to get the last non-NULL value of all columns - If the values of a column in the result set are all NULL, NULL is returned for that column; if all columns in the result are all NULL, no result will be returned. - When it's used on a STable, if there are multiple values with the timestamp in the result set, one of them will be returned randomly and it's not guaranteed that the same value is returned if the same query is run multiple times. + ### LAST_ROW -``` +```sql SELECT LAST_ROW(field_name) FROM { tb_name | stb_name }; ``` -**Description**: The last row of a table or STable. +**Description**: The last row of a table or STable -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: All data type. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables **More explanations**: -- When it's used against a STable, multiple rows with the same and largest timestamp may exist, in this case one of them is returned randomly and it's not guaranteed that the result is same if the query is run multiple times. -- Cannot be used with `INTERVAL`. +- When it's used on a STable, if there are multiple values with the timestamp in the result set, one of them will be returned randomly and it's not guaranteed that the same value is returned if the same query is run multiple times. +- Can't be used with `INTERVAL`. ### MAX -``` +```sql SELECT MAX(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The maximum value of a specific column of a table or STable. +**Description**: The maximum value of a specific column of a table or STable + +**Return value type**:Same as the data type of the column being operated upon -**Return value type**: Same as the data type of the column being operated upon. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. ### MIN -``` +```sql SELECT MIN(field_name) FROM {tb_name | stb_name} [WHERE clause]; ``` -**Description**: The minimum value of a specific column in a table or STable. +**Description**: The minimum value of a specific column in a table or STable + +**Return value type**:Same as the data type of the column being operated upon -**Return value type**: Same as the data type of the column being operated upon. +**Applicable data types**: Numeric -**Applicable data types**: Numeric types. +**Applicable table types**: standard tables and supertables -**Applicable table types**: table, STable. ### MODE -``` +```sql SELECT MODE(field_name) FROM tb_name [WHERE clause]; ``` **Description**:The value which has the highest frequency of occurrence. NULL is returned if there are multiple values which have highest frequency of occurrence. -**Return value type**:Same as the data type of the column being operated upon. +**Return value type**: Same as the input data -**Applicable data types**: All data types. +**Applicable data types**: Numeric + +**Applicable table types**: standard tables and supertables -**More explanations**:Considering the number of returned result set is unpredictable, it's suggested to limit the number of unique values to 100,000, otherwise error will be returned. ### SAMPLE ```sql - SELECT SAMPLE(field_name, K) FROM { tb_name | stb_name } [WHERE clause] +SELECT SAMPLE(field_name, K) FROM { tb_name | stb_name } [WHERE clause] ``` **Description**: _k_ sampling values of a specific column. The applicable range of _k_ is [1,1000]. -**Return value type**: Same as the column being operated. +**Return value type**: Same as the column being operated plus the associated timestamp -**Applicable data types**: All data types. +**Applicable data types**: Any data type except for tags of STable -**Applicable table types**: table, STable. +**Applicable nested query**: Inner query and Outer query -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanations**: + +This function cannot be used in expression calculation. +- Must be used with `PARTITION BY tbname` when it's used on a STable to force the result on each single timeline -- Arithmetic operation cannot be operated on the result of `SAMPLE` function -- Must be used with `Partition by tbname` when it's used on a STable to force the result on each single timeline. ### TAIL -``` +```sql SELECT TAIL(field_name, k, offset_val) FROM {tb_name | stb_name} [WHERE clause]; ``` **Description**: The next _k_ rows are returned after skipping the last `offset_val` rows, NULL values are not ignored. `offset_val` is optional parameter. When it's not specified, the last _k_ rows are returned. When `offset_val` is used, the effect is same as `order by ts desc LIMIT k OFFSET offset_val`. -**Parameter value range**: k: [1,100] offset_val: [0,100]. +**Parameter value range**: k: [1,100] offset_val: [0,100] + +**Return value type**:Same as the data type of the column being operated upon -**Return value type**: Same as the column being operated upon. +**Applicable data types**: Any data type except for timestamp, i.e. the primary key + +**Applicable table types**: standard tables and supertables -**Applicable data types**: All data types. ### TOP -``` +```sql SELECT TOP(field_name, K) FROM { tb_name | stb_name } [WHERE clause]; ``` **Description**: The greatest _k_ values of a specific column in a table or STable. If a value has multiple occurrences in the column but counting all of them in will exceed the upper limit _k_, then a part of them will be returned randomly. -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanation**: -- _k_ must be in range [1,100]. -- The timestamp associated with the selected values are returned too. -- Cannot be used with `FILL`. +- _k_ must be in range [1,100] +- The timestamp associated with the selected values are returned too +- Can't be used with `FILL` ### UNIQUE -``` +```sql SELECT UNIQUE(field_name) FROM {tb_name | stb_name} [WHERE clause]; ``` -**Description**: The values that occur the first time in the specified column. The effect is similar to `distinct` keyword, but it can also be used to match tags or timestamp. +**Description**: The values that occur the first time in the specified column. The effect is similar to `distinct` keyword, but it can also be used to match tags or timestamp. The first occurrence of a timestamp or tag is used. -**Return value type**: Same as the column or tag being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: All data types. +**Applicable column types**: Any data types except for timestamp -**More explanations**: +**Applicable table types**: table, STable -- It can be used against table or STable, but can't be used together with time window, like `interval`, `state_window` or `session_window` . -- Considering the number of result sets is unpredictable, it's suggested to limit the distinct values under 100,000 to control the memory usage, otherwise error will be returned. -## Time-Series Specific Functions +## Time-Series Extensions -TDengine provides a set of time-series specific functions to better meet the requirements in querying time-series data. In general databases, similar functionalities can only be achieved with much more complex syntax and much worse performance. TDengine provides these functionalities in builtin functions so that the burden on user side is minimized. +TDengine includes extensions to standard SQL that are intended specifically for time-series use cases. The functions enabled by these extensions require complex queries to implement in general-purpose databases. By offering them as built-in extensions, TDengine reduces user workload. ### CSUM ```sql - SELECT CSUM(field_name) FROM { tb_name | stb_name } [WHERE clause] +SELECT CSUM(field_name) FROM { tb_name | stb_name } [WHERE clause] ``` **Description**: The cumulative sum of each row for a specific column. The number of output rows is same as that of the input rows. -**Return value type**: BIGINT for signed integer input types; UNSIGNED BIGINT for unsigned integer input types; DOUBLE for floating point input types. +**Return value type**: Long integer for integers; Double for floating points. uint64_t for unsigned integers -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable nested query**: Inner query and Outer query. +**Applicable table types**: standard tables and supertables + +**More explanations**: + +- Arithmetic operation can't be performed on the result of `csum` function +- Can only be used with aggregate functions This function can be used with supertables and standard tables. +- Must be used with `PARTITION BY tbname` when it's used on a STable to force the result on each single timeline -**More explanations**: -- Arithmetic operation cannot be performed on the result of `csum` function. -- Can only be used with aggregate functions. -- `Partition by tbname` must be used together on a STable to force the result on a single timeline. ### DERIVATIVE -``` +```sql SELECT DERIVATIVE(field_name, time_interval, ignore_negative) FROM tb_name [WHERE clause]; ``` **Description**: The derivative of a specific column. The time rage can be specified by parameter `time_interval`, the minimum allowed time range is 1 second (1s); the value of `ignore_negative` can be 0 or 1, 1 means negative values are ignored. -**Return value type**: DOUBLE. - -**Applicable data types**: Numeric types. +**Return value type**: DOUBLE -**Applicable table types**: table, STable. +**Applicable data types**: Numeric -**More explanations**: +**Applicable table types**: standard tables and supertables -- The number of result rows is the number of total rows in the time range subtracted by one, no output for the first row. +**More explanation**: + - It can be used together with `PARTITION BY tbname` against a STable. -- Can be used together with selection of relative columns. E.g. select \_rowts, DERIVATIVE() from. +- It can be used together with a selected column. For example: select \_rowts, DERIVATIVE() from。 ### DIFF @@ -1070,159 +1080,159 @@ SELECT {DIFF(field_name, ignore_negative) | DIFF(field_name)} FROM tb_name [WHER **Description**: The different of each row with its previous row for a specific column. `ignore_negative` can be specified as 0 or 1, the default value is 1 if it's not specified. `1` means negative values are ignored. -**Return value type**: Same as the column being operated upon. +**Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanation**: + +- The number of result rows is the number of rows subtracted by one, no output for the first row +- It can be used together with a selected column. For example: select \_rowts, DIFF() from。 -- The number of result rows is the number of rows subtracted by one, no output for the first row. -- It can be used on STable with `PARTITION by tbname`. -- Can be used together with selection of relative columns. E.g. select \_rowts, DIFF() from. ### IRATE -``` +```sql SELECT IRATE(field_name) FROM tb_name WHERE clause; ``` **Description**: instantaneous rate on a specific column. The last two samples in the specified time range are used to calculate instantaneous rate. If the last sample value is smaller, then only the last sample value is used instead of the difference between the last two sample values. -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: - -- It can be used on stble with `PARTITION BY`, i.e. timelines generated by `PARTITION BY tbname` on a STable. ### MAVG ```sql - SELECT MAVG(field_name, K) FROM { tb_name | stb_name } [WHERE clause] +SELECT MAVG(field_name, K) FROM { tb_name | stb_name } [WHERE clause] ``` **Description**: The moving average of continuous _k_ values of a specific column. If the number of input rows is less than _k_, nothing is returned. The applicable range of _k_ is [1,1000]. -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable nested query**: Inner query and Outer query. +**Nested query**: It can be used in both the outer query and inner query in a nested query. -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +**More explanations**: + +- Arithmetic operation can't be performed on the result of `MAVG`. +- Can only be used with data columns, can't be used with tags. - Can't be used with aggregate functions. +- Must be used with `PARTITION BY tbname` when it's used on a STable to force the result on each single timeline -- Arithmetic operation cannot be performed on the result of `MAVG`. -- Cannot be used with aggregate functions. -- Must be used with `PARTITION BY tbname` when it's used on a STable to force the result on each single timeline. ### STATECOUNT -``` +```sql SELECT STATECOUNT(field_name, oper, val) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The number of continuous rows satisfying the specified conditions for a specific column. If the specified condition is evaluated as true, the number is increased by 1; otherwise the number is reset to -1. If the input value is NULL, then the corresponding row is skipped. +**Description**: The number of continuous rows satisfying the specified conditions for a specific column. The result is shown as an extra column for each row. If the specified condition is evaluated as true, the number is increased by 1; otherwise the number is reset to -1. If the input value is NULL, then the corresponding row is skipped. **Applicable parameter values**: -- oper : Can be one of "LT" (lower than), "GT" (greater than), "LE" (lower than or euqal to), "GE" (greater than or equal to), "NE" (not equal to), "EQ" (equal to). -- val ： Numeric types. +- oper : Can be one of `LT` (lower than), `GT` (greater than), `LE` (lower than or equal to), `GE` (greater than or equal to), `NE` (not equal to), `EQ` (equal to), the value is case insensitive +- val ： Numeric types -**Return value type**: INTEGER. +**Return value type**: Integer -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable nested query**: Outer query only -**Applicable nested query**: Outer query only. +**Applicable table types**: standard tables and supertables **More explanations**: -- Must be used together with `PARTITION BY tbname` when it's used on a STable to force the result into each single timeline. -- Cannot be used with window operation, like interval/state_window/session_window. +- Must be used together with `PARTITION BY tbname` when it's used on a STable to force the result into each single timeline] +- Can't be used with window operation, like interval/state_window/session_window + ### STATEDURATION -``` +```sql SELECT stateDuration(field_name, oper, val, unit) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The length of time range in which all rows satisfy the specified condition for a specific column. The length for the first row that satisfies the condition is 0. Next, if the condition is evaluated as true for a row, the time interval between current row and its previous row is added up to the time range; otherwise the time range length is reset to -1. If the value of the column is NULL, the corresponding row is skipped. +**Description**: The length of time range in which all rows satisfy the specified condition for a specific column. The result is shown as an extra column for each row. The length for the first row that satisfies the condition is 0. Next, if the condition is evaluated as true for a row, the time interval between current row and its previous row is added up to the time range; otherwise the time range length is reset to -1. If the value of the column is NULL, the corresponding row is skipped. **Applicable parameter values**: -- oper : Can be one of "LT" (lower than), "GT" (greater than), "LE" (lower than or euqal to), "GE" (greater than or equal to), "NE" (not equal to), "EQ" (equal to). -- val : Numeric types. -- unit : The unit of time interval, can be: 1b(nanosecond), 1u(microsecond),1a(millisecond),1s(second),1m(minute),1h(hour),1d(day),1w(week). If not specified, default is same as the current database time precision in use. +- oper : Can be one of `LT` (lower than), `GT` (greater than), `LE` (lower than or equal to), `GE` (greater than or equal to), `NE` (not equal to), `EQ` (equal to), the value is case insensitive +- val ： Numeric types +- unit: The unit of time interval. Enter one of the following options: 1b (nanoseconds), 1u (microseconds), 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), or 1w (weeks) If you do not enter a unit of time, the precision of the current database is used by default. -**Return value type**: INTEGER. +**Return value type**: Integer -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable nested query**: Outer query only -**Applicable nested query**: Outer query only. +**Applicable table types**: standard tables and supertables **More explanations**: -- Must be used together with `PARTITION BY tbname` when it's used on a STable to force the result into each single timeline. -- Cannot be used with window operation, like interval/state_window/session_window. +- Must be used together with `PARTITION BY tbname` when it's used on a STable to force the result into each single timeline] +- Can't be used with window operation, like interval/state_window/session_window + ### TWA -``` +```sql SELECT TWA(field_name) FROM tb_name WHERE clause; ``` -**Description**: Time weighted average on a specific column within a time range. +**Description**: Time weighted average on a specific column within a time range -**Return value type**: DOUBLE. +**Return value type**: DOUBLE -**Applicable data types**: Numeric types. +**Applicable data types**: Numeric -**Applicable table types**: table, STable. +**Applicable table types**: standard tables and supertables -**More explanations**: +- Must be used together with `PARTITION BY tbname` to force the result into each single timeline. -- It can be used on stable with `PARTITION BY`, i.e. timelines generated by `PARTITION BY tbname` on a STable. ## System Information Functions ### DATABASE -``` +```sql SELECT DATABASE(); ``` -**Description**：Return the current database being used. If the user doesn't specify database when logon and doesn't use `USE` SQL command to switch the datbase, this function returns NULL. +**Description**: The current database. If no database is specified upon logging in and no database is specified with `USE` after login, NULL will be returned by `select database()`. + ### CLIENT_VERSION -``` +```sql SELECT CLIENT_VERSION(); ``` -**Description**：Return the client version. +**Description**: The client version. ### SERVER_VERSION -``` +```sql SELECT SERVER_VERSION(); ``` -**Description**：Returns the server version. +**Description**: The server version. ### SERVER_STATUS -``` +```sql SELECT SERVER_VERSION(); ``` -**Description**：Returns the server's status. +**Description**: The server status. diff --git a/docs/en/12-taos-sql/12-distinguished.md b/docs/en/12-taos-sql/12-distinguished.md index d2f7cf66b63521d157a6e05f1dd8d93658d65549..7390fe983f0dff7476f606f9989aae8f73c52c0e 100644 --- a/docs/en/12-taos-sql/12-distinguished.md +++ b/docs/en/12-taos-sql/12-distinguished.md @@ -1,60 +1,35 @@ --- -sidebar_label: Distinguished -title: Distinguished Query for Time Series Database +sidebar_label: Time-Series Extensions +title: Time-Series Extensions --- -Aggregation by time window is supported in TDengine. For example, in the case where temperature sensors report the temperature every seconds, the average temperature for every 10 minutes can be retrieved by performing a query with a time window. -Window related clauses are used to divide the data set to be queried into subsets and then aggregation is performed across the subsets. There are three kinds of windows: time window, status window, and session window. There are two kinds of time windows: sliding window and flip time/tumbling window. +As a purpose-built database for storing and processing time-series data, TDengine provides time-series-specific extensions to standard SQL. -## Time Window +These extensions include tag-partitioned queries and windowed queries. -The `INTERVAL` clause is used to generate time windows of the same time interval. The `SLIDING` parameter is used to specify the time step for which the time window moves forward. The query is performed on one time window each time, and the time window moves forward with time. When defining a continuous query, both the size of the time window and the step of forward sliding time need to be specified. As shown in the figure blow, [t0s, t0e] ，[t1s , t1e]， [t2s, t2e] are respectively the time ranges of three time windows on which continuous queries are executed. The time step for which time window moves forward is marked by `sliding time`. Query, filter and aggregate operations are executed on each time window respectively. When the time step specified by `SLIDING` is same as the time interval specified by `INTERVAL`, the sliding time window is actually a flip time/tumbling window. - -![TDengine Database Time Window](./timewindow-1.webp) +## Tag-Partitioned Queries -`INTERVAL` and `SLIDING` should be used with aggregate functions and select functions. The SQL statement below is illegal because no aggregate or selection function is used with `INTERVAL`. +When you query a supertable, you may need to partition the supertable by tag and perform additional operations on a specific partition. In this case, you can use the following SQL clause: +```sql +PARTITION BY part_list ``` -SELECT * FROM temp_tb_1 INTERVAL(1m); -``` - -The time step specified by `SLIDING` cannot exceed the time interval specified by `INTERVAL`. The SQL statement below is illegal because the time length specified by `SLIDING` exceeds that specified by `INTERVAL`. - -``` -SELECT COUNT(*) FROM temp_tb_1 INTERVAL(1m) SLIDING(2m); -``` - -When the time length specified by `SLIDING` is the same as that specified by `INTERVAL`, the sliding window is actually a flip/tumbling window. The minimum time range specified by `INTERVAL` is 10 milliseconds (10a) prior to version 2.1.5.0. Since version 2.1.5.0, the minimum time range by `INTERVAL` can be 1 microsecond (1u). However, if the DB precision is millisecond, the minimum time range is 1 millisecond (1a). Please note that the `timezone` parameter should be configured to be the same value in the `taos.cfg` configuration file on client side and server side. - -## Status Window - -In case of using integer, bool, or string to represent the status of a device at any given moment, continuous rows with the same status belong to a status window. Once the status changes, the status window closes. As shown in the following figure, there are two status windows according to status, [2019-04-28 14:22:07，2019-04-28 14:22:10] and [2019-04-28 14:22:11，2019-04-28 14:22:12]. Status window is not applicable to STable for now. -![TDengine Database Status Window](./timewindow-3.webp) +part_list can be any scalar expression, such as a column, constant, scalar function, or a combination of the preceding items. -`STATE_WINDOW` is used to specify the column on which the status window will be based. For example： +A PARTITION BY clause with a tag is processed as follows: -``` -SELECT COUNT(*), FIRST(ts), status FROM temp_tb_1 STATE_WINDOW(status); -``` - -## Session Window +- The PARTITION BY clause must occur after the WHERE clause and cannot be used with a JOIN clause. +- The PARTITION BY clause partitions the super table by the specified tag group, and the specified calculation is performed on each partition. The calculation performed is determined by the rest of the statement - a window clause, GROUP BY clause, or SELECT clause. +- You can use PARTITION BY together with a window clause or GROUP BY clause. In this case, the window or GROUP BY clause takes effect on every partition. For example, the following statement partitions the table by the location tag, performs downsampling over a 10 minute window, and returns the maximum value: ```sql -SELECT COUNT(*), FIRST(ts) FROM temp_tb_1 SESSION(ts, tol_val); +select max(current) from meters partition by location interval(10m) ``` -The primary key, i.e. timestamp, is used to determine which session window a row belongs to. If the time interval between two adjacent rows is within the time range specified by `tol_val`, they belong to the same session window; otherwise they belong to two different session windows. As shown in the figure below, if the limit of time interval for the session window is specified as 12 seconds, then the 6 rows in the figure constitutes 2 time windows, [2019-04-28 14:22:10，2019-04-28 14:22:30] and [2019-04-28 14:23:10，2019-04-28 14:23:30], because the time difference between 2019-04-28 14:22:30 and 2019-04-28 14:23:10 is 40 seconds, which exceeds the time interval limit of 12 seconds. - -![TDengine Database Session Window](./timewindow-2.webp) - -If the time interval between two continuous rows are within the time interval specified by `tol_value` they belong to the same session window; otherwise a new session window is started automatically. Session window is not supported on STable for now. - -## More On Window Aggregate - -### Syntax +## Windowed Queries -The full syntax of aggregate by window is as follows: +Aggregation by time window is supported in TDengine. For example, in the case where temperature sensors report the temperature every seconds, the average temperature for every 10 minutes can be retrieved by performing a query with a time window. Window related clauses are used to divide the data set to be queried into subsets and then aggregation is performed across the subsets. There are three kinds of windows: time window, status window, and session window. There are two kinds of time windows: sliding window and flip time/tumbling window. The query syntax is as follows: ```sql SELECT function_list FROM tb_name @@ -63,27 +38,36 @@ SELECT function_list FROM tb_name [STATE_WINDOW(col)] [INTERVAL(interval [, offset]) [SLIDING sliding]] [FILL({NONE | VALUE | PREV | NULL | LINEAR | NEXT})] - -SELECT function_list FROM stb_name - [WHERE where_condition] - [INTERVAL(interval [, offset]) [SLIDING sliding]] - [FILL({NONE | VALUE | PREV | NULL | LINEAR | NEXT})] - [GROUP BY tags] ``` -### Restrictions +The following restrictions apply: + +### Restricted Functions - Aggregate functions and select functions can be used in `function_list`, with each function having only one output. For example COUNT, AVG, SUM, STDDEV, LEASTSQUARES, PERCENTILE, MIN, MAX, FIRST, LAST. Functions having multiple outputs, such as DIFF or arithmetic operations can't be used. - `LAST_ROW` can't be used together with window aggregate. - Scalar functions, like CEIL/FLOOR, can't be used with window aggregate. + +### Other Rules + +- The window clause must occur after the PARTITION BY clause and before the GROUP BY clause. It cannot be used with a GROUP BY clause. +- SELECT clauses on windows can contain only the following expressions: + - Constants + - Aggregate functions + - Expressions that include the preceding expressions. +- The window clause cannot be used with a GROUP BY clause. - `WHERE` clause can be used to specify the starting and ending time and other filter conditions -- `FILL` clause is used to specify how to fill when there is data missing in any window, including: - 1. NONE: No fill (the default fill mode) - 2. VALUE：Fill with a fixed value, which should be specified together, for example `FILL(VALUE, 1.23)` - 3. PREV：Fill with the previous non-NULL value, `FILL(PREV)` - 4. NULL：Fill with NULL, `FILL(NULL)` - 5. LINEAR：Fill with the closest non-NULL value, `FILL(LINEAR)` - 6. NEXT：Fill with the next non-NULL value, `FILL(NEXT)` + +### FILL Clause + +`FILL` clause is used to specify how to fill when there is data missing in any window, including: + +1. NONE: No fill (the default fill mode) +2. VALUE：Fill with a fixed value, which should be specified together, for example `FILL(VALUE, 1.23)` Note: The value filled depends on the data type. For example, if you run FILL(VALUE 1.23) on an integer column, the value 1 is filled. +3. PREV：Fill with the previous non-NULL value, `FILL(PREV)` +4. NULL：Fill with NULL, `FILL(NULL)` +5. LINEAR：Fill with the closest non-NULL value, `FILL(LINEAR)` +6. NEXT：Fill with the next non-NULL value, `FILL(NEXT)` :::info @@ -93,17 +77,66 @@ SELECT function_list FROM stb_name ::: -Aggregate by time window is also used in continuous query, please refer to [Continuous Query](/develop/continuous-query). +### Time Window -## Examples +There are two kinds of time windows: sliding window and flip time/tumbling window. + +The `INTERVAL` clause is used to generate time windows of the same time interval. The `SLIDING` parameter is used to specify the time step for which the time window moves forward. The query is performed on one time window each time, and the time window moves forward with time. When defining a continuous query, both the size of the time window and the step of forward sliding time need to be specified. As shown in the figure blow, [t0s, t0e] ，[t1s , t1e]， [t2s, t2e] are respectively the time ranges of three time windows on which continuous queries are executed. The time step for which time window moves forward is marked by `sliding time`. Query, filter and aggregate operations are executed on each time window respectively. When the time step specified by `SLIDING` is same as the time interval specified by `INTERVAL`, the sliding time window is actually a flip time/tumbling window. + +![TDengine Database Time Window](./timewindow-1.webp) + +`INTERVAL` and `SLIDING` should be used with aggregate functions and select functions. The SQL statement below is illegal because no aggregate or selection function is used with `INTERVAL`. + +``` +SELECT * FROM temp_tb_1 INTERVAL(1m); +``` + +The time step specified by `SLIDING` cannot exceed the time interval specified by `INTERVAL`. The SQL statement below is illegal because the time length specified by `SLIDING` exceeds that specified by `INTERVAL`. + +``` +SELECT COUNT(*) FROM temp_tb_1 INTERVAL(1m) SLIDING(2m); +``` + +When using time windows, note the following: + +- The window length for aggregation depends on the value of INTERVAL. The minimum interval is 10 ms. You can configure a window as an offset from UTC 0:00. The offset cannot be smaler than the interval. You can use SLIDING to specify the length of time that the window moves forward. +Please note that the `timezone` parameter should be configured to be the same value in the `taos.cfg` configuration file on client side and server side. +- The result set is in ascending order of timestamp when you aggregate by time window. + +### Status Window + +In case of using integer, bool, or string to represent the status of a device at any given moment, continuous rows with the same status belong to a status window. Once the status changes, the status window closes. As shown in the following figure, there are two status windows according to status, [2019-04-28 14:22:07，2019-04-28 14:22:10] and [2019-04-28 14:22:11，2019-04-28 14:22:12]. Status window is not applicable to STable for now. + +![TDengine Database Status Window](./timewindow-3.webp) + +`STATE_WINDOW` is used to specify the column on which the status window will be based. For example: + +``` +SELECT COUNT(*), FIRST(ts), status FROM temp_tb_1 STATE_WINDOW(status); +``` + +### Session Window + +The primary key, i.e. timestamp, is used to determine which session window a row belongs to. As shown in the figure below, if the limit of time interval for the session window is specified as 12 seconds, then the 6 rows in the figure constitutes 2 time windows, [2019-04-28 14:22:10，2019-04-28 14:22:30] and [2019-04-28 14:23:10，2019-04-28 14:23:30] because the time difference between 2019-04-28 14:22:30 and 2019-04-28 14:23:10 is 40 seconds, which exceeds the time interval limit of 12 seconds. + +![TDengine Database Session Window](./timewindow-2.webp) + +If the time interval between two continuous rows are within the time interval specified by `tol_value` they belong to the same session window; otherwise a new session window is started automatically. Session window is not supported on STable for now. + +``` + +SELECT COUNT(*), FIRST(ts) FROM temp_tb_1 SESSION(ts, tol_val); +``` + +### Examples A table of intelligent meters can be created by the SQL statement below： -```sql +``` CREATE TABLE meters (ts TIMESTAMP, current FLOAT, voltage INT, phase FLOAT) TAGS (location BINARY(64), groupId INT); ``` -The average current, maximum current and median of current in every 10 minutes for the past 24 hours can be calculated using the SQL statement below, with missing values filled with the previous non-NULL values. +The average current, maximum current and median of current in every 10 minutes for the past 24 hours can be calculated using the SQL statement below, with missing values filled with the previous non-NULL values. The query statement is as follows: ``` SELECT AVG(current), MAX(current), APERCENTILE(current, 50) FROM meters diff --git a/docs/en/12-taos-sql/13-tmq.md b/docs/en/12-taos-sql/13-tmq.md index 4d9c475a3829456916175d8a0518c47d67bc18ee..befab4f4f01e595564e93ffcfbb0723e13294af0 100644 --- a/docs/en/12-taos-sql/13-tmq.md +++ b/docs/en/12-taos-sql/13-tmq.md @@ -1,41 +1,34 @@ --- -sidebar_label: 消息队列 -title: 消息队列 +sidebar_label: Data Subscription +title: Data Subscription --- -TDengine 3.0.0.0 开始对消息队列做了大幅的优化和增强以简化用户的解决方案。 +The information in this document is related to the TDengine data subscription feature. -## 创建订阅主题 +## Create a Topic ```sql -CREATE TOPIC [IF NOT EXISTS] topic_name AS {subquery | DATABASE db_name | STABLE stb_name }; +CREATE TOPIC [IF NOT EXISTS] topic_name AS subquery; ``` -订阅主题包括三种：列订阅、超级表订阅和数据库订阅。 -**列订阅是**用 subquery 描述，支持过滤和标量函数和 UDF 标量函数，不支持 JOIN、GROUP BY、窗口切分子句、聚合函数和 UDF 聚合函数。列订阅规则如下： +You can use filtering, scalar functions, and user-defined scalar functions with a topic. JOIN, GROUP BY, windows, aggregate functions, and user-defined aggregate functions are not supported. The following rules apply to subscribing to a column: -1. TOPIC 一旦创建则返回结果的字段确定 -2. 被订阅或用于计算的列不可被删除、修改 -3. 列可以新增，但新增的列不出现在订阅结果字段中 -4. 对于 select \*，则订阅展开为创建时所有的列（子表、普通表为数据列，超级表为数据列加标签列） +1. The returned field is determined when the topic is created. +2. Columns to which a consumer is subscribed or that are involved in calculations cannot be deleted or modified. +3. If you add a column, the new column will not appear in the results for the subscription. +4. If you run `SELECT \*`, all columns in the subscription at the time of its creation are displayed. This includes columns in supertables, standard tables, and subtables. Supertables are shown as data columns plus tag columns. -**超级表订阅和数据库订阅**规则如下： -1. 被订阅主体的 schema 变更不受限 -2. 返回消息中 schema 是块级别的，每块的 schema 可能不一样 -3. 列变更后写入的数据若未落盘，将以写入时的 schema 返回 -4. 列变更后写入的数据若未已落盘，将以落盘时的 schema 返回 - -## 删除订阅主题 +## Delete a Topic ```sql DROP TOPIC [IF EXISTS] topic_name; ``` -此时如果该订阅主题上存在 consumer，则此 consumer 会收到一个错误。 +If a consumer is subscribed to the topic that you delete, the consumer will receive an error. -## 查看订阅主题 +## View Topics ## SHOW TOPICS @@ -43,24 +36,24 @@ DROP TOPIC [IF EXISTS] topic_name; SHOW TOPICS; ``` -显示当前数据库下的所有主题的信息。 +The preceding command displays all topics in the current database. -## 创建消费组 +## Create Consumer Group -消费组的创建只能通过 TDengine 客户端驱动或者连接器所提供的 API 创建。 +You can create consumer groups only through the TDengine Client driver or the API provided by a connector. -## 删除消费组 +## Delete Consumer Group ```sql DROP CONSUMER GROUP [IF EXISTS] cgroup_name ON topic_name; ``` -删除主题 topic_name 上的消费组 cgroup_name。 +This deletes the cgroup_name in the topic_name. -## 查看消费组 +## View Consumer Groups ```sql SHOW CONSUMERS; ``` -显示当前数据库下所有活跃的消费者的信息。 +The preceding command displays all consumer groups in the current database. diff --git a/docs/en/12-taos-sql/14-stream.md b/docs/en/12-taos-sql/14-stream.md index 7ff7da2bfb82e282cefb1a554283860d0e683de2..fcd78765104af17285b43749969821ceb98da33b 100644 --- a/docs/en/12-taos-sql/14-stream.md +++ b/docs/en/12-taos-sql/14-stream.md @@ -1,13 +1,13 @@ --- -sidebar_label: 流式计算 -title: 流式计算 +sidebar_label: Stream Processing +title: Stream Processing --- -在时序数据的处理中，经常要对原始数据进行清洗、预处理，再使用时序数据库进行长久的储存。用户通常需要在时序数据库之外再搭建 Kafka、Flink、Spark 等流计算处理引擎，增加了用户的开发成本和维护成本。 +Raw time-series data is often cleaned and preprocessed before being permanently stored in a database. Stream processing components like Kafka, Flink, and Spark are often deployed alongside a time-series database to handle these operations, increasing system complexity and maintenance costs. -使用 TDengine 3.0 的流式计算引擎能够最大限度的减少对这些额外中间件的依赖，真正将数据的写入、预处理、长期存储、复杂分析、实时计算、实时报警触发等功能融为一体，并且，所有这些任务只需要使用 SQL 完成，极大降低了用户的学习成本、使用成本。 +Because stream processing is built in to TDengine, you are no longer reliant on middleware. TDengine offers a unified platform for writing, preprocessing, permanent storage, complex analysis, and real-time computation and alerting. Additionally, you can use SQL to perform all these tasks. -## 创建流式计算 +## Create a Stream ```sql CREATE STREAM [IF NOT EXISTS] stream_name [stream_options] INTO stb_name AS subquery @@ -18,7 +18,7 @@ stream_options: { ``` -其中 subquery 是 select 普通查询语法的子集: +The subquery is a subset of standard SELECT query syntax: ```sql subquery: SELECT [DISTINCT] select_list @@ -26,97 +26,74 @@ subquery: SELECT [DISTINCT] select_list [WHERE condition] [PARTITION BY tag_list] [window_clause] - [group_by_clause] ``` -不支持 order_by，limit，slimit，fill 语句 +Session windows, state windows, and sliding windows are supported. When you configure a session or state window for a supertable, you must use PARTITION BY TBNAME. -例如，如下语句创建流式计算，同时自动创建名为 avg_vol 的超级表，此流计算以一分钟为时间窗口、30 秒为前向增量统计这些电表的平均电压，并将来自 meters 表的数据的计算结果写入 avg_vol 表，不同 partition 的数据会分别创建子表并写入不同子表。 +```sql +window_clause: { + SESSION(ts_col, tol_val) + | STATE_WINDOW(col) + | INTERVAL(interval_val [, interval_offset]) [SLIDING (sliding_val)] +} +``` + +`SESSION` indicates a session window, and `tol_val` indicates the maximum range of the time interval. If the time interval between two continuous rows are within the time interval specified by `tol_val` they belong to the same session window; otherwise a new session window is started automatically. + +For example, the following SQL statement creates a stream and automatically creates a supertable named `avg_vol`. The stream has a 1 minute time window that slides forward in 30 second intervals to calculate the average voltage of the meters supertable. ```sql CREATE STREAM avg_vol_s INTO avg_vol AS SELECT _wstartts, count(*), avg(voltage) FROM meters PARTITION BY tbname INTERVAL(1m) SLIDING(30s); ``` -## 删除流式计算 +## Delete a Stream ```sql DROP STREAM [IF NOT EXISTS] stream_name ``` -仅删除流式计算任务，由流式计算写入的数据不会被删除。 +This statement deletes the stream processing service only. The data generated by the stream is retained. -## 展示流式计算 +## View Streams ```sql SHOW STREAMS; ``` -## 流式计算的触发模式 - -在创建流时，可以通过 TRIGGER 指令指定流式计算的触发模式。 - -对于非窗口计算，流式计算的触发是实时的；对于窗口计算，目前提供 3 种触发模式： - -1. AT_ONCE：写入立即触发 - -2. WINDOW_CLOSE：窗口关闭时触发（窗口关闭由事件时间决定，可配合 watermark 使用，详见《流式计算的乱序数据容忍策略》） - -3. MAX_DELAY time：若窗口关闭，则触发计算。若窗口未关闭，且未关闭时长超过 max delay 指定的时间，则触发计算。 - -由于窗口关闭是由事件时间决定的，如事件流中断、或持续延迟，则事件时间无法更新，可能导致无法得到最新的计算结果。 +## Trigger Stream Processing -因此，流式计算提供了以事件时间结合处理时间计算的 MAX_DELAY 触发模式。 +When you create a stream, you can use the TRIGGER parameter to specify triggering conditions for it. -MAX_DELAY 模式在窗口关闭时会立即触发计算。此外，当数据写入后，计算触发的时间超过 max delay 指定的时间，则立即触发计算 +For non-windowed processing, triggering occurs in real time. For windowed processing, there are three methods of triggering: -## 流式计算的乱序数据容忍策略 +1. AT_ONCE: triggers on write -在创建流时，可以在 stream_option 中指定 watermark。 +2. WINDOW_CLOSE: triggers when the window closes. This is determined by the event time. You can use WINDOW_CLOSE together with `watermark`. For more information, see Stream Processing Strategy for Out-of-Order Data. -流式计算通过 watermark 来度量对乱序数据的容忍程度，watermark 默认为 0。 +3. MAX_DELAY: triggers when the window closes. If the window has not closed but the time elapsed exceeds MAX_DELAY, stream processing is also triggered. -T = 最新事件时间 - watermark +Because the window closing is determined by the event time, a delay or termination of an event stream will prevent the event time from being updated. This may result in an inability to obtain the latest results. -每批到来的数据都会以上述公式更新窗口关闭时间，并将窗口结束时间 < T 的所有打开的窗口关闭，若触发模式为 WINDOW_CLOSE 或 MAX_DELAY，则推送窗口聚合结果。 +For this reason, MAX_DELAY is provided as a way to ensure that processing occurs even if the window does not close. -流式计算的过期数据处理策略 -对于已关闭的窗口，再次落入该窗口中的数据被标记为过期数据，对于过期数据，流式计算提供两种处理方式： +MAX_DELAY also triggers when the window closes. Additionally, if a write occurs but the processing is not triggered before MAX_DELAY expires, processing is also triggered. -1. 直接丢弃：这是常见流式计算引擎提供的默认（甚至是唯一）计算模式 +## Stream Processing Strategy for Out-of-Order Data -2. 重新计算：从 TSDB 中重新查找对应窗口的所有数据并重新计算得到最新结果 +When you create a stream, you can specify a watermark in the `stream_option` parameter. -无论在哪种模式下，watermark 都应该被妥善设置，来得到正确结果（直接丢弃模式）或避免频繁触发重算带来的性能开销（重新计算模式）。 +The watermark is used to specify the tolerance for out-of-order data. The default value is 0. -## 流式计算的数据填充策略 +T = latest event time - watermark -TODO +The window closing time for each batch of data that arrives at the system is updated using the preceding formula, and all windows are closed whose closing time is less than T. If the triggering method is WINDOW_CLOSE or MAX_DELAY, the aggregate result for the window is pushed. -## 流式计算与会话窗口（session window） +Stream processing strategy for expired data +The data in expired windows is tagged as expired. TDengine stream processing provides two methods for handling such data: -```sql -window_clause: { - SESSION(ts_col, tol_val) - | STATE_WINDOW(col) - | INTERVAL(interval_val [, interval_offset]) [SLIDING (sliding_val)] [FILL(fill_mod_and_val)] -} -``` - -其中，SESSION 是会话窗口，tol_val 是时间间隔的最大范围。在 tol_val 时间间隔范围内的数据都属于同一个窗口，如果连续的两条数据的时间超过 tol_val，则自动开启下一个窗口。 +1. Drop the data. This is the default and often only handling method for most stream processing engines. -## 流式计算的监控与流任务分布查询 +2. Recalculate the data. In this method, all data in the window is reobtained from the database and recalculated. The latest results are then returned. -TODO - -## 流式计算的内存控制与存算分离 - -TODO - -## 流式计算的暂停与恢复 - -```sql -STOP STREAM stream_name; - -RESUME STREAM stream_name; -``` +In both of these methods, configuring the watermark is essential for obtaining accurate results (if expired data is dropped) and avoiding repeated triggers that affect system performance (if expired data is recalculated). diff --git a/docs/en/12-taos-sql/16-operators.md b/docs/en/12-taos-sql/16-operators.md index 0ca9ec49430a66384400bc41cd08562b3d5d28c7..c426e2879342e430c61c4f8133aa9f8186888941 100644 --- a/docs/en/12-taos-sql/16-operators.md +++ b/docs/en/12-taos-sql/16-operators.md @@ -5,62 +5,62 @@ title: Operators ## Arithmetic Operators -| # | **Operator** | **Data Types** | **Description** | -| --- | :----------: | -------------- | --------------------------------------------------------- | -| 1 | +, - | Numeric Types | Representing positive or negative numbers, unary operator | -| 2 | +, - | Numeric Types | Addition and substraction, binary operator | -| 3 | \*, / | Numeric Types | Multiplication and division, binary oeprator | -| 4 | % | Numeric Types | Taking the remainder, binary operator | +| # | **Operator** | **Supported Data Types** | **Description** | +| --- | :--------: | -------------- | -------------------------- | +| 1 | +, - | Numeric | Expresses sign. Unary operators. | +| 2 | +, - | Numeric | Expresses addition and subtraction. Binary operators. | +| 3 | \*, / | Numeric | Expresses multiplication and division. Binary operators. | +| 4 | % | Numeric | Expresses modulo. Binary operator. | ## Bitwise Operators -| # | **Operator** | **Data Types** | **Description** | -| --- | :----------: | -------------- | ----------------------------- | -| 1 | & | Numeric Types | Bitewise AND, binary operator | -| 2 | \| | Numeric Types | Bitewise OR, binary operator | +| # | **Operator** | **Supported Data Types** | **Description** | +| --- | :--------: | -------------- | ------------------ | +| 1 | & | Numeric | Bitwise AND. Binary operator. | +| 2 | \| | Numeric | Bitwise OR. Binary operator. | -## JSON Operator +## JSON Operators -`->` operator can be used to get the value of a key in a column of JSON type, the left oeprand is the column name, the right operand is a string constant. For example, `col->'name'` returns the value of key `'name'`. +The `->` operator returns the value for a key in JSON column. Specify the column indicator on the left of the operator and the key name on the right of the operator. For example, `col->name` returns the value of the name key. -## Set Operator +## Set Operators -Set operators are used to combine the results of two queries into single result. A query including set operators is called a combined query. The number of rows in each result in a combined query must be same, and the type is determined by the first query's result, the type of the following queriess result must be able to be converted to the type of the first query's result, the conversion rule is same as `CAST` function. +Set operators combine the results of two queries. Queries that include set operators are known as compound queries. The expressions corresponding to each query in the select list in a compound query must match in number. The results returned take the data type of the first query, and the data type returned by subsequent queries must be convertible into the data type of the first query. The conditions of the `CAST` function apply to this conversion. -TDengine provides 2 set operators: `UNION ALL` and `UNION`. `UNION ALL` combines the results without removing duplicate data. `UNION` combines the results and remove duplicate data rows. In single SQL statement, at most 100 set operators can be used. +TDengine supports the `UNION` and `UNION ALL` operations. UNION ALL collects all query results and returns them as a composite result without deduplication. UNION collects all query results and returns them as a deduplicated composite result. In a single SQL statement, at most 100 set operators can be supported. -## Comparsion Operator +## Comparison Operators -| # | **Operator** | **Data Types** | **Description** | -| --- | :---------------: | ------------------------------------------------------------------- | ----------------------------------------------- | -| 1 | = | Except for BLOB, MEDIUMBLOB and JSON | Equal | -| 2 | <\>, != | Except for BLOB, MEDIUMBLOB, JSON and primary key of timestamp type | Not equal | -| 3 | \>, < | Except for BLOB, MEDIUMBLOB and JSON | Greater than, less than | -| 4 | \>=, <= | Except for BLOB, MEDIUMBLOB and JSON | Greater than or equal to, less than or equal to | -| 5 | IS [NOT] NULL | Any types | Is NULL or NOT | -| 6 | [NOT] BETWEEN AND | Except for BLOB, MEDIUMBLOB and JSON | In a value range or not | -| 7 | IN | Except for BLOB, MEDIUMBLOB, JSON and primary key of timestamp type | In a list of values or not | -| 8 | LIKE | BINARY, NCHAR and VARCHAR | Wildcard matching | -| 9 | MATCH, NMATCH | BINARY, NCHAR and VARCHAR | Regular expression matching | -| 10 | CONTAINS | JSON | If A key exists in JSON | +| # | **Operator** | **Supported Data Types** | **Description** | +| --- | :---------------: | -------------------------------------------------------------------- | -------------------- | +| 1 | = | All types except BLOB, MEDIUMBLOB, and JSON | Equal to | +| 2 | <\>, != | All types except BLOB, MEDIUMBLOB, and JSON; the primary key (timestamp) is also not supported | Not equal to | +| 3 | \>, < | All types except BLOB, MEDIUMBLOB, and JSON | Greater than and less than | +| 4 | \>=, <= | All types except BLOB, MEDIUMBLOB, and JSON | Greater than or equal to and less than or equal to | +| 5 | IS [NOT] NULL | All types | Indicates whether the value is null | +| 6 | [NOT] BETWEEN AND | All types except BLOB, MEDIUMBLOB, and JSON | Closed interval comparison | +| 7 | IN | All types except BLOB, MEDIUMBLOB, and JSON; the primary key (timestamp) is also not supported | Equal to any value in the list | +| 8 | LIKE | BINARY, NCHAR, and VARCHAR | Wildcard match | +| 9 | MATCH, NMATCH | BINARY, NCHAR, and VARCHAR | Regular expression match | +| 10 | CONTAINS | JSON | Indicates whether the key exists | -`LIKE` operator uses wildcard to match a string, the rules are： +LIKE is used together with wildcards to match strings. Its usage is described as follows: -- '%' matches 0 to any number of characters; '\_' matches any single ASCII character. -- \_ can be used to match a `_` in the string, i.e. using escape character backslash `\` -- Wildcard string is 100 bytes at most. Longer a wildcard string is, worse the performance of LIKE operator is. +- '%' matches 0 or any number of characters, '\_' matches any single ASCII character. +- `\_` is used to match the \_ in the string. +- The maximum length of wildcard string is 100 bytes. A very long wildcard string may slowdown the execution performance of `LIKE` operator. -`MATCH` and `NMATCH` operators use regular expressions to match a string, the rules are: +MATCH and NMATCH are used together with regular expressions to match strings. Their usage is described as follows: -- Regular expressions of POSIX standard are supported. -- Only `tbname`, i.e. table name of sub tables, and tag columns of string types can be matched with regular expression, data columns are not supported. -- Regular expression string is 128 bytes at most, and can be adjusted by setting parameter `maxRegexStringLen`, which is a client side configuration and needs to restart the client to take effect. +- Use POSIX regular expression syntax. For more information, see Regular Expressions. +- Regular expression can be used against only table names, i.e. `tbname`, and tags of binary/nchar types, but can't be used against data columns. +- The maximum length of regular expression string is 128 bytes. Configuration parameter `maxRegexStringLen` can be used to set the maximum allowed regular expression. It's a configuration parameter on the client side, and will take effect after restarting the client. ## Logical Operators -| # | **Operator** | **Data Types** | **Description** | -| --- | :----------: | -------------- | ---------------------------------------------------------------------------------------- | -| 1 | AND | BOOL | Logical AND, return TRUE if both conditions are TRUE; return FALSE if any one is FALSE. | -| 2 | OR | BOOL | Logical OR, return TRUE if any condition is TRUE; return FALSE if both are FALSE | +| # | **Operator** | **Supported Data Types** | **Description** | +| --- | :--------: | -------------- | --------------------------------------------------------------------------- | +| 1 | AND | BOOL | Logical AND; if both conditions are true, TRUE is returned; If either condition is false, FALSE is returned. +| 2 | OR | BOOL | Logical OR; if either condition is true, TRUE is returned; If both conditions are false, FALSE is returned. -TDengine uses shortcircut optimization when performing logical operations. For AND operator, if the first condition is evaluated to FALSE, then the second one is not evaluated. For OR operator, if the first condition is evaluated to TRUE, then the second one is not evaluated. +TDengine performs short-path optimization when calculating logical conditions. If the first condition for AND is false, FALSE is returned without calculating the second condition. If the first condition for OR is true, TRUE is returned without calculating the second condition diff --git a/docs/en/12-taos-sql/17-json.md b/docs/en/12-taos-sql/17-json.md index 7460a5e0ba3ce78ee7744569cda460c477cac19c..77f774303316b466a15226f548f84da69be8f92d 100644 --- a/docs/en/12-taos-sql/17-json.md +++ b/docs/en/12-taos-sql/17-json.md @@ -1,60 +1,64 @@ --- +sidebar_label: JSON Type title: JSON Type --- + ## Syntax 1. Tag of type JSON - ```sql - create STable s1 (ts timestamp, v1 int) tags (info json); + ``` + create stable s1 (ts timestamp, v1 int) tags (info json) - create table s1_1 using s1 tags ('{"k1": "v1"}'); + create table s1_1 using s1 tags ('{"k1": "v1"}') ``` 2. "->" Operator of JSON - ```sql - select * from s1 where info->'k1' = 'v1'; + ``` + select * from s1 where info->'k1' = 'v1' - select info->'k1' from s1; + select info->'k1' from s1 ``` 3. "contains" Operator of JSON - ```sql - select * from s1 where info contains 'k2'; + ``` + select * from s1 where info contains 'k2' - select * from s1 where info contains 'k1'; + select * from s1 where info contains 'k1' ``` ## Applicable Operations 1. When a JSON data type is used in `where`, `match/nmatch/between and/like/and/or/is null/is no null` can be used but `in` can't be used. - ```sql + ``` select * from s1 where info->'k1' match 'v*'; select * from s1 where info->'k1' like 'v%' and info contains 'k2'; select * from s1 where info is null; - select * from s1 where info->'k1' is not null; + select * from s1 where info->'k1' is not null ``` 2. A tag of JSON type can be used in `group by`, `order by`, `join`, `union all` and sub query; for example `group by json->'key'` 3. `Distinct` can be used with a tag of type JSON - ```sql - select distinct info->'k1' from s1; + ``` + select distinct info->'k1' from s1 ``` 4. Tag Operations The value of a JSON tag can be altered. Please note that the full JSON will be overriden when doing this. - The name of a JSON tag can be altered. A tag of JSON type can't be added or removed. The column length of a JSON tag can't be changed. + The name of a JSON tag can be altered. + + A tag of JSON type can't be added or removed. The column length of a JSON tag can't be changed. ## Other Restrictions @@ -64,19 +68,24 @@ title: JSON Type - JSON format： - - The input string for JSON can be empty, i.e. "", "\t", or NULL, but it can't be non-NULL string, bool or array. - - object can be {}, and the entire JSON is empty if so. Key can be "", and it's ignored if so. - - value can be int, double, string, bool or NULL, and it can't be an array. Nesting is not allowed which means that the value of a key can't be JSON. - - If one key occurs twice in JSON, only the first one is valid. - - Escape characters are not allowed in JSON. + - The input string for JSON can be empty, i.e. "", "\t", or NULL, but it can't be non-NULL string, bool or array. + - object can be {}, and the entire JSON is empty if so. Key can be "", and it's ignored if so. + - value can be int, double, string, bool or NULL, and it can't be an array. Nesting is not allowed which means that the value of a key can't be JSON. + - If one key occurs twice in JSON, only the first one is valid. + - Escape characters are not allowed in JSON. - NULL is returned when querying a key that doesn't exist in JSON. - If a tag of JSON is the result of inner query, it can't be parsed and queried in the outer query. -For example, the SQL statements below are not supported. + For example, the SQL statements below are not supported. -```sql; -select jtag->'key' from (select jtag from STable); -select jtag->'key' from (select jtag from STable) where jtag->'key'>0; -``` + ``` + select jtag->'key' from (select jtag from stable) + ``` + + and + + ``` + select jtag->'key' from (select jtag from stable) where jtag->'key'>0 + ``` diff --git a/docs/en/12-taos-sql/18-escape.md b/docs/en/12-taos-sql/18-escape.md index 34ce9f7848a9d60811a23286a6675e8afa4f04fe..46ab35a276ee15b6540b0eaf096482aed210b79c 100644 --- a/docs/en/12-taos-sql/18-escape.md +++ b/docs/en/12-taos-sql/18-escape.md @@ -2,7 +2,7 @@ title: Escape Characters --- -Below table is the list of escape characters used in TDengine. +## Escape Characters | Escape Character | **Actual Meaning** | | :--------------: | ------------------------ | diff --git a/docs/en/12-taos-sql/19-limit.md b/docs/en/12-taos-sql/19-limit.md index ff552fc9771f5b428554acc62e9aeac03a305ecc..0486ea30940cdcb5d034bb730d12c0c120a59cd1 100644 --- a/docs/en/12-taos-sql/19-limit.md +++ b/docs/en/12-taos-sql/19-limit.md @@ -1,59 +1,59 @@ --- -sidebar_label: 命名与边界限制 -title: 命名与边界限制 +sidebar_label: Name and Size Limits +title: Name and Size Limits --- -## 名称命名规则 +## Naming Rules -1. 合法字符：英文字符、数字和下划线 -2. 允许英文字符或下划线开头，不允许以数字开头 -3. 不区分大小写 -4. 转义后表（列）名规则： - 为了兼容支持更多形式的表（列）名，TDengine 引入新的转义符 "`"。可用让表名与关键词不冲突，同时不受限于上述表名称合法性约束检查 - 转义后的表（列）名同样受到长度限制要求，且长度计算的时候不计算转义符。使用转义字符以后，不再对转义字符中的内容进行大小写统一 +1. Names can include letters, digits, and underscores (_). +2. Names can begin with letters or underscores (_) but not with digits. +3. Names are not case-sensitive. +4. Rules for names with escape characters are as follows: + You can escape a name by enclosing it in backticks (`). In this way, you can reuse keyword names for table names. However, the first three naming rules no longer apply. + Table and column names that are enclosed in escape characters are still subject to length limits. When the length of such a name is calculated, the escape characters are not included. Names specified using escape character are case-sensitive. - 例如：\`aBc\` 和 \`abc\` 是不同的表（列）名，但是 abc 和 aBc 是相同的表（列）名。 - 需要注意的是转义字符中的内容必须是可打印字符。 + For example, \`aBc\` and \`abc\` are different table or column names, but "abc" and "aBc" are same names because internally they are all "abc". + Only ASCII visible characters can be used with escape character. -## 密码合法字符集 +## Password Rules `[a-zA-Z0-9!?$%^&*()_–+={[}]:;@~#|<,>.?/]` -去掉了 `` ‘“`\ `` (单双引号、撇号、反斜杠、空格) +The following characters cannot occur in a password: single quotation marks ('), double quotation marks ("), backticks (`), backslashes (\\), and spaces. -## 一般限制 +## General Limits -- 数据库名最大长度为 32 -- 表名最大长度为 192，不包括数据库名前缀和分隔符 -- 每行数据最大长度 48KB （注意：数据行内每个 BINARY/NCHAR 类型的列还会额外占用 2 个字节的存储位置） -- 列名最大长度为 64 -- 最多允许 4096 列，最少需要 2 列，第一列必须是时间戳。 -- 标签名最大长度为 64 -- 最多允许 128 个，至少要有 1 个标签，一个表中标签值的总长度不超过 16KB -- SQL 语句最大长度 1048576 个字符，也可通过客户端配置参数 maxSQLLength 修改，取值范围 65480 ~ 1048576 -- SELECT 语句的查询结果，最多允许返回 4096 列（语句中的函数调用可能也会占用一些列空间），超限时需要显式指定较少的返回数据列，以避免语句执行报错 -- 库的数目，超级表的数目、表的数目，系统不做限制，仅受系统资源限制 -- 数据库的副本数只能设置为 1 或 3 -- 用户名的最大长度是 23 个字节 -- 用户密码的最大长度是 15 个字节 -- 总数据行数取决于可用资源 -- 单个数据库的虚拟结点数上限为 1024 +- Maximum length of database name is 32 bytes +- Maximum length of table name is 192 bytes, excluding the database name prefix and the separator. +- Maximum length of each data row is 48K bytes. Note that the upper limit includes the extra 2 bytes consumed by each column of BINARY/NCHAR type. +- The maximum length of a column name is 64 bytes. +- Maximum number of columns is 4096. There must be at least 2 columns, and the first column must be timestamp. +- The maximum length of a tag name is 64 bytes +- Maximum number of tags is 128. There must be at least 1 tag. The total length of tag values cannot exceed 16 KB. +- Maximum length of single SQL statement is 1 MB (1048576 bytes). It can be configured in the parameter `maxSQLLength` in the client side, the applicable range is [65480, 1048576]. +- At most 4096 columns can be returned by `SELECT`. Functions in the query statement constitute columns. An error is returned if the limit is exceeded. +- Maximum numbers of databases, STables, tables are dependent only on the system resources. +- The number of replicas can only be 1 or 3. +- The maximum length of a username is 23 bytes. +- The maximum length of a password is 15 bytes. +- The maximum number of rows depends on system resources. +- The maximum number of vnodes in a database is 1024. -## 表(列)名合法性说明 +## Restrictions of Table/Column Names -### TDengine 中的表（列）名命名规则如下： +### Name Restrictions of Table/Column -只能由字母、数字、下划线构成，数字不能在首位，长度不能超过 192 字节，不区分大小写。这里表名称不包括数据库名的前缀和分隔符。 +The name of a table or column can only be composed of ASCII characters, digits and underscore and it cannot start with a digit. The maximum length is 192 bytes. Names are case insensitive. The name mentioned in this rule doesn't include the database name prefix and the separator. -### 转义后表（列）名规则： +### Name Restrictions After Escaping -为了兼容支持更多形式的表（列）名，TDengine 引入新的转义符 "`"，可以避免表名与关键词的冲突，同时不受限于上述表名合法性约束检查，转义符不计入表名的长度。 -转义后的表（列）名同样受到长度限制要求，且长度计算的时候不计算转义符。使用转义字符以后，不再对转义字符中的内容进行大小写统一。 +To support more flexible table or column names, new escape character "\`" is introduced in TDengine to avoid the conflict between table name and keywords and break the above restrictions for table names. The escape character is not counted in the length of table name. +With escaping, the string inside escape characters are case sensitive, i.e. will not be converted to lower case internally. The table names specified using escape character are case sensitive. -例如： -\`aBc\` 和 \`abc\` 是不同的表（列）名，但是 abc 和 aBc 是相同的表（列）名。 +For example: +\`aBc\` and \`abc\` are different table or column names, but "abc" and "aBc" are same names because internally they are all "abc". :::note -转义字符中的内容必须是可打印字符。 +The characters inside escape characters must be printable characters. ::: diff --git a/docs/en/12-taos-sql/20-keywords.md b/docs/en/12-taos-sql/20-keywords.md index 6d40deb5a696141cbd7bf8dd01bba6a251ef8908..6f166c8034382b0613845d18470556622106e673 100644 --- a/docs/en/12-taos-sql/20-keywords.md +++ b/docs/en/12-taos-sql/20-keywords.md @@ -1,10 +1,11 @@ --- -title: Keywords +sidebar_label: Reserved Keywords +title: Reserved Keywords --- -There are about 200 keywords reserved by TDengine, they can't be used as the name of database, STable or table with either upper case, lower case or mixed case. +## Keyword List -## Keywords List +There are about 200 keywords reserved by TDengine, they can't be used as the name of database, STable or table with either upper case, lower case or mixed case. The following list shows all reserved keywords: ### A @@ -57,70 +58,70 @@ There are about 200 keywords reserved by TDengine, they can't be used as the nam ### D -- DATABASE -- DATABASES -- DAYS -- DBS -- DEFERRED +- DATABASE +- DATABASES +- DAYS +- DBS +- DEFERRED - DELETE - DELIMITERS -- DESC -- DESCRIBE -- DETACH -- DISTINCT -- DIVIDE -- DNODE -- DNODES -- DOT -- DOUBLE -- DROP +- DESC +- DESCRIBE +- DETACH +- DISTINCT +- DIVIDE +- DNODE +- DNODES +- DOT +- DOUBLE +- DROP ### E -- END -- EQ -- EXISTS -- EXPLAIN +- END +- EQ +- EXISTS +- EXPLAIN ### F -- FAIL -- FILE -- FILL -- FLOAT -- FOR -- FROM -- FSYNC +- FAIL +- FILE +- FILL +- FLOAT +- FOR +- FROM +- FSYNC ### G -- GE -- GLOB +- GE +- GLOB - GRANTS -- GROUP -- GT +- GROUP +- GT ### H -- HAVING +- HAVING ### I - ID - IF -- IGNORE +- IGNORE - IMMEDIA -- IMPORT -- IN +- IMPORT +- IN - INITIAL -- INSERT +- INSERT - INSTEAD -- INT +- INT - INTEGER - INTERVA -- INTO -- IS -- ISNULL +- INTO +- IS +- ISNULL ### J @@ -129,187 +130,147 @@ There are about 200 keywords reserved by TDengine, they can't be used as the nam ### K - KEEP -- KEY +- KEY - KILL ### L -- LE -- LIKE -- LIMIT +- LE +- LIKE +- LIMIT - LINEAR -- LOCAL -- LP +- LOCAL +- LP - LSHIFT -- LT +- LT ### M -- MATCH -- MAXROWS -- MINROWS -- MINUS -- MNODES -- MODIFY -- MODULES +- MATCH +- MAXROWS +- MINROWS +- MINUS +- MNODES +- MODIFY +- MODULES ### N -- NE -- NONE -- NOT +- NE +- NONE +- NOT - NOTNULL -- NOW +- NOW - NULL ### O -- OF +- OF - OFFSET -- OR -- ORDER +- OR +- ORDER ### P - PARTITION -- PASS -- PLUS -- PPS +- PASS +- PLUS +- PPS - PRECISION -- PREV +- PREV - PRIVILEGE ### Q -- QTIME +- QTIME - QUERIE -- QUERY +- QUERY - QUORUM ### R -- RAISE -- REM +- RAISE +- REM - REPLACE - REPLICA -- RESET +- RESET - RESTRIC -- ROW -- RP +- ROW +- RP - RSHIFT ### S -- SCORES -- SELECT -- SEMI +- SCORES +- SELECT +- SEMI - SESSION -- SET -- SHOW -- SLASH +- SET +- SHOW +- SLASH - SLIDING -- SLIMIT +- SLIMIT - SMALLIN - SOFFSET -- STable +- STable - STableS -- STAR -- STATE +- STAR +- STATE - STATEMEN - STATE_WI -- STORAGE -- STREAM -- STREAMS -- STRING -- SYNCDB +- STORAGE +- STREAM +- STREAMS +- STRING +- SYNCDB ### T -- TABLE -- TABLES -- TAG -- TAGS -- TBNAME -- TIMES -- TIMESTAMP -- TINYINT -- TOPIC -- TOPICS -- TRIGGER -- TSERIES +- TABLE +- TABLES +- TAG +- TAGS +- TBNAME +- TIMES +- TIMESTAMP +- TINYINT +- TOPIC +- TOPICS +- TRIGGER +- TSERIES ### U -- UMINUS -- UNION -- UNSIGNED -- UPDATE -- UPLUS -- USE -- USER -- USERS -- USING +- UMINUS +- UNION +- UNSIGNED +- UPDATE +- UPLUS +- USE +- USER +- USERS +- USING ### V -- VALUES -- VARIABLE +- VALUES +- VARIABLE - VARIABLES -- VGROUPS -- VIEW -- VNODES +- VGROUPS +- VIEW +- VNODES ### W - WAL - WHERE -### _ - -- _C0 -- _QSTART -- _QSTOP -- _QDURATION -- _WSTART -- _WSTOP -- _WDURATION - -## Explanations -### TBNAME -`TBNAME` can be considered as a special tag, which represents the name of the subtable, in a STable. - -Get the table name and tag values of all subtables in a STable. -```mysql -SELECT TBNAME, location FROM meters; -``` - -Count the number of subtables in a STable. -```mysql -SELECT COUNT(TBNAME) FROM meters; -``` - -Only filter on TAGS can be used in WHERE clause in the above two query statements. -```mysql -taos> SELECT TBNAME, location FROM meters; - tbname | location | -================================================================== - d1004 | California.SanFrancisco | - d1003 | California.SanFrancisco | - d1002 | California.LosAngeles | - d1001 | California.LosAngeles | -Query OK, 4 row(s) in set (0.000881s) - -taos> SELECT COUNT(tbname) FROM meters WHERE groupId > 2; - count(tbname) | -======================== - 2 | -Query OK, 1 row(s) in set (0.001091s) -``` -### _QSTART/_QSTOP/_QDURATION -The start, stop and duration of a query time window. - -### _WSTART/_WSTOP/_WDURATION -The start, stop and duration of aggegate query by time window, like interval, session window, state window. - -### _c0/_ROWTS -_c0 is equal to _ROWTS, it means the first column of a table or STable. +### \_ + +- \_C0 +- \_QSTART +- \_QSTOP +- \_QDURATION +- \_WSTART +- \_WSTOP +- \_WDURATION diff --git a/docs/en/12-taos-sql/21-node.md b/docs/en/12-taos-sql/21-node.md index 4816daf42042c0607aebf37c8b57961e5b1927fe..8bb895f73cd26edd1ec2ddabf08b842ceecf76fe 100644 --- a/docs/en/12-taos-sql/21-node.md +++ b/docs/en/12-taos-sql/21-node.md @@ -1,37 +1,37 @@ --- -sidebar_label: 集群管理 -title: 集群管理 +sidebar_label: Cluster +title: Cluster --- -组成 TDengine 集群的物理实体是 dnode (data node 的缩写)，它是一个运行在操作系统之上的进程。在 dnode 中可以建立负责时序数据存储的 vnode (virtual node)，在多节点集群环境下当某个数据库的 replica 为 3 时，该数据库中的每个 vgroup 由 3 个 vnode 组成；当数据库的 replica 为 1 时，该数据库中的每个 vgroup 由 1 个 vnode 组成。如果要想配置某个数据库为多副本，则集群中的 dnode 数量至少为 3。在 dnode 还可以创建 mnode (management node)，单个集群中最多可以创建三个 mnode。在 TDengine 3.0.0.0 中为了支持存算分离，引入了一种新的逻辑节点 qnode (query node)，qnode 和 vnode 既可以共存在一个 dnode 中，也可以完全分离在不同的 dnode 上。 +The physical entities that form TDengine clusters are known as data nodes (dnodes). Each dnode is a process running on the operating system of the physical machine. Dnodes can contain virtual nodes (vnodes), which store time-series data. Virtual nodes are formed into vgroups, which have 1 or 3 vnodes depending on the replica setting. If you want to enable replication on your cluster, it must contain at least three nodes. Dnodes can also contain management nodes (mnodes). Each cluster has up to three mnodes. Finally, dnodes can contain query nodes (qnodes), which compute time-series data, thus separating compute from storage. A single dnode can contain a vnode, qnode, and mnode. -## 创建数据节点 +## Create a Dnode ```sql CREATE DNODE {dnode_endpoint | dnode_host_name PORT port_val} ``` -其中 `dnode_endpoint` 是形成 `hostname:port`的格式。也可以分开指定 hostname 和 port。 +Enter the dnode_endpoint in hostname:port format. You can also specify the hostname and port as separate parameters. -实际操作中推荐先创建 dnode，再启动相应的 dnode 进程，这样该 dnode 就可以立即根据其配置文件中的 firstEP 加入集群。每个 dnode 在加入成功后都会被分配一个 ID。 +Create the dnode before starting the corresponding dnode process. The dnode can then join the cluster based on the value of the firstEp parameter. Each dnode is assigned an ID after it joins a cluster. -## 查看数据节点 +## View Dnodes ```sql SHOW DNODES; ``` -可以列出集群中所有的数据节点，所列出的字段有 dnode 的 ID, endpoint, status。 +The preceding SQL command shows all dnodes in the cluster with the ID, endpoint, and status. -## 删除数据节点 +## Delete a DNODE ```sql DROP DNODE {dnode_id | dnode_endpoint} ``` -可以用 dnoe_id 或 endpoint 两种方式从集群中删除一个 dnode。注意删除 dnode 不等于停止相应的进程。实际中推荐先将一个 dnode 删除之后再停止其所对应的进程。 +You can delete a dnode by its ID or by its endpoint. Note that deleting a dnode does not stop its process. You must stop the process after the dnode is deleted. -## 修改数据节点配置 +## Modify Dnode Configuration ```sql ALTER DNODE dnode_id dnode_option @@ -62,59 +62,59 @@ dnode_option: { } ``` -上面语法中的这些可修改配置项其配置方式与 dnode 配置文件中的配置方式相同，区别是修改是动态的立即生效，且不需要重启 dnode。 +The parameters that you can modify through this statement are the same as those located in the dnode configuration file. Modifications that you make through this statement take effect immediately, while modifications to the configuration file take effect when the dnode restarts. -## 添加管理节点 +## Add an Mnode ```sql CREATE MNODE ON DNODE dnode_id ``` -系统启动默认在 firstEP 节点上创建一个 MNODE，用户可以使用此语句创建更多的 MNODE 来提高系统可用性。一个集群最多存在三个 MNODE，一个 DNODE 上只能创建一个 MNODE。 +TDengine automatically creates an mnode on the firstEp node. You can use this statement to create more mnodes for higher system availability. A cluster can have a maximum of three mnodes. Each dnode can contain only one mnode. -## 查看管理节点 +## View Mnodes ```sql SHOW MNODES; ``` -列出集群中所有的管理节点，包括其 ID，所在 DNODE 以及状态。 +This statement shows all mnodes in the cluster with the ID, dnode, and status. -## 删除管理节点 +## Delete an Mnode ```sql DROP MNODE ON DNODE dnode_id; ``` -删除 dnode_id 所指定的 DNODE 上的 MNODE。 +This statement deletes the mnode located on the specified dnode. -## 创建查询节点 +## Create a Qnode ```sql CREATE QNODE ON DNODE dnode_id; ``` -系统启动默认没有 QNODE，用户可以创建 QNODE 来实现计算和存储的分离。一个 DNODE 上只能创建一个 QNODE。一个 DNODE 的 `supportVnodes` 参数如果不为 0，同时又在其上创建上 QNODE，则在该 dnode 中既有负责存储管理的 vnode 又有负责查询计算的 qnode，如果还在该 dnode 上创建了 mnode，则一个 dnode 上最多三种逻辑节点都可以存在。但通过配置也可以使其彻底分离。将一个 dnode 的`supportVnodes`配置为 0，可以选择在其上创建 mnode 或者 qnode 中的一种，这样可以实现三种逻辑节点在物理上的彻底分离。 +TDengine does not automatically create qnodes on startup. You can create qnodes as necessary for compute/storage separation. Each dnode can contain only one qnode. If a qnode is created on a dnode whose supportVnodes parameter is not 0, a vnode and qnode may coexist on the dnode. Each dnode can have a maximum of one vnode, one qnode, and one mnode. However, you can configure your cluster so that vnodes, qnodes, and mnodes are located on separate dnodes. If you set supportVnodes to 0 for a dnode, you can then decide whether to deploy an mnode or a qnode on it. In this way you can physically separate virtual node types. -## 查看查询节点 +## View Qnodes ```sql SHOW QNODES; ``` -列出集群中所有查询节点，包括 ID，及所在 DNODE。 +This statement shows all qnodes in the cluster with the ID and dnode. -## 删除查询节点 +## Delete a Qnode ```sql DROP QNODE ON DNODE dnode_id; ``` -删除 ID 为 dnode_id 的 DNODE 上的 QNODE，但并不会影响该 dnode 的状态。 +This statement deletes the mnode located on the specified dnode. This does not affect the status of the dnode. -## 修改客户端配置 +## Modify Client Configuration -如果将客户端也看作广义的集群的一部分，可以通过如下命令动态修改客户端配置参数。 +The client configuration can also be modified in a similar way to other cluster components. ```sql ALTER LOCAL local_option @@ -129,26 +129,26 @@ local_option: { } ``` -上面语法中的参数与在配置文件中配置客户端的用法相同，但不需要重启客户端，修改后立即生效。 +The parameters that you can modify through this statement are the same as those located in the client configuration file. Modifications that you make through this statement take effect immediately, while modifications to the configuration file take effect when the client restarts. -## 查看客户端配置 +## View Client Configuration ```sql SHOW LOCAL VARIABLES; ``` -## 合并 vgroup +## Combine Vgroups ```sql MERGE VGROUP vgroup_no1 vgroup_no2; ``` -如果在系统实际运行一段时间后，因为不同时间线的数据特征不同导致在 vgroups 之间的数据和负载分布不均衡，可以通过合并或拆分 vgroups 的方式逐步实现负载均衡。 +If load and data are not properly balanced among vgroups due to the data in different tim lines having different characteristics, you can combine or separate vgroups. -## 拆分 vgroup +## Separate Vgroups ```sql SPLIT VGROUP vgroup_no; ``` -会创建一个新的 vgroup，并将指定 vgroup 中的数据按照一致性 HASH 迁移一部分到新的 vgroup 中。此过程中，原 vgroup 可以正常提供读写服务。 +This statement creates a new vgroup and migrates part of the data from the original vgroup to the new vgroup with consistent hashing. During this process, the original vgroup can continue to provide services normally. diff --git a/docs/en/12-taos-sql/22-meta.md b/docs/en/12-taos-sql/22-meta.md index 1e178706859a3e5fa5dbabc00777b92639d76617..796b25dcb0a425aa0ffd76a6e9b8de45ba069357 100644 --- a/docs/en/12-taos-sql/22-meta.md +++ b/docs/en/12-taos-sql/22-meta.md @@ -1,247 +1,247 @@ --- -sidebar_label: 元数据库 -title: 元数据库 +sidebar_label: Metadata +title: Information_Schema Database --- -TDengine 内置了一个名为 `INFORMATION_SCHEMA` 的数据库，提供对数据库元数据、数据库系统信息和状态的访问，例如数据库或表的名称，当前执行的 SQL 语句等。该数据库存储有关 TDengine 维护的所有其他数据库的信息。它包含多个只读表。实际上，这些表都是视图，而不是基表，因此没有与它们关联的文件。所以对这些表只能查询，不能进行 INSERT 等写入操作。`INFORMATION_SCHEMA` 数据库旨在以一种更一致的方式来提供对 TDengine 支持的各种 SHOW 语句（如 SHOW TABLES、SHOW DATABASES）所提供的信息的访问。与 SHOW 语句相比，使用 SELECT ... FROM INFORMATION_SCHEMA.tablename 具有以下优点： +TDengine includes a built-in database named `INFORMATION_SCHEMA` to provide access to database metadata, system information, and status information. This information includes database names, table names, and currently running SQL statements. All information related to TDengine maintenance is stored in this database. It contains several read-only tables. These tables are more accurately described as views, and they do not correspond to specific files. You can query these tables but cannot write data to them. The INFORMATION_SCHEMA database is intended to provide a unified method for SHOW commands to access data. However, using SELECT ... FROM INFORMATION_SCHEMA.tablename offers several advantages over SHOW commands: -1. 可以使用 USE 语句将 INFORMATION_SCHEMA 设为默认数据库 -2. 可以使用 SELECT 语句熟悉的语法，只需要学习一些表名和列名 -3. 可以对查询结果进行筛选、排序等操作。事实上，可以使用任意 TDengine 支持的 SELECT 语句对 INFORMATION_SCHEMA 中的表进行查询 -4. TDengine 在后续演进中可以灵活的添加已有 INFORMATION_SCHEMA 中表的列，而不用担心对既有业务系统造成影响 -5. 与其他数据库系统更具互操作性。例如，Oracle 数据库用户熟悉查询 Oracle 数据字典中的表 +1. You can use a USE statement to specify the INFORMATION_SCHEMA database as the current database. +2. You can use the familiar SELECT syntax to access information, provided that you know the table and column names. +3. You can filter and order the query results. More generally, you can use any SELECT syntax that TDengine supports to query the INFORMATION_SCHEMA database. +4. Future versions of TDengine can add new columns to INFORMATION_SCHEMA tables without affecting existing business systems. +5. It is easier for users coming from other database management systems. For example, Oracle users can query data dictionary tables. -Note: 由于 SHOW 语句已经被开发者熟悉和广泛使用，所以它们仍然被保留。 +Note: SHOW statements are still supported for the convenience of existing users. -本章将详细介绍 `INFORMATION_SCHEMA` 这个内置元数据库中的表和表结构。 +This document introduces the tables of INFORMATION_SCHEMA and their structure. ## INS_DNODES -提供 dnode 的相关信息。也可以使用 SHOW DNODES 来查询这些信息。 +Provides information about dnodes. Similar to SHOW DNODES. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :------------: | ------------ | ------------------------- | -| 1 | vnodes | SMALLINT | dnode 中的实际 vnode 个数 | -| 2 | support_vnodes | SMALLINT | 最多支持的 vnode 个数 | -| 3 | status | BINARY(10) | 当前状态 | -| 4 | note | BINARY(256) | 离线原因等信息 | -| 5 | id | SMALLINT | dnode id | -| 6 | endpoint | BINARY(134) | dnode 的地址 | -| 7 | create | TIMESTAMP | 创建时间 | +| 1 | vnodes | SMALLINT | Current number of vnodes on the dnode | +| 2 | vnodes | SMALLINT | Maximum number of vnodes on the dnode | +| 3 | status | BINARY(10) | Current status | +| 4 | note | BINARY(256) | Reason for going offline or other information | +| 5 | id | SMALLINT | Dnode ID | +| 6 | endpoint | BINARY(134) | Dnode endpoint | +| 7 | create | TIMESTAMP | Creation time | ## INS_MNODES -提供 mnode 的相关信息。也可以使用 SHOW MNODES 来查询这些信息。 +Provides information about mnodes. Similar to SHOW MNODES. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | ------------------ | -| 1 | id | SMALLINT | mnode id | -| 2 | endpoint | BINARY(134) | mnode 的地址 | -| 3 | role | BINARY(10) | 当前角色 | -| 4 | role_time | TIMESTAMP | 成为当前角色的时间 | -| 5 | create_time | TIMESTAMP | 创建时间 | +| 1 | id | SMALLINT | Mnode ID | +| 2 | endpoint | BINARY(134) | Mnode endpoint | +| 3 | role | BINARY(10) | Current role | +| 4 | role_time | TIMESTAMP | Time at which the current role was assumed | +| 5 | create_time | TIMESTAMP | Creation time | ## INS_MODULES -提供组件的相关信息。也可以使用 SHOW MODULES 来查询这些信息 +Provides information about modules. Similar to SHOW MODULES. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :------: | ------------ | ---------- | -| 1 | id | SMALLINT | module id | -| 2 | endpoint | BINARY(134) | 组件的地址 | -| 3 | module | BINARY(10) | 组件状态 | +| 1 | id | SMALLINT | Module ID | +| 2 | endpoint | BINARY(134) | Module endpoint | +| 3 | module | BINARY(10) | Module status | ## INS_QNODES -当前系统中 QNODE 的信息。也可以使用 SHOW QNODES 来查询这些信息。 +Provides information about qnodes. Similar to SHOW QNODES. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | ------------ | -| 1 | id | SMALLINT | qnode id | -| 2 | endpoint | BINARY(134) | qnode 的地址 | -| 3 | create_time | TIMESTAMP | 创建时间 | +| 1 | id | SMALLINT | Qnode ID | +| 2 | endpoint | BINARY(134) | Qnode endpoint | +| 3 | create_time | TIMESTAMP | Creation time | ## INS_CLUSTER -存储集群相关信息。 +Provides information about the cluster. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | ---------- | -| 1 | id | BIGINT | cluster id | -| 2 | name | BINARY(134) | 集群名称 | -| 3 | create_time | TIMESTAMP | 创建时间 | +| 1 | id | BIGINT | Cluster ID | +| 2 | name | BINARY(134) | Cluster name | +| 3 | create_time | TIMESTAMP | Creation time | ## INS_DATABASES -提供用户创建的数据库对象的相关信息。也可以使用 SHOW DATABASES 来查询这些信息。 +Provides information about user-created databases. Similar to SHOW DATABASES. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :------------------: | ---------------- | ------------------------------------------------ | -| 1 | name | BINARY(32) | 数据库名 | -| 2 | create_time | TIMESTAMP | 创建时间 | -| 3 | ntables | INT | 数据库中表的数量，包含子表和普通表但不包含超级表 | -| 4 | vgroups | INT | 数据库中有多少个 vgroup | -| 6 | replica | INT | 副本数 | -| 7 | quorum | BINARY(3) | 强一致性 | -| 8 | duration | INT | 单文件存储数据的时间跨度 | -| 9 | keep | INT | 数据保留时长 | -| 10 | buffer | INT | 每个 vnode 写缓存的内存块大小，单位 MB | -| 11 | pagesize | INT | 每个 VNODE 中元数据存储引擎的页大小，单位为 KB | -| 12 | pages | INT | 每个 vnode 元数据存储引擎的缓存页个数 | -| 13 | minrows | INT | 文件块中记录的最大条数 | -| 14 | maxrows | INT | 文件块中记录的最小条数 | -| 15 | comp | INT | 数据压缩方式 | -| 16 | precision | BINARY(2) | 时间分辨率 | -| 17 | status | BINARY(10) | 数据库状态 | -| 18 | retention | BINARY (60) | 数据的聚合周期和保存时长 | -| 19 | single_stable | BOOL | 表示此数据库中是否只可以创建一个超级表 | -| 20 | cachemodel | BINARY(60) | 表示是否在内存中缓存子表的最近数据 | -| 21 | cachesize | INT | 表示每个 vnode 中用于缓存子表最近数据的内存大小 | -| 22 | wal_level | INT | WAL 级别 | -| 23 | wal_fsync_period | INT | 数据落盘周期 | -| 24 | wal_retention_period | INT | WAL 的保存时长 | -| 25 | wal_retention_size | INT | WAL 的保存上限 | -| 26 | wal_roll_period | INT | wal 文件切换时长 | -| 27 | wal_segment_size | wal 单个文件大小 | +| 1| name| BINARY(32)| Database name | +| 2 | create_time | TIMESTAMP | Creation time | +| 3 | ntables | INT | Number of standard tables and subtables (not including supertables) | +| 4 | vgroups | INT | Number of vgroups | +| 6 | replica | INT | Number of replicas | +| 7 | quorum | BINARY(3) | Strong consistency | +| 8 | duration | INT | Duration for storage of single files | +| 9 | keep | INT | Data retention period | +| 10 | buffer | INT | Write cache size per vnode, in MB | +| 11 | pagesize | INT | Page size for vnode metadata storage engine, in KB | +| 12 | pages | INT | Number of pages per vnode metadata storage engine | +| 13 | minrows | INT | Maximum number of records per file block | +| 14 | maxrows | INT | Minimum number of records per file block | +| 15 | comp | INT | Compression method | +| 16 | precision | BINARY(2) | Time precision | +| 17 | status | BINARY(10) | Current database status | +| 18 | retention | BINARY (60) | Aggregation interval and retention period | +| 19 | single_stable | BOOL | Whether the database can contain multiple supertables | +| 20 | cachemodel | BINARY(60) | Caching method for the newest data | +| 21 | cachesize | INT | Memory per vnode used for caching the newest data | +| 22 | wal_level | INT | WAL level | +| 23 | wal_fsync_period | INT | Interval at which WAL is written to disk | +| 24 | wal_retention_period | INT | WAL retention period | +| 25 | wal_retention_size | INT | Maximum WAL size | +| 26 | wal_roll_period | INT | WAL rotation period | +| 27 | wal_segment_size | WAL file size | ## INS_FUNCTIONS -用户创建的自定义函数的信息。 +Provides information about user-defined functions. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | -------------- | -| 1 | name | BINARY(64) | 函数名 | -| 2 | comment | BINARY(255) | 补充说明 | -| 3 | aggregate | INT | 是否为聚合函数 | -| 4 | output_type | BINARY(31) | 输出类型 | -| 5 | create_time | TIMESTAMP | 创建时间 | -| 6 | code_len | INT | 代码长度 | -| 7 | bufsize | INT | buffer 大小 | +| 1 | name | BINARY(64) | Function name | +| 2 | comment | BINARY(255) | Function description | +| 3 | aggregate | INT | Whether the UDF is an aggregate function | +| 4 | output_type | BINARY(31) | Output data type | +| 5 | create_time | TIMESTAMP | Creation time | +| 6 | code_len | INT | Length of the source code | +| 7 | bufsize | INT | Buffer size | ## INS_INDEXES -提供用户创建的索引的相关信息。也可以使用 SHOW INDEX 来查询这些信息。 +Provides information about user-created indices. Similar to SHOW INDEX. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :--------------: | ------------ | ---------------------------------------------------------------------------------- | -| 1 | db_name | BINARY(32) | 包含此索引的表所在的数据库名 | -| 2 | table_name | BINARY(192) | 包含此索引的表的名称 | -| 3 | index_name | BINARY(192) | 索引名 | -| 4 | column_name | BINARY(64) | 建索引的列的列名 | -| 5 | index_type | BINARY(10) | 目前有 SMA 和 FULLTEXT | -| 6 | index_extensions | BINARY(256) | 索引的额外信息。对 SMA 类型的索引，是函数名的列表。对 FULLTEXT 类型的索引为 NULL。 | +| 1 | db_name | BINARY(32) | Database containing the table with the specified index | +| 2 | table_name | BINARY(192) | Table containing the specified index | +| 3 | index_name | BINARY(192) | Index name | +| 4 | db_name | BINARY(64) | Index column | +| 5 | index_type | BINARY(10) | SMA or FULLTEXT index | +| 6 | index_extensions | BINARY(256) | Other information For SMA indices, this shows a list of functions. For FULLTEXT indices, this is null. | ## INS_STABLES -提供用户创建的超级表的相关信息。 +Provides information about supertables. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :-----------: | ------------ | ------------------------ | -| 1 | stable_name | BINARY(192) | 超级表表名 | -| 2 | db_name | BINARY(64) | 超级表所在的数据库的名称 | -| 3 | create_time | TIMESTAMP | 创建时间 | -| 4 | columns | INT | 列数目 | -| 5 | tags | INT | 标签数目 | -| 6 | last_update | TIMESTAMP | 最后更新时间 | -| 7 | table_comment | BINARY(1024) | 表注释 | -| 8 | watermark | BINARY(64) | 窗口的关闭时间 | -| 9 | max_delay | BINARY(64) | 推送计算结果的最大延迟 | -| 10 | rollup | BINARY(128) | rollup 聚合函数 | +| 1 | stable_name | BINARY(192) | Supertable name | +| 2 | db_name | BINARY(64) | All databases in the supertable | +| 3 | create_time | TIMESTAMP | Creation time | +| 4 | columns | INT | Number of columns | +| 5 | tags | INT | Number of tags | +| 6 | last_update | TIMESTAMP | Last updated time | +| 7 | table_comment | BINARY(1024) | Table description | +| 8 | watermark | BINARY(64) | Window closing time | +| 9 | max_delay | BINARY(64) | Maximum delay for pushing stream processing results | +| 10 | rollup | BINARY(128) | Rollup aggregate function | ## INS_TABLES -提供用户创建的普通表和子表的相关信息 +Provides information about standard tables and subtables. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :-----------: | ------------ | ---------------- | -| 1 | table_name | BINARY(192) | 表名 | -| 2 | db_name | BINARY(64) | 数据库名 | -| 3 | create_time | TIMESTAMP | 创建时间 | -| 4 | columns | INT | 列数目 | -| 5 | stable_name | BINARY(192) | 所属的超级表表名 | -| 6 | uid | BIGINT | 表 id | -| 7 | vgroup_id | INT | vgroup id | -| 8 | ttl | INT | 表的生命周期 | -| 9 | table_comment | BINARY(1024) | 表注释 | -| 10 | type | BINARY(20) | 表类型 | +| 1 | table_name | BINARY(192) | Table name | +| 2 | db_name | BINARY(64) | Database name | +| 3 | create_time | TIMESTAMP | Creation time | +| 4 | columns | INT | Number of columns | +| 5 | stable_name | BINARY(192) | Supertable name | +| 6 | uid | BIGINT | Table ID | +| 7 | vgroup_id | INT | Vgroup ID | +| 8 | ttl | INT | Table time-to-live | +| 9 | table_comment | BINARY(1024) | Table description | +| 10 | type | BINARY(20) | Table type | ## INS_TAGS -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------- | ---------------------- | -| 1 | table_name | BINARY(192) | 表名 | -| 2 | db_name | BINARY(64) | 该表所在的数据库的名称 | -| 3 | stable_name | BINARY(192) | 所属的超级表表名 | -| 4 | tag_name | BINARY(64) | tag 的名称 | -| 5 | tag_type | BINARY(64) | tag 的类型 | -| 6 | tag_value | BINARY(16384) | tag 的值 | +| 1 | table_name | BINARY(192) | Table name | +| 2 | db_name | BINARY(64) | Database name | +| 3 | stable_name | BINARY(192) | Supertable name | +| 4 | tag_name | BINARY(64) | Tag name | +| 5 | tag_type | BINARY(64) | Tag type | +| 6 | tag_value | BINARY(16384) | Tag value | ## INS_USERS -提供系统中创建的用户的相关信息。 +Provides information about TDengine users. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | -------- | -| 1 | user_name | BINARY(23) | 用户名 | -| 2 | privilege | BINARY(256) | 权限 | -| 3 | create_time | TIMESTAMP | 创建时间 | +| 1 | user_name | BINARY(23) | User name | +| 2 | privilege | BINARY(256) | User permissions | +| 3 | create_time | TIMESTAMP | Creation time | ## INS_GRANTS -提供企业版授权的相关信息。 +Provides information about TDengine Enterprise Edition permissions. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :---------: | ------------ | -------------------------------------------------- | -| 1 | version | BINARY(9) | 企业版授权说明：official(官方授权的)/trial(试用的) | -| 2 | cpu_cores | BINARY(9) | 授权使用的 CPU 核心数量 | -| 3 | dnodes | BINARY(10) | 授权使用的 dnode 节点数量 | -| 4 | streams | BINARY(10) | 授权创建的流数量 | -| 5 | users | BINARY(10) | 授权创建的用户数量 | -| 6 | accounts | BINARY(10) | 授权创建的帐户数量 | -| 7 | storage | BINARY(21) | 授权使用的存储空间大小 | -| 8 | connections | BINARY(21) | 授权使用的客户端连接数量 | -| 9 | databases | BINARY(11) | 授权使用的数据库数量 | -| 10 | speed | BINARY(9) | 授权使用的数据点每秒写入数量 | -| 11 | querytime | BINARY(9) | 授权使用的查询总时长 | -| 12 | timeseries | BINARY(21) | 授权使用的测点数量 | -| 13 | expired | BINARY(5) | 是否到期，true：到期，false：未到期 | -| 14 | expire_time | BINARY(19) | 试用期到期时间 | +| 1 | version | BINARY(9) | Whether the deployment is a licensed or trial version | +| 2 | cpu_cores | BINARY(9) | CPU cores included in license | +| 3 | dnodes | BINARY(10) | Dnodes included in license | +| 4 | streams | BINARY(10) | Streams included in license | +| 5 | users | BINARY(10) | Users included in license | +| 6 | streams | BINARY(10) | Accounts included in license | +| 7 | storage | BINARY(21) | Storage space included in license | +| 8 | connections | BINARY(21) | Client connections included in license | +| 9 | databases | BINARY(11) | Databases included in license | +| 10 | speed | BINARY(9) | Write speed specified in license (data points per second) | +| 11 | querytime | BINARY(9) | Total query time specified in license | +| 12 | timeseries | BINARY(21) | Number of metrics included in license | +| 13 | expired | BINARY(5) | Whether the license has expired | +| 14 | expire_time | BINARY(19) | When the trial period expires | ## INS_VGROUPS -系统中所有 vgroups 的信息。 +Provides information about vgroups. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :-------: | ------------ | ------------------------------------------------------ | -| 1 | vgroup_id | INT | vgroup id | -| 2 | db_name | BINARY(32) | 数据库名 | -| 3 | tables | INT | 此 vgroup 内有多少表 | -| 4 | status | BINARY(10) | 此 vgroup 的状态 | -| 5 | v1_dnode | INT | 第一个成员所在的 dnode 的 id | -| 6 | v1_status | BINARY(10) | 第一个成员的状态 | -| 7 | v2_dnode | INT | 第二个成员所在的 dnode 的 id | -| 8 | v2_status | BINARY(10) | 第二个成员的状态 | -| 9 | v3_dnode | INT | 第三个成员所在的 dnode 的 id | -| 10 | v3_status | BINARY(10) | 第三个成员的状态 | -| 11 | nfiles | INT | 此 vgroup 中数据/元数据文件的数量 | -| 12 | file_size | INT | 此 vgroup 中数据/元数据文件的大小 | -| 13 | tsma | TINYINT | 此 vgroup 是否专用于 Time-range-wise SMA，1: 是, 0: 否 | +| 1 | vgroup_id | INT | Vgroup ID | +| 2 | db_name | BINARY(32) | Database name | +| 3 | tables | INT | Tables in vgroup | +| 4 | status | BINARY(10) | Vgroup status | +| 5 | v1_dnode | INT | Dnode ID of first vgroup member | +| 6 | v1_status | BINARY(10) | Status of first vgroup member | +| 7 | v2_dnode | INT | Dnode ID of second vgroup member | +| 8 | v2_status | BINARY(10) | Status of second vgroup member | +| 9 | v3_dnode | INT | Dnode ID of third vgroup member | +| 10 | v3_status | BINARY(10) | Status of third vgroup member | +| 11 | nfiles | INT | Number of data and metadata files in the vgroup | +| 12 | file_size | INT | Size of the data and metadata files in the vgroup | +| 13 | tsma | TINYINT | Whether time-range-wise SMA is enabled. 1 means enabled; 0 means disabled. | ## INS_CONFIGS -系统配置参数。 +Provides system configuration information. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :------: | ------------ | ------------ | -| 1 | name | BINARY(32) | 配置项名称 | -| 2 | value | BINARY(64) | 该配置项的值 | +| 1 | name | BINARY(32) | Parameter | +| 2 | value | BINARY(64) | Value | ## INS_DNODE_VARIABLES -系统中每个 dnode 的配置参数。 +Provides dnode configuration information. -| # | **列名** | **数据类型** | **说明** | +| # | **Column** | **Data Type** | **Description** | | --- | :------: | ------------ | ------------ | -| 1 | dnode_id | INT | dnode 的 ID | -| 2 | name | BINARY(32) | 配置项名称 | -| 3 | value | BINARY(64) | 该配置项的值 | +| 1 | dnode_id | INT | Dnode ID | +| 2 | name | BINARY(32) | Parameter | +| 3 | value | BINARY(64) | Value | diff --git a/docs/en/12-taos-sql/23-perf.md b/docs/en/12-taos-sql/23-perf.md new file mode 100644 index 0000000000000000000000000000000000000000..10a93380220d357261914066d2fe036b8470e224 --- /dev/null +++ b/docs/en/12-taos-sql/23-perf.md @@ -0,0 +1,129 @@ +--- +sidebar_label: Statistics +title: Performance_Schema Database +--- + +TDengine includes a built-in database named `PERFORMANCE_SCHEMA` to provide access to database performance statistics. This document introduces the tables of PERFORMANCE_SCHEMA and their structure. + +## PERF_APP + +Provides information about clients (such as applications) that connect to the cluster. Similar to SHOW APPS. + +| # | **Column** | **Data Type** | **Description** | +| --- | :----------: | ------------ | ------------------------------- | +| 1 | app_id | UBIGINT | Client ID | +| 2 | ip | BINARY(16) | Client IP address | +| 3 | pid | INT | Client process | +| 4 | name | BINARY(24) | Client name | +| 5 | start_time | TIMESTAMP | Time when client was started | +| 6 | insert_req | UBIGINT | Insert requests | +| 7 | insert_row | UBIGINT | Rows inserted | +| 8 | insert_time | UBIGINT | Time spent processing insert requests in microseconds | +| 9 | insert_bytes | UBIGINT | Size of data inserted in byted | +| 10 | fetch_bytes | UBIGINT | Size of query results in bytes | +| 11 | query_time | UBIGINT | Time spend processing query requests | +| 12 | slow_query | UBIGINT | Number of slow queries (greater than or equal to 3 seconds) | +| 13 | total_req | UBIGINT | Total requests | +| 14 | current_req | UBIGINT | Requests currently being processed | +| 15 | last_access | TIMESTAMP | Last update time | + +## PERF_CONNECTIONS + +Provides information about connections to the database. Similar to SHOW CONNECTIONS. + +| # | **Column** | **Data Type** | **Description** | +| --- | :---------: | ------------ | -------------------------------------------------- | +| 1 | conn_id | INT | Connection ID | +| 2 | user | BINARY(24) | User name | +| 3 | app | BINARY(24) | Client name | +| 4 | pid | UINT | Client process ID on client device that initiated the connection | +| 5 | end_point | BINARY(128) | Client endpoint | +| 6 | login_time | TIMESTAMP | Login time | +| 7 | last_access | TIMESTAMP | Last update time | + +## PERF_QUERIES + +Provides information about SQL queries currently running. Similar to SHOW QUERIES. + +| # | **Column** | **Data Type** | **Description** | +| --- | :----------: | ------------ | ---------------------------- | +| 1 | kill_id | UBIGINT | ID used to stop the query | +| 2 | query_id | INT | Query ID | +| 3 | conn_id | UINT | Connection ID | +| 4 | app | BINARY(24) | Client name | +| 5 | pid | INT | Client process ID on client device | +| 6 | user | BINARY(24) | User name | +| 7 | end_point | BINARY(16) | Client endpoint | +| 8 | create_time | TIMESTAMP | Creation time | +| 9 | exec_usec | BIGINT | Elapsed time | +| 10 | stable_query | BOOL | Whether the query is on a supertable | +| 11 | sub_num | INT | Number of subqueries | +| 12 | sub_status | BINARY(1000) | Subquery status | +| 13 | sql | BINARY(1024) | SQL statement | + +## PERF_TOPICS + +| # | **Column** | **Data Type** | **Description** | +| --- | :---------: | ------------ | ------------------------------ | +| 1 | topic_name | BINARY(192) | Topic name | +| 2 | db_name | BINARY(64) | Database for the topic | +| 3 | create_time | TIMESTAMP | Creation time | +| 4 | sql | BINARY(1024) | SQL statement used to create the topic | + +## PERF_CONSUMERS + +| # | **Column** | **Data Type** | **Description** | +| --- | :------------: | ------------ | ----------------------------------------------------------- | +| 1 | consumer_id | BIGINT | Consumer ID | +| 2 | consumer_group | BINARY(192) | Consumer group | +| 3 | client_id | BINARY(192) | Client ID (user-defined) | +| 4 | status | BINARY(20) | Consumer status | +| 5 | topics | BINARY(204) | Subscribed topic. Returns one row for each topic. | +| 6 | up_time | TIMESTAMP | Time of first connection to TDengine Server | +| 7 | subscribe_time | TIMESTAMP | Time of first subscription | +| 8 | rebalance_time | TIMESTAMP | Time of first rebalance triggering | + +## PERF_SUBSCRIPTIONS + +| # | **Column** | **Data Type** | **Description** | +| --- | :------------: | ------------ | ------------------------ | +| 1 | topic_name | BINARY(204) | Subscribed topic | +| 2 | consumer_group | BINARY(193) | Subscribed consumer group | +| 3 | vgroup_id | INT | Vgroup ID for the consumer | +| 4 | consumer_id | BIGINT | Consumer ID | + +## PERF_TRANS + +| # | **Column** | **Data Type** | **Description** | +| --- | :--------------: | ------------ | -------------------------------------------------------------- | +| 1 | id | INT | ID of the transaction currently running | +| 2 | create_time | TIMESTAMP | Creation time | +| 3 | stage | BINARY(12) | Transaction stage (redoAction, undoAction, or commit) | +| 4 | db1 | BINARY(64) | First database having a conflict with the transaction | +| 5 | db2 | BINARY(64) | Second database having a conflict with the transaction | +| 6 | failed_times | INT | Times the transaction has failed | +| 7 | last_exec_time | TIMESTAMP | Previous time the transaction was run | +| 8 | last_action_info | BINARY(511) | Reason for failure on previous run | + +## PERF_SMAS + +| # | **Column** | **Data Type** | **Description** | +| --- | :---------: | ------------ | ------------------------------------------- | +| 1 | sma_name | BINARY(192) | Time-range-wise SMA name | +| 2 | create_time | TIMESTAMP | Creation time | +| 3 | stable_name | BINARY(192) | Supertable name | +| 4 | vgroup_id | INT | Dedicated vgroup name | + +## PERF_STREAMS + +| # | **Column** | **Data Type** | **Description** | +| --- | :----------: | ------------ | --------------------------------------- | +| 1 | stream_name | BINARY(64) | Stream name | +| 2 | create_time | TIMESTAMP | Creation time | +| 3 | sql | BINARY(1024) | SQL statement used to create the stream | +| 4 | status | BIANRY(20) | Current status | +| 5 | source_db | BINARY(64) | Source database | +| 6 | target_db | BIANRY(64) | Target database | +| 7 | target_table | BINARY(192) | Target table | +| 8 | watermark | BIGINT | Watermark (see stream processing documentation) | +| 9 | trigger | INT | Method of triggering the result push (see stream processing documentation) | diff --git a/docs/en/12-taos-sql/24-show.md b/docs/en/12-taos-sql/24-show.md index 781f94324c78e7975abde33803cffdb914da020c..96503c95989b4ae2e99fa0c38181a74232e6dc23 100644 --- a/docs/en/12-taos-sql/24-show.md +++ b/docs/en/12-taos-sql/24-show.md @@ -1,9 +1,9 @@ --- -sidebar_label: SHOW 命令 -title: 使用 SHOW 命令查看系统元数据 +sidebar_label: SHOW Statement +title: SHOW Statement for Metadata --- -除了使用 `select` 语句查询 `INFORMATION_SCHEMA` 数据库中的表获得系统中的各种元数据、系统信息和状态之外，也可以用 `SHOW` 命令来实现同样的目的。 +In addition to running SELECT statements on INFORMATION_SCHEMA, you can also use SHOW to obtain system metadata, information, and status. ## SHOW ACCOUNTS @@ -11,9 +11,9 @@ title: 使用 SHOW 命令查看系统元数据 SHOW ACCOUNTS; ``` -显示当前系统中所有租户的信息。 +Shows information about tenants on the system. -注：企业版独有 +Note: TDengine Enterprise Edition only. ## SHOW APPS @@ -21,7 +21,7 @@ SHOW ACCOUNTS; SHOW APPS; ``` -显示接入集群的应用（客户端）信息。 +Shows all clients (such as applications) that connect to the cluster. ## SHOW BNODES @@ -29,7 +29,7 @@ SHOW APPS; SHOW BNODES; ``` -显示当前系统中存在的 BNODE (backup node, 即备份节点）的信息。 +Shows information about backup nodes (bnodes) in the system. ## SHOW CLUSTER @@ -37,7 +37,7 @@ SHOW BNODES; SHOW CLUSTER; ``` -显示当前集群的信息 +Shows information about the current cluster. ## SHOW CONNECTIONS @@ -45,7 +45,7 @@ SHOW CLUSTER; SHOW CONNECTIONS; ``` -显示当前系统中存在的连接的信息。 +Shows information about connections to the system. ## SHOW CONSUMERS @@ -53,7 +53,7 @@ SHOW CONNECTIONS; SHOW CONSUMERS; ``` -显示当前数据库下所有活跃的消费者的信息。 +Shows information about all active consumers in the system. ## SHOW CREATE DATABASE @@ -61,7 +61,7 @@ SHOW CONSUMERS; SHOW CREATE DATABASE db_name; ``` -显示 db_name 指定的数据库的创建语句。 +Shows the SQL statement used to create the specified database. ## SHOW CREATE STABLE @@ -69,7 +69,7 @@ SHOW CREATE DATABASE db_name; SHOW CREATE STABLE [db_name.]stb_name; ``` -显示 tb_name 指定的超级表的创建语句 +Shows the SQL statement used to create the specified supertable. ## SHOW CREATE TABLE @@ -77,7 +77,7 @@ SHOW CREATE STABLE [db_name.]stb_name; SHOW CREATE TABLE [db_name.]tb_name ``` -显示 tb_name 指定的表的创建语句。支持普通表、超级表和子表。 +Shows the SQL statement used to create the specified table. This statement can be used on supertables, standard tables, and subtables. ## SHOW DATABASES @@ -85,7 +85,7 @@ SHOW CREATE TABLE [db_name.]tb_name SHOW DATABASES; ``` -显示用户定义的所有数据库。 +Shows all user-created databases. ## SHOW DNODES @@ -93,7 +93,7 @@ SHOW DATABASES; SHOW DNODES; ``` -显示当前系统中 DNODE 的信息。 +Shows all dnodes in the system. ## SHOW FUNCTIONS @@ -101,7 +101,7 @@ SHOW DNODES; SHOW FUNCTIONS; ``` -显示用户定义的自定义函数。 +Shows all user-defined functions in the system. ## SHOW LICENSE @@ -110,9 +110,9 @@ SHOW LICENSE; SHOW GRANTS; ``` -显示企业版许可授权的信息。 +Shows information about the TDengine Enterprise Edition license. -注：企业版独有 +Note: TDengine Enterprise Edition only. ## SHOW INDEXES @@ -120,7 +120,7 @@ SHOW GRANTS; SHOW INDEXES FROM tbl_name [FROM db_name]; ``` -显示已创建的索引。 +Shows indices that have been created. ## SHOW LOCAL VARIABLES @@ -128,7 +128,7 @@ SHOW INDEXES FROM tbl_name [FROM db_name]; SHOW LOCAL VARIABLES; ``` -显示当前客户端配置参数的运行值。 +Shows the working configuration of the client. ## SHOW MNODES @@ -136,7 +136,7 @@ SHOW LOCAL VARIABLES; SHOW MNODES; ``` -显示当前系统中 MNODE 的信息。 +Shows information about mnodes in the system. ## SHOW MODULES @@ -144,7 +144,7 @@ SHOW MNODES; SHOW MODULES; ``` -显示当前系统中所安装的组件的信息。 +Shows information about modules installed in the system. ## SHOW QNODES @@ -152,7 +152,7 @@ SHOW MODULES; SHOW QNODES; ``` -显示当前系统中 QNODE （查询节点）的信息。 +Shows information about qnodes in the system. ## SHOW SCORES @@ -160,9 +160,9 @@ SHOW QNODES; SHOW SCORES; ``` -显示系统被许可授权的容量的信息。 +Shows information about the storage space allowed by the license. -注：企业版独有 +Note: TDengine Enterprise Edition only. ## SHOW SNODES @@ -170,7 +170,7 @@ SHOW SCORES; SHOW SNODES; ``` -显示当前系统中 SNODE （流计算节点）的信息。 +Shows information about stream processing nodes (snodes) in the system. ## SHOW STABLES @@ -178,7 +178,7 @@ SHOW SNODES; SHOW [db_name.]STABLES [LIKE 'pattern']; ``` -显示当前数据库下的所有超级表的信息。可以使用 LIKE 对表名进行模糊匹配。 +Shows all supertables in the current database. You can use LIKE for fuzzy matching. ## SHOW STREAMS @@ -186,7 +186,7 @@ SHOW [db_name.]STABLES [LIKE 'pattern']; SHOW STREAMS; ``` -显示当前系统内所有流计算的信息。 +Shows information about streams in the system. ## SHOW SUBSCRIPTIONS @@ -194,7 +194,7 @@ SHOW STREAMS; SHOW SUBSCRIPTIONS; ``` -显示当前数据库下的所有的订阅关系 +Shows all subscriptions in the current database. ## SHOW TABLES @@ -202,7 +202,7 @@ SHOW SUBSCRIPTIONS; SHOW [db_name.]TABLES [LIKE 'pattern']; ``` -显示当前数据库下的所有普通表和子表的信息。可以使用 LIKE 对表名进行模糊匹配。 +Shows all standard tables and subtables in the current database. You can use LIKE for fuzzy matching. ## SHOW TABLE DISTRIBUTED @@ -210,7 +210,7 @@ SHOW [db_name.]TABLES [LIKE 'pattern']; SHOW TABLE DISTRIBUTED table_name; ``` -显示表的数据分布信息。 +Shows how table data is distributed. ## SHOW TAGS @@ -218,7 +218,7 @@ SHOW TABLE DISTRIBUTED table_name; SHOW TAGS FROM child_table_name [FROM db_name]; ``` -显示子表的标签信息。 +Shows all tag information in a subtable. ## SHOW TOPICS @@ -226,7 +226,7 @@ SHOW TAGS FROM child_table_name [FROM db_name]; SHOW TOPICS; ``` -显示当前数据库下的所有主题的信息。 +Shows all topics in the current database. ## SHOW TRANSACTIONS @@ -234,7 +234,7 @@ SHOW TOPICS; SHOW TRANSACTIONS; ``` -显示当前系统中正在执行的事务的信息 +Shows all running transactions in the system. ## SHOW USERS @@ -242,7 +242,7 @@ SHOW TRANSACTIONS; SHOW USERS; ``` -显示当前系统中所有用户的信息。包括用户自定义的用户和系统默认用户。 +Shows information about users on the system. This includes user-created users and system-defined users. ## SHOW VARIABLES @@ -251,7 +251,7 @@ SHOW VARIABLES; SHOW DNODE dnode_id VARIABLES; ``` -显示当前系统中各节点需要相同的配置参数的运行值，也可以指定 DNODE 来查看其的配置参数。 +Shows the working configuration of the parameters that must be the same on each node. You can also specify a dnode to show the working configuration for that node. ## SHOW VGROUPS @@ -259,7 +259,7 @@ SHOW DNODE dnode_id VARIABLES; SHOW [db_name.]VGROUPS; ``` -显示当前系统中所有 VGROUP 或某个 db 的 VGROUPS 的信息。 +Shows information about all vgroups in the system or about the vgroups for a specified database. ## SHOW VNODES @@ -267,4 +267,4 @@ SHOW [db_name.]VGROUPS; SHOW VNODES [dnode_name]; ``` -显示当前系统中所有 VNODE 或某个 DNODE 的 VNODE 的信息。 +Shows information about all vnodes in the system or about the vnodes for a specified dnode. diff --git a/docs/en/12-taos-sql/25-grant.md b/docs/en/12-taos-sql/25-grant.md index 0c290350cc155e975e5a817c991bebc74944cd04..37438ee780cac17b463e0dbb1b5385d0f3965de7 100644 --- a/docs/en/12-taos-sql/25-grant.md +++ b/docs/en/12-taos-sql/25-grant.md @@ -1,29 +1,29 @@ --- -sidebar_label: 权限管理 -title: 权限管理 +sidebar_label: Permissions Management +title: Permissions Management --- -本节讲述如何在 TDengine 中进行权限管理的相关操作。 +This document describes how to manage permissions in TDengine. -## 创建用户 +## Create a User ```sql -CREATE USER use_name PASS password; +CREATE USER use_name PASS 'password'; ``` -创建用户。 +This statement creates a user account. -use_name最长为23字节。 +The maximum length of use_name is 23 bytes. -password最长为128字节，合法字符包括"a-zA-Z0-9!?$%^&*()_–+={[}]:;@~#|<,>.?/"，不可以出现单双引号、撇号、反斜杠和空格，且不可以为空。 +The maximum length of password is 128 bytes. The password can include leters, digits, and special characters excluding single quotation marks, double quotation marks, backticks, backslashes, and spaces. The password cannot be empty. -## 删除用户 +## Delete a User ```sql DROP USER user_name; ``` -## 修改用户信息 +## Modify User Information ```sql ALTER USER user_name alter_user_clause @@ -35,12 +35,12 @@ alter_user_clause: { } ``` -- PASS：修改用户密码。 -- ENABLE：修改用户是否启用。1表示启用此用户，0表示禁用此用户。 -- SYSINFO：修改用户是否可查看系统信息。1表示可以查看系统信息，0表示不可以查看系统信息。 +- PASS: Modify the user password. +- ENABLE: Specify whether the user is enabled or disabled. 1 indicates enabled and 0 indicates disabled. +- SYSINFO: Specify whether the user can query system information. 1 indicates that the user can query system information and 0 indicates that the user cannot query system information. -## 授权 +## Grant Permissions ```sql GRANT privileges ON priv_level TO user_name @@ -61,15 +61,15 @@ priv_level : { } ``` -对用户授权。 +Grant permissions to a user. -授权级别支持到DATABASE，权限有READ和WRITE两种。 +Permissions are granted on the database level. You can grant read or write permissions. -TDengine 有超级用户和普通用户两类用户。超级用户缺省创建为root，拥有所有权限。使用超级用户创建出来的用户为普通用户。在未授权的情况下，普通用户可以创建DATABASE，并拥有自己创建的DATABASE的所有权限，包括删除数据库、修改数据库、查询时序数据和写入时序数据。超级用户可以给普通用户授予其他DATABASE的读写权限，使其可以在此DATABASE上读写数据，但不能对其进行删除和修改数据库的操作。 +TDengine has superusers and standard users. The default superuser name is root. This account has all permissions. You can use the superuser account to create standard users. With no permissions, standard users can create databases and have permissions on the databases that they create. These include deleting, modifying, querying, and writing to their own databases. Superusers can grant users permission to read and write other databases. However, standard users cannot delete or modify databases created by other users. -对于非DATABASE的对象，如USER、DNODE、UDF、QNODE等，普通用户只有读权限（一般为SHOW命令），不能创建和修改。 +For non-database objects such as users, dnodes, and user-defined functions, standard users have read permissions only, generally by means of the SHOW statement. Standard users cannot create or modify these objects. -## 撤销授权 +## Revoke Permissions ```sql REVOKE privileges ON priv_level FROM user_name @@ -91,4 +91,4 @@ priv_level : { ``` -收回对用户的授权。 \ No newline at end of file +Revoke permissions from a user. diff --git a/docs/en/12-taos-sql/26-udf.md b/docs/en/12-taos-sql/26-udf.md index bd8d61a5844241efae9eee99a73c65afd3d0926f..e6199e8b315c2311be509a3eb819f33ac9a8b8bc 100644 --- a/docs/en/12-taos-sql/26-udf.md +++ b/docs/en/12-taos-sql/26-udf.md @@ -1,28 +1,68 @@ --- -sidebar_label: 自定义函数 -title: 用户自定义函数 +sidebar_label: User-Defined Functions +title: User-Defined Functions (UDF) --- -除了 TDengine 的内置函数以外，用户还可以编写自己的函数逻辑并加入TDengine系统中。 +You can create user-defined functions and import them into TDengine. +## Create UDF -## 创建函数 +SQL command can be executed on the host where the generated UDF DLL resides to load the UDF DLL into TDengine. This operation cannot be done through REST interface or web console. Once created, any client of the current TDengine can use these UDF functions in their SQL commands. UDF are stored in the management node of TDengine. The UDFs loaded in TDengine would be still available after TDengine is restarted. +When creating UDF, the type of UDF, i.e. a scalar function or aggregate function must be specified. If the specified type is wrong, the SQL statements using the function would fail with errors. The input data type and output data type must be consistent with the UDF definition. + +- Create Scalar Function ```sql -CREATE [AGGREGATE] FUNCTION func_name AS library_path OUTPUTTYPE type_name [BUFSIZE value] +CREATE FUNCTION function_name AS library_path OUTPUTTYPE output_type; ``` -语法说明： + - function_name: The scalar function name to be used in SQL statement which must be consistent with the UDF name and is also the name of the compiled DLL (.so file). + - library_path: The absolute path of the DLL file including the name of the shared object file (.so). The path must be quoted with single or double quotes. + - output_type: The data type of the results of the UDF. + + For example, the following SQL statement can be used to create a UDF from `libbitand.so`. + + ```sql + CREATE FUNCTION bit_and AS "/home/taos/udf_example/libbitand.so" OUTPUTTYPE INT; + ``` + +- Create Aggregate Function +```sql +CREATE AGGREGATE FUNCTION function_name AS library_path OUTPUTTYPE output_type [ BUFSIZE buffer_size ]; +``` + + - function_name: The aggregate function name to be used in SQL statement which must be consistent with the udfNormalFunc name and is also the name of the compiled DLL (.so file). + - library_path: The absolute path of the DLL file including the name of the shared object file (.so). The path must be quoted with single or double quotes. + - output_type: The output data type, the value is the literal string of the supported TDengine data type. + - buffer_size: The size of the intermediate buffer in bytes. This parameter is optional. + + For example, the following SQL statement can be used to create a UDF from `libl2norm.so`. + + ```sql + CREATE AGGREGATE FUNCTION l2norm AS "/home/taos/udf_example/libl2norm.so" OUTPUTTYPE DOUBLE bufsize 8; + ``` +For more information about user-defined functions, see [User-Defined Functions](../../develop/udf). -AGGREGATE：标识此函数是标量函数还是聚集函数。 -func_name：函数名，必须与函数实现中udfNormalFunc的实际名称一致。 -library_path：包含UDF函数实现的动态链接库的绝对路径，是在客户端侧主机上的绝对路径。 -OUTPUTTYPE：标识此函数的返回类型。 -BUFSIZE：中间结果的缓冲区大小，单位是字节。不设置则默认为0。最大不可超过512字节。 +## Manage UDF -关于如何开发自定义函数，请参考 [UDF使用说明](../../develop/udf)。 +- The following statement deleted the specified user-defined function. +``` +DROP FUNCTION function_name; +``` -## 删除自定义函数 +- function_name: The value of function_name in the CREATE statement used to import the UDF for example `bit_and` or `l2norm`. +```sql +DROP FUNCTION bit_and; +``` +- Show Available UDF +```sql +SHOW FUNCTIONS; +``` + +## Call UDF +The function name specified when creating UDF can be used directly in SQL statements, just like builtin functions. For example: ```sql -DROP FUNCTION func_name -``` \ No newline at end of file +SELECT X(c1,c2) FROM table/stable; +``` + +The above SQL statement invokes function X for column c1 and c2. You can use query keywords like WHERE with user-defined functions. diff --git a/docs/en/12-taos-sql/27-index.md b/docs/en/12-taos-sql/27-index.md index 2c0907723e76f304566e6a19bdef2d63225f903f..7d09bc43ab06932b82019923d4a8fda48cd99c97 100644 --- a/docs/en/12-taos-sql/27-index.md +++ b/docs/en/12-taos-sql/27-index.md @@ -1,11 +1,11 @@ --- -sidebar_label: 索引 -title: 使用索引 +sidebar_label: Index +title: Using Indices --- -TDengine 从 3.0.0.0 版本开始引入了索引功能，支持 SMA 索引和 FULLTEXT 索引。 +TDengine supports SMA and FULLTEXT indexing. -## 创建索引 +## Create an Index ```sql CREATE FULLTEXT INDEX index_name ON tb_name (col_name [, col_name] ...) @@ -19,29 +19,29 @@ functions: function [, function] ... ``` -### SMA 索引 +### SMA Indexing -对指定列按 INTERVAL 子句定义的时间窗口创建进行预聚合计算，预聚合计算类型由 functions_string 指定。SMA 索引能提升指定时间段的聚合查询的性能。目前，限制一个超级表只能创建一个 SMA INDEX。 +Performs pre-aggregation on the specified column over the time window defined by the INTERVAL clause. The type is specified in functions_string. SMA indexing improves aggregate query performance for the specified time period. One supertable can only contain one SMA index. -- 支持的函数包括 MAX、MIN 和 SUM。 -- WATERMARK: 最小单位毫秒，取值范围 [0ms, 900000ms]，默认值为 5 秒，只可用于超级表。 -- MAX_DELAY: 最小单位毫秒，取值范围 [1ms, 900000ms]，默认值为 interval 的值(但不能超过最大值)，只可用于超级表。注：不建议 MAX_DELAY 设置太小，否则会过于频繁的推送结果，影响存储和查询性能，如无特殊需求，取默认值即可。 +- The max, min, and sum functions are supported. +- WATERMARK: Enter a value between 0ms and 900000ms. The most precise unit supported is milliseconds. The default value is 5 seconds. This option can be used only on supertables. +- MAX_DELAY: Enter a value between 1ms and 900000ms. The most precise unit supported is milliseconds. The default value is the value of interval provided that it does not exceed 900000ms. This option can be used only on supertables. Note: Retain the default value if possible. Configuring a small MAX_DELAY may cause results to be frequently pushed, affecting storage and query performance. -### FULLTEXT 索引 +### FULLTEXT Indexing -对指定列建立文本索引，可以提升含有文本过滤的查询的性能。FULLTEXT 索引不支持 index_option 语法。现阶段只支持对 JSON 类型的标签列创建 FULLTEXT 索引。不支持多列联合索引，但可以为每个列分布创建 FULLTEXT 索引。 +Creates a text index for the specified column. FULLTEXT indexing improves performance for queries with text filtering. The index_option syntax is not supported for FULLTEXT indexing. FULLTEXT indexing is supported for JSON tag columns only. Multiple columns cannot be indexed together. However, separate indices can be created for each column. -## 删除索引 +## Delete an Index ```sql DROP INDEX index_name; ``` -## 查看索引 +## View Indices ````sql ```sql SHOW INDEXES FROM tbl_name [FROM db_name]; ```` -显示在所指定的数据库或表上已创建的索引。 +Shows indices that have been created for the specified database or table. diff --git a/docs/en/12-taos-sql/28-recovery.md b/docs/en/12-taos-sql/28-recovery.md index 72b220b8ff44917831ac16301237702c991b9b15..14ac14f8673fba05fee09317de927df00effed0f 100644 --- a/docs/en/12-taos-sql/28-recovery.md +++ b/docs/en/12-taos-sql/28-recovery.md @@ -1,38 +1,38 @@ --- -sidebar_label: 异常恢复 -title: 异常恢复 +sidebar_label: Error Recovery +title: Error Recovery --- -在一个复杂的应用场景中，连接和查询任务等有可能进入一种错误状态或者耗时过长迟迟无法结束，此时需要有能够终止这些连接或任务的方法。 +In a complex environment, connections and query tasks may encounter errors or fail to return in a reasonable time. If this occurs, you can terminate the connection or task. -## 终止连接 +## Terminate a Connection ```sql KILL CONNECTION conn_id; ``` -conn_id 可以通过 `SHOW CONNECTIONS` 获取。 +You can use the SHOW CONNECTIONS statement to find the conn_id. -## 终止查询 +## Terminate a Query ```sql SHOW QUERY query_id; ``` -query_id 可以通过 `SHOW QUERIES` 获取。 +You can use the SHOW QUERIES statement to find the query_id. -## 终止事务 +## Terminate a Transaction ```sql KILL TRANSACTION trans_id ``` -trans_id 可以通过 `SHOW TRANSACTIONS` 获取。 +You can use the SHOW TRANSACTIONS statement to find the trans_id. -## 重置客户端缓存 +## Reset Client Cache ```sql RESET QUERY CACHE; ``` -如果在多客户端情况下出现元数据不同步的情况，可以用这条命令强制清空客户端缓存，随后客户端会从服务端拉取最新的元数据。 +If metadata becomes desynchronized among multiple clients, you can use this command to clear the client-side cache. Clients then obtain the latest metadata from the server. diff --git a/docs/en/12-taos-sql/29-changes.md b/docs/en/12-taos-sql/29-changes.md new file mode 100644 index 0000000000000000000000000000000000000000..8532eeac5d599ca2739393c9e38eec52631e407a --- /dev/null +++ b/docs/en/12-taos-sql/29-changes.md @@ -0,0 +1,95 @@ +--- +sidebar_label: Changes in TDengine 3.0 +title: Changes in TDengine 3.0 +description: "This document explains how TDengine SQL has changed in version 3.0." +--- + +## Basic SQL Elements + +| # | **Element** | **
Change
** | **Description** | +| - | :------- | :-------- | :------- | +| 1 | VARCHAR | Added | Alias of BINARY. +| 2 | TIMESTAMP literal | Added | TIMESTAMP 'timestamp format' syntax now supported. +| 3 | _ROWTS pseudocolumn | Added | Indicates the primary key. Alias of _C0. +| 4 | INFORMATION_SCHEMA | Added | Database for system metadata containing all schema definitions +| 5 | PERFORMANCE_SCHEMA | Added | Database for system performance information. +| 6 | Connection queries | Deprecated | Connection queries are no longer supported. The syntax and interfaces are deprecated. +| 7 | Mixed operations | Enhanced | Mixing scalar and vector operations in queries has been enhanced and is supported in all SELECT clauses. +| 8 | Tag operations | Added | Tag columns can be used in queries and clauses like data columns. +| 9 | Timeline clauses and time functions in supertables | Enhanced | When PARTITION BY is not used, data in supertables is merged into a single timeline. + +## SQL Syntax + +The following data types can be used in the schema for standard tables. + +| # | **Statement** | **
Change
** | **Description** | +| - | :------- | :-------- | :------- | +| 1 | ALTER ACCOUNT | Deprecated| This Enterprise Edition-only statement has been removed. It returns the error "This statement is no longer supported." +| 2 | ALTER ALL DNODES | Added | Modifies the configuration of all dnodes. +| 3 | ALTER DATABASE | Modified | Deprecated
  • QUORUM: Specified the required number of confirmations. STRICT is now used to specify strong or weak consistency. The STRICT parameter cannot be modified.
  • BLOCKS: Specified the memory blocks used by each vnode. BUFFER is now used to specify the size of the write cache pool for each vnode.
  • UPDATE: Specified whether update operations were supported. All databases now support updating data in certain columns.
  • CACHELAST: Specified how to cache the newest row of data. CACHEMODEL now replaces CACHELAST.
  • COMP: Cannot be modified.
    Added
  • CACHEMODEL: Specifies whether to cache the latest subtable data.
  • CACHESIZE: Specifies the size of the cache for the newest subtable data.
  • WAL_FSYNC_PERIOD: Replaces the FSYNC parameter.
  • WAL_LEVEL: Replaces the WAL parameter.
    Modified
  • REPLICA: Cannot be modified.
  • KEEP: Now supports units.
+| 4 | ALTER STABLE | Modified | Deprecated
  • CHANGE TAG: Modified the name of a tag. Replaced by RENAME TAG.
    Added
  • RENAME TAG: Replaces CHANGE TAG.
  • COMMENT: Specifies comments for a supertable.
+| 5 | ALTER TABLE | Modified | Deprecated
  • CHANGE TAG: Modified the name of a tag. Replaced by RENAME TAG.
    Added
  • RENAME TAG: Replaces CHANGE TAG.
  • COMMENT: Specifies comments for a standard table.
  • TTL: Specifies the time-to-live for a standard table.
+| 6 | ALTER USER | Modified | Deprecated
  • PRIVILEGE: Specified user permissions. Replaced by GRANT and REVOKE.
    Added
  • ENABLE: Enables or disables a user.
  • SYSINFO: Specifies whether a user can query system information.
+| 7 | COMPACT VNODES | Not supported | Compacted the data on a vnode. Not supported. +| 8 | CREATE ACCOUNT | Deprecated| This Enterprise Edition-only statement has been removed. It returns the error "This statement is no longer supported." +| 9 | CREATE DATABASE | Modified | Deprecated
  • BLOCKS: Specified the number of blocks for each vnode. BUFFER is now used to specify the size of the write cache pool for each vnode.
  • CACHE: Specified the size of the memory blocks used by each vnode. BUFFER is now used to specify the size of the write cache pool for each vnode.
  • CACHELAST: Specified how to cache the newest row of data. CACHEMODEL now replaces CACHELAST.
  • DAYS: The length of time to store in a single file. Replaced by DURATION.
  • FSYNC: Specified the fsync interval when WAL was set to 2. Replaced by WAL_FSYNC_PERIOD.
  • QUORUM: Specified the number of confirmations required. STRICT is now used to specify strong or weak consistency.
  • UPDATE: Specified whether update operations were supported. All databases now support updating data in certain columns.
  • WAL: Specified the WAL level. Replaced by WAL_LEVEL.
    Added
  • BUFFER: Specifies the size of the write cache pool for each vnode.
  • CACHEMODEL: Specifies whether to cache the latest subtable data.
  • CACHESIZE: Specifies the size of the cache for the newest subtable data.
  • DURATION: Replaces DAYS. Now supports units.
  • PAGES: Specifies the number of pages in the metadata storage engine cache on each vnode.
  • PAGESIZE: specifies the size (in KB) of each page in the metadata storage engine cache on each vnode.
  • RETENTIONS: Specifies the aggregation interval and retention period
  • STRICT: Specifies whether strong data consistency is enabled.
  • SINGLE_STABLE: Specifies whether a database can contain multiple supertables.
  • VGROUPS: Specifies the initial number of vgroups when a database is created.
  • WAL_FSYNC_PERIOD: Replaces the FSYNC parameter.
  • WAL_LEVEL: Replaces the WAL parameter.
  • WAL_RETENTION_PERIOD: specifies the time after which WAL files are deleted. This parameter is used for data subscription.
  • WAL_RETENTION_SIZE: specifies the size at which WAL files are deleted. This parameter is used for data subscription.
  • WAL_ROLL_PERIOD: Specifies the WAL rotation period.
  • WAL_SEGMENT_SIZE: specifies the maximum size of a WAL file.
    Modified
  • KEEP: Now supports units.
+| 10 | CREATE DNODE | Modified | Now supports specifying hostname and port separately
  • CREATE DNODE dnode_host_name PORT port_val
+| 11 | CREATE INDEX | Added | Creates an SMA index. +| 12 | CREATE MNODE | Added | Creates an mnode. +| 13 | CREATE QNODE | Added | Creates a qnode. +| 14 | CREATE STABLE | Modified | New parameter added
  • COMMENT: Specifies comments for the supertable.
    • +| 15 | CREATE STREAM | Added | Creates a stream. +| 16 | CREATE TABLE | Modified | New parameters added
    • COMMENT: Specifies comments for the table
    • WATERMARK: Specifies the window closing time.
    • MAX_DELAY: Specifies the maximum delay for pushing stream processing results.
    • ROLLUP: Specifies aggregate functions to roll up. Rolling up a function provides downsampled results based on multiple axes.
    • SMA: Provides user-defined precomputation of aggregates based on data blocks.
    • TTL: Specifies the time-to-live for a standard table.
    +| 17 | CREATE TOPIC | Added | Creates a topic. +| 18 | DROP ACCOUNT | Deprecated| This Enterprise Edition-only statement has been removed. It returns the error "This statement is no longer supported." +| 19 | DROP CONSUMER GROUP | Added | Deletes a consumer group. +| 20 | DROP INDEX | Added | Deletes an index. +| 21 | DROP MNODE | Added | Creates an mnode. +| 22 | DROP QNODE | Added | Creates a qnode. +| 23 | DROP STREAM | Added | Deletes a stream. +| 24 | DROP TABLE | Modified | Added batch deletion syntax. +| 25 | DROP TOPIC | Added | Deletes a topic. +| 26 | EXPLAIN | Added | Query the execution plan of a query statement. +| 27 | GRANT | Added | Grants permissions to a user. +| 28 | KILL TRANSACTION | Added | Terminates an mnode transaction. +| 29 | KILL STREAM | Deprecated | Terminated a continuous query. The continuous query feature has been replaced with the stream processing feature. +| 30 | MERGE VGROUP | Added | Merges vgroups. +| 31 | REVOKE | Added | Revokes permissions from a user. +| 32 | SELECT | Modified |
    • SELECT does not use the implicit results column. Output columns must be specified in the SELECT clause.
    • DISTINCT support is enhanced. In previous versions, DISTINCT only worked on the tag column and could not be used with JOIN or GROUP BY.
    • JOIN support is enhanced. The following are now supported after JOIN: a WHERE clause with OR, operations on multiple tables, and GROUP BY on multiple tables.
    • Subqueries after FROM are enhanced. Levels of nesting are no longer restricted. Subqueries can be used with UNION ALL. Other syntax restrictions are eliminated.
    • All scalar functions can be used after WHERE.
    • GROUP BY is enhanced. You can group by any scalar expression or combination thereof.
    • SESSION can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline.
    • STATE_WINDOW can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline.
    • ORDER BY is enhanced. It is no longer required to use ORDER BY and GROUP BY together. There is no longer a restriction on the number of order expressions. NULLS FIRST and NULLS LAST syntax has been added. Any expression that conforms to the ORDER BY semantics can be used.
    • Added PARTITION BY syntax. PARTITION BY replaces GROUP BY tags.
    +| 33 | SHOW ACCOUNTS | Deprecated | This Enterprise Edition-only statement has been removed. It returns the error "This statement is no longer supported." +| 34 | SHOW APPS | Added | Shows all clients (such as applications) that connect to the cluster. +| 35 | SHOW CONSUMERS | Added | Shows information about all active consumers in the system. +| 36 | SHOW DATABASES | Modified | Only shows database names. +| 37 | SHOW FUNCTIONS | Modified | Only shows UDF names. +| 38 | SHOW LICENCE | Added | Alias of SHOW GRANTS. +| 39 | SHOW INDEXES | Added | Shows indices that have been created. +| 40 | SHOW LOCAL VARIABLES | Added | Shows the working configuration of the client. +| 41 | SHOW MODULES | Deprecated | Shows information about modules installed in the system. +| 42 | SHOW QNODES | Added | Shows information about qnodes in the system. +| 43 | SHOW STABLES | Modified | Only shows supertable names. +| 44 | SHOW STREAMS | Modified | This statement previously showed continuous queries. The continuous query feature has been replaced with the stream processing feature. This statement now shows streams that have been created. +| 45 | SHOW SUBSCRIPTIONS | Added | Shows all subscriptions in the current database. +| 46 | SHOW TABLES | Modified | Only shows table names. +| 47 | SHOW TABLE DISTRIBUTED | Added | Shows how table data is distributed. This replaces the `SELECT _block_dist() FROM { tb_name | stb_name }` command. +| 48 | SHOW TOPICS | Added | Shows all subscribed topics in the current database. +| 49 | SHOW TRANSACTIONS | Added | Shows all running transactions in the system. +| 50 | SHOW DNODE VARIABLES | Added | Shows the configuration of the specified dnode. +| 51 | SHOW VNODES | Not supported | Shows information about vnodes in the system. Not supported. +| 52 | SPLIT VGROUP | Added | Splits a vgroup into two vgroups. +| 53 | TRIM DATABASE | Added | Deletes data that has expired and orders the remaining data in accordance with the storage configuration. + +## SQL Functions + +| # | **Function** | **
    Change
    ** | **Description** | +| - | :------- | :-------- | :------- | +| 1 | TWA | Added | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 2 | IRATE | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 3 | LEASTSQUARES | Enhanced | Can be used on supertables. +| 4 | ELAPSED | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 5 | DIFF | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 6 | DERIVATIVE | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 7 | CSUM | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 8 | MAVG | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 9 | SAMPLE | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 10 | STATECOUNT | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. +| 11 | STATEDURATION | Enhanced | Can be used on supertables. When PARTITION BY is not used, data in supertables is merged into a single timeline. diff --git a/docs/en/12-taos-sql/index.md b/docs/en/12-taos-sql/index.md index 33656338a7bba38dc55cf536bdba8e95309c5acf..e243cd23186a6b9286d3297e467567c26c316112 100644 --- a/docs/en/12-taos-sql/index.md +++ b/docs/en/12-taos-sql/index.md @@ -3,20 +3,21 @@ title: TDengine SQL description: "The syntax supported by TDengine SQL " --- -This section explains the syntax of SQL to perform operations on databases, tables and STables, insert data, select data and use functions. We also provide some tips that can be used in TDengine SQL. If you have previous experience with SQL this section will be fairly easy to understand. If you do not have previous experience with SQL, you'll come to appreciate the simplicity and power of SQL. +This section explains the syntax of SQL to perform operations on databases, tables and STables, insert data, select data and use functions. We also provide some tips that can be used in TDengine SQL. If you have previous experience with SQL this section will be fairly easy to understand. If you do not have previous experience with SQL, you'll come to appreciate the simplicity and power of SQL. TDengine SQL has been enhanced in version 3.0, and the query engine has been rearchitected. For information about how TDengine SQL has changed, see [Changes in TDengine 3.0](../taos-sql/changes). -TDengine SQL is the major interface for users to write data into or query from TDengine. For ease of use, the syntax is similar to that of standard SQL. However, please note that TDengine SQL is not standard SQL. For instance, TDengine doesn't provide a delete function for time series data and so corresponding statements are not provided in TDengine SQL. +TDengine SQL is the major interface for users to write data into or query from TDengine. It uses standard SQL syntax and includes extensions and optimizations for time-series data and services. The maximum length of a TDengine SQL statement is 1 MB. Note that keyword abbreviations are not supported. For example, DELETE cannot be entered as DEL. Syntax Specifications used in this chapter: -- The content inside <\> needs to be input by the user, excluding <\> itself. +- Keywords are given in uppercase, although SQL is not case-sensitive. +- Information that you input is given in lowercase. - \[ \] means optional input, excluding [] itself. - | means one of a few options, excluding | itself. - … means the item prior to it can be repeated multiple times. To better demonstrate the syntax, usage and rules of TAOS SQL, hereinafter it's assumed that there is a data set of data from electric meters. Each meter collects 3 data measurements: current, voltage, phase. The data model is shown below: -```sql +``` taos> DESCRIBE meters; Field | Type | Length | Note | ================================================================================= @@ -29,3 +30,10 @@ taos> DESCRIBE meters; ``` The data set includes the data collected by 4 meters, the corresponding table name is d1001, d1002, d1003 and d1004 based on the data model of TDengine. + +```mdx-code-block +import DocCardList from '@theme/DocCardList'; +import {useCurrentSidebarCategory} from '@docusaurus/theme-common'; + + +``` diff --git a/docs/en/14-reference/02-rest-api/02-rest-api.mdx b/docs/en/14-reference/02-rest-api/02-rest-api.mdx index fe18349a6dae3ad44772b4a30a2c3d4ad75b0f47..8d4186a36bb983e688ae2824f13c71f4461bebf2 100644 --- a/docs/en/14-reference/02-rest-api/02-rest-api.mdx +++ b/docs/en/14-reference/02-rest-api/02-rest-api.mdx @@ -2,10 +2,10 @@ title: REST API --- -To support the development of various types of applications and platforms, TDengine provides an API that conforms to REST principles; namely REST API. To minimize the learning cost, unlike REST APIs for other database engines, TDengine allows insertion of SQL commands in the BODY of an HTTP POST request, to operate the database. +To support the development of various types of applications and platforms, TDengine provides an API that conforms to REST principles; namely REST API. To minimize the learning cost, unlike REST APIs for other database engines, TDengine allows insertion of SQL commands in the BODY of an HTTP POST request, to operate the database. -:::note -One difference from the native connector is that the REST interface is stateless and so the `USE db_name` command has no effect. All references to table names and super table names need to specify the database name in the prefix. (Since version 2.2.0.0, TDengine supports specification of the db_name in RESTful URL. If the database name prefix is not specified in the SQL command, the `db_name` specified in the URL will be used. Since version 2.4.0.0, REST service is provided by taosAdapter by default and it requires that the `db_name` must be specified in the URL.) +:::note +One difference from the native connector is that the REST interface is stateless and so the `USE db_name` command has no effect. All references to table names and super table names need to specify the database name in the prefix. TDengine supports specification of the db_name in RESTful URL. If the database name prefix is not specified in the SQL command, the `db_name` specified in the URL will be used. ::: ## Installation @@ -20,84 +20,75 @@ The following example is in an Ubuntu environment and uses the `curl` tool to ve The following example lists all databases on the host h1.taosdata.com. To use it in your environment, replace `h1.taosdata.com` and `6041` (the default port) with the actual running TDengine service FQDN and port number. -```html -curl -L -H "Authorization: Basic cm9vdDp0YW9zZGF0YQ==" -d "show databases;" h1.taosdata.com:6041/rest/sql +```bash +curl -L -H "Authorization: Basic cm9vdDp0YW9zZGF0YQ==" \ + -d "select name, ntables, status from information_schema.ins_databases;" \ + h1.taosdata.com:6041/rest/sql ``` The following return value results indicate that the verification passed. ```json { - "status": "succ", - "head": [ - "name", - "created_time", - "ntables", - "vgroups", - "replica", - "quorum", - "days", - "keep1,keep2,keep(D)", - "cache(MB)", - "blocks", - "minrows", - "maxrows", - "wallevel", - "fsync", - "comp", - "precision", - "status" - ], - "data": [ - [ - "log", - "2020-09-02 17:23:00.039", - 4, - 1, - 1, - 1, - 10, - "30,30,30", - 1, - 3, - 100, - 4096, - 1, - 3000, - 2, - "us", - "ready" - ] - ], - "rows": 1 + "code": 0, + "column_meta": [ + [ + "name", + "VARCHAR", + 64 + ], + [ + "ntables", + "BIGINT", + 8 + ], + [ + "status", + "VARCHAR", + 10 + ] + ], + "data": [ + [ + "information_schema", + 16, + "ready" + ], + [ + "performance_schema", + 9, + "ready" + ] + ], + "rows": 2 } ``` ## HTTP request URL format -``` +```text http://:/rest/sql/[db_name] ``` Parameter Description: -- fqnd: FQDN or IP address of any host in the cluster -- port: httpPort configuration item in the configuration file, default is 6041 -- db_name: Optional parameter that specifies the default database name for the executed SQL command. (supported since version 2.2.0.0) +- fqnd: FQDN or IP address of any host in the cluster. +- port: httpPort configuration item in the configuration file, default is 6041. +- db_name: Optional parameter that specifies the default database name for the executed SQL command. For example, `http://h1.taos.com:6041/rest/sql/test` is a URL to `h1.taos.com:6041` and sets the default database name to `test`. TDengine supports both Basic authentication and custom authentication mechanisms, and subsequent versions will provide a standard secure digital signature mechanism for authentication. -- The custom authentication information is as follows. More details about "token" later. +- authentication information is shown below： - ``` + ```text Authorization: Taosd ``` -- Basic authentication information is shown below +- Basic authentication information is shown below： - ``` + ```text Authorization: Basic ``` @@ -109,51 +100,148 @@ Use `curl` to initiate an HTTP request with a custom authentication method, with curl -L -H "Authorization: Basic " -d "" :/rest/sql/[db_name] ``` -Or +or ```bash curl -L -u username:password -d "" :/rest/sql/[db_name] ``` -where `TOKEN` is the string after Base64 encoding of `{username}:{password}`, e.g. `root:taosdata` is encoded as `cm9vdDp0YW9zZGF0YQ==`. +where `TOKEN` is the string after Base64 encoding of `{username}:{password}`, e.g. `root:taosdata` is encoded as `cm9vdDp0YW9zZGF0YQ==`.. ## HTTP Return Format -The return result is in JSON format, as follows: +### HTTP Response Code + +| **Response Code** | **Description** | +|-------------------|----------------| +| 200 | Success. (Also used for C interface errors.) | +| 400 | Parameter error | +| 401 | Authentication failure | +| 404 | Interface not found | +| 500 | Internal error | +| 503 | Insufficient system resources | + +### HTTP body structure + +#### Successful Operation + +Example: ```json { - "status": "succ", - "head": ["ts", "current", ...], - "column_meta": [["ts",9,8],["current",6,4], ...], - "data": [ - ["2018-10-03 14:38:05.000", 10.3, ...], - ["2018-10-03 14:38:15.000", 12.6, ...] + "code": 0, + "column_meta": [["affected_rows", "INT", 4]], + "data": [[0]], + "rows": 1 +} +``` + +Description: + +- code: (`int`) 0 indicates success. +- column_meta: (`[1][3]any`) Only returns `[["affected_rows", "INT", 4]]`. +- rows: (`int`) Only returns `1`. +- data: (`[][]any`) Returns the number of rows affected. + +#### Successful Query + +Example: + +```json +{ + "code": 0, + "column_meta": [ + ["ts", "TIMESTAMP", 8], + ["count", "BIGINT", 8], + ["endpoint", "VARCHAR", 45], + ["status_code", "INT", 4], + ["client_ip", "VARCHAR", 40], + ["request_method", "VARCHAR", 15], + ["request_uri", "VARCHAR", 128] + ], + "data": [ + [ + "2022-06-29T05:50:55.401Z", + 2, + "LAPTOP-NNKFTLTG:6041", + 200, + "172.23.208.1", + "POST", + "/rest/sql" ], - "rows": 2 + [ + "2022-06-29T05:52:16.603Z", + 1, + "LAPTOP-NNKFTLTG:6041", + 200, + "172.23.208.1", + "POST", + "/rest/sql" + ], + [ + "2022-06-29T06:28:14.118Z", + 1, + "LAPTOP-NNKFTLTG:6041", + 200, + "172.23.208.1", + "POST", + "/rest/sql" + ], + [ + "2022-06-29T05:52:16.603Z", + 2, + "LAPTOP-NNKFTLTG:6041", + 401, + "172.23.208.1", + "POST", + "/rest/sql" + ] + ], + "rows": 4 +} +``` + +Description: + +- code: `int` 0 indicates success. +- column_meta: (`[][3]any`) Column information. Each column is described with three values: column name (string), column type (string), and type length (int). +- rows: (`int`) The number of rows returned. +- data: (`[][]any`) + +The following types may be returned: + +- "NULL" +- "BOOL" +- "TINYINT" +- "SMALLINT" +- "INT" +- "BIGINT" +- "FLOAT" +- "DOUBLE" +- "VARCHAR" +- "TIMESTAMP" +- "NCHAR" +- "TINYINT UNSIGNED" +- "SMALLINT UNSIGNED" +- "INT UNSIGNED" +- "BIGINT UNSIGNED" +- "JSON" + +#### Errors + +Example: + +```json +{ + "code": 9728, + "desc": "syntax error near \"1\"" } ``` Description: -- status: tells you whethre the operation result is success or failure. -- head: the definition of the table, or just one column "affected_rows" if no result set is returned. (As of version 2.0.17.0, it is recommended not to rely on the head return value to determine the data column type but rather use column_meta. In later versions, the head item may be removed from the return value.) -- column_meta: this item is added to the return value to indicate the data type of each column in the data with version 2.0.17.0 and later versions. Each column is described by three values: column name, column type, and type length. For example, `["current",6,4]` means that the column name is "current", the column type is 6, which is the float type, and the type length is 4, which is the float type with 4 bytes. If the column type is binary or nchar, the type length indicates the maximum length of content stored in the column, not the length of the specific data in this return value. When the column type is nchar, the type length indicates the number of Unicode characters that can be saved, not bytes. -- data: The exact data returned, presented row by row, or just [[affected_rows]] if no result set is returned. The order of the data columns in each row of data is the same as that of the data columns described in column_meta. -- rows: Indicates how many rows of data there are. - -The column types in column_meta are described as follows: - -- 1:BOOL -- 2:TINYINT -- 3:SMALLINT -- 4:INT -- 5:BIGINT -- 6:FLOAT -- 7:DOUBLE -- 8:BINARY -- 9:TIMESTAMP -- 10:NCHAR +- code: (`int`) Error code. +- desc: (`string`): Error code description. ## Custom Authorization Code @@ -165,11 +253,9 @@ curl http://:/rest/login// Where `fqdn` is the FQDN or IP address of the TDengine database. `port` is the port number of the TDengine service. `username` is the database username. `password` is the database password. The return value is in `JSON` format, and the meaning of each field is as follows. -- status: flag bit of the request result - -- code: return value code - -- desc: authorization code +- status: flag bit of the request result. +- code: return value code. +- desc: authorization code. Example of getting authorization code. @@ -187,7 +273,7 @@ Response body: } ``` -## For example +## Usage examples - query all records from table d1001 of database demo @@ -199,19 +285,44 @@ Response body: ```json { - "status": "succ", - "head": ["ts", "current", "voltage", "phase"], - "column_meta": [ - ["ts", 9, 8], - ["current", 6, 4], - ["voltage", 4, 4], - ["phase", 6, 4] - ], - "data": [ - ["2018-10-03 14:38:05.000", 10.3, 219, 0.31], - ["2018-10-03 14:38:15.000", 12.6, 218, 0.33] - ], - "rows": 2 + "code": 0, + "column_meta": [ + [ + "ts", + "TIMESTAMP", + 8 + ], + [ + "current", + "FLOAT", + 4 + ], + [ + "voltage", + "INT", + 4 + ], + [ + "phase", + "FLOAT", + 4 + ] + ], + "data": [ + [ + "2022-07-30T06:44:40.32Z", + 10.3, + 219, + 0.31 + ], + [ + "2022-07-30T06:44:41.32Z", + 12.6, + 218, + 0.33 + ] + ], + "rows": 2 } ``` @@ -225,83 +336,23 @@ Response body: ```json { - "status": "succ", - "head": ["affected_rows"], - "column_meta": [["affected_rows", 4, 4]], - "data": [[1]], - "rows": 1 + "code": 0, + "column_meta": [ + [ + "affected_rows", + "INT", + 4 + ] + ], + "data": [ + [ + 0 + ] + ], + "rows": 1 } ``` -## Other Uses - -### Unix timestamps for result sets - -When the HTTP request URL uses `/rest/sqlt`, the returned result set's timestamp value will be in Unix timestamp format, for example: - -```bash -curl -L -H "Authorization: Basic cm9vdDp0YW9zZGF0YQ==" -d "select * from demo.d1001" 192.168.0.1:6041/rest/sqlt -``` - -Response body: - -```json -{ - "status": "succ", - "head": ["ts", "current", "voltage", "phase"], - "column_meta": [ - ["ts", 9, 8], - ["current", 6, 4], - ["voltage", 4, 4], - ["phase", 6, 4] - ], - "data": [ - [1538548685000, 10.3, 219, 0.31], - [1538548695000, 12.6, 218, 0.33] - ], - "rows": 2 -} -``` - -### UTC format for the result set +## Reference -When the HTTP request URL uses `/rest/sqlutc`, the timestamp of the returned result set will be expressed as a UTC format, for example: - -```bash - curl -L -H "Authorization: Basic cm9vdDp0YW9zZGF0YQ==" -d "select * from demo.t1" 192.168.0.1:6041/rest/sqlutc -``` - -Response body: - -```json -{ - "status": "succ", - "head": ["ts", "current", "voltage", "phase"], - "column_meta": [ - ["ts", 9, 8], - ["current", 6, 4], - ["voltage", 4, 4], - ["phase", 6, 4] - ], - "data": [ - ["2018-10-03T14:38:05.000+0800", 10.3, 219, 0.31], - ["2018-10-03T14:38:15.000+0800", 12.6, 218, 0.33] - ], - "rows": 2 -} -``` - -## Important configuration items - -Only some configuration parameters related to the RESTful interface are listed below. Please see the description in the configuration file for other system parameters. - -- The port number of the external RESTful service is bound to 6041 by default (the actual value is serverPort + 11, so it can be changed by modifying the setting of the serverPort parameter). -- httpMaxThreads: the number of threads to start, default is 2 (the default value is rounded down to half of the CPU cores with version 2.0.17.0 and later versions). -- restfulRowLimit: the maximum number of result sets (in JSON format) to return. The default value is 10240. -- httpEnableCompress: whether to support compression, the default is not supported. Currently, TDengine only supports the gzip compression format. -- httpDebugFlag: logging switch, default is 131. 131: error and alarm messages only, 135: debug messages, 143: very detailed debug messages. -- httpDbNameMandatory: users must specify the default database name in the RESTful URL. The default is 0, which turns off this check. If set to 1, users must put a default database name in every RESTful URL. Otherwise, it will return an execution error and reject this SQL statement, regardless of whether the SQL statement executed at this time requires a specified database. - -:::note -If you are using the REST API provided by taosd, you should write the above configuration in taosd's configuration file taos.cfg. If you use the REST API of taosAdapter, you need to refer to taosAdapter [corresponding configuration method](/reference/taosadapter/). -::: +[taosAdapter](/reference/taosadapter/) diff --git a/docs/en/14-reference/03-connector/03-connector.mdx b/docs/en/14-reference/03-connector/03-connector.mdx index c3f4530023db19e807565573bd10d41dafcd6f8e..49e2dceec531cf8449749ea9dbb111079771a788 100644 --- a/docs/en/14-reference/03-connector/03-connector.mdx +++ b/docs/en/14-reference/03-connector/03-connector.mdx @@ -8,13 +8,13 @@ TDengine provides a rich set of APIs (application development interface). To fac ## Supported platforms -Currently, TDengine's native interface connectors can support platforms such as X64/ARM64 hardware platforms and Linux/Win64 development environments. The comparison matrix is as follows. +Currently, TDengine's native interface connectors can support platforms such as x64 and ARM hardware platforms and Linux and Windows development environments. The comparison matrix is as follows. -| **CPU** | **OS** | **JDBC** | **Python** | **Go** | **Node.js** | **C#** | **Rust** | C/C++ | -| ------- | ------ | -------- | ---------- | ------ | ----------- | ------ | -------- | ----- | -| **X86 64bit** | **Linux** | ● | ● | ● | ● | ● | ● | ● | -| **X86 64bit** | **Win64** | ● | ● | ● | ● | ● | ● | ● | -| **ARM64** | **Linux** | ● | ● | ● | ● | ○ | ○ | ● | +| **CPU** | **OS** | **Java** | **Python** | **Go** | **Node.js** | **C#** | **Rust** | C/C++ | +| -------------- | --------- | -------- | ---------- | ------ | ----------- | ------ | -------- | ----- | +| **X86 64bit** | **Linux** | ● | ● | ● | ● | ● | ● | ● | +| **X86 64bit** | **Win64** | ● | ● | ● | ● | ● | ● | ● | +| **ARM64** | **Linux** | ● | ● | ● | ● | ○ | ○ | ● | Where ● means the official test verification passed, ○ means the unofficial test verification passed, -- means no assurance. @@ -26,6 +26,7 @@ TDengine version updates often add new features, and the connector versions in t | **TDengine Versions** | **Java** | **Python** | **Go** | **C#** | **Node.js** | **Rust** | | --------------------- | -------- | ---------- | ------------ | ------------- | --------------- | -------- | +| **3.0.0.0 and later** | 3.0.0 | current version | 3.0 branch | 3.0.0 | 3.0.0 | current version | | **2.4.0.14 and up** | 2.0.38 | current version | develop branch | 1.0.2 - 1.0.6 | 2.0.10 - 2.0.12 | current version | | **2.4.0.6 and up** | 2.0.37 | current version | develop branch | 1.0.2 - 1.0.6 | 2.0.10 - 2.0.12 | current version | | **2.4.0.4 - 2.4.0.5** | 2.0.37 | current version | develop branch | 1.0.2 - 1.0.6 | 2.0.10 - 2.0.12 | current version | @@ -42,9 +43,8 @@ Comparing the connector support for TDengine functional features as follows. | -------------- | -------- | ---------- | ------ | ------ | ----------- | -------- | | **Connection Management** | Support | Support | Support | Support | Support | Support | | **Regular Query** | Support | Support | Support | Support | Support | Support | -| **Continuous Query** | Support | Support | Support | Support | Support | Support | | **Parameter Binding** | Support | Support | Support | Support | Support | Support | -| **Subscription** | Support | Support | Support | Support | Support | Not Supported | +| ** TMQ ** | Support | Support | Support | Support | Support | Support | | **Schemaless** | Support | Support | Support | Support | Support | Support | | **DataFrame** | Not Supported | Support | Not Supported | Not Supported | Not Supported | Not Supported | @@ -52,17 +52,17 @@ Comparing the connector support for TDengine functional features as follows. The different database framework specifications for various programming languages do not mean that all C/C++ interfaces need a wrapper. ::: -### Using the REST interface +### Use HTTP Interfaces (REST or WebSocket) | **Functional Features** | **Java** | **Python** | **Go** | **C# (not supported yet)** | **Node.js** | **Rust** | | ------------------------------ | -------- | ---------- | -------- | ------------------ | ----------- | -------- | | **Connection Management** | Support | Support | Support | N/A | Support | Support | | **Regular Query** | Support | Support | Support | N/A | Support | Support | | **Continous Query ** | Support | Support | Support | N/A | Support | Support | -| **Parameter Binding** | Not Supported | Not Supported | Not Supported | N/A | Not Supported | Not Supported | -| **Subscription** | Not Supported | Not Supported | Not Supported | N/A | Not Supported | Not Supported | -| **Schemaless** | Not supported | Not Supported | Not Supported | N/A | Not Supported | Not supported | -| **Bulk Pulling (based on WebSocket) **| Support | Not Supported | Not Supported | N/A | Not Supported | Not Supported | +| **Parameter Binding** | Not supported | Not supported | Not supported | N/A | Not supported | Support | +| ** TMQ ** | Not supported | Not supported | Not supported | N/A | Not supported | Support | +| **Schemaless** | Not supported | Not supported | Not supported | N/A | Not supported | Not supported | +| **Bulk Pulling (based on WebSocket) **| Support | Support | Not Supported | N/A | Not Supported | Supported | | **DataFrame** | Not supported | Support | Not supported | N/A | Not supported | Not supported | :::warning @@ -85,7 +85,7 @@ The client driver needs to be installed if you use the native interface connecto ::: -### Installation steps +### Install @@ -96,7 +96,7 @@ The client driver needs to be installed if you use the native interface connecto -### Installation Verification +### Verify After completing the above installation and configuration and you have confirmed that the TDengine service is up and running, you can execute the TDengine CLI tool to log in. diff --git a/docs/en/14-reference/03-connector/_linux_install.mdx b/docs/en/14-reference/03-connector/_linux_install.mdx index 3fa123794cd8ff304a1bc13449591194e7320aa9..07f8fb5c7118d84c53017f44d9811a3357944cfc 100644 --- a/docs/en/14-reference/03-connector/_linux_install.mdx +++ b/docs/en/14-reference/03-connector/_linux_install.mdx @@ -1,16 +1,15 @@ -import PkgList from "/components/PkgList"; +import PkgListV3 from "/components/PkgListV3"; -1. Download the TDengine client installation package +1. Download the client installation package - + - [All Packages](https://www.taosdata.com/en/all-downloads/) + [All Downloads](../../releases) 2. Unzip Download the package to any directory the current user has read/write permission. Then execute `tar -xzvf TDengine-client-VERSION.tar.gz` command. The VERSION should be the version of the package you just downloaded. - 3. Execute the install script Once the package is unzipped, you will see the following files in the directory: @@ -18,17 +17,14 @@ import PkgList from "/components/PkgList"; - _ taos.tar.gz_: client driver package - _ driver_: TDengine client driver - _examples_: some example programs of different programming languages (C/C#/go/JDBC/MATLAB/python/R) - You can run `install_client.sh` to install it. - -4. Edit taos.cfg +4. configure taos.cfg Edit `taos.cfg` file (full path is `/etc/taos/taos.cfg` by default), modify `firstEP` with actual TDengine server's End Point, for example `h1.tdengine.com:6030` :::tip 1. If the computer does not run the TDengine service but installs the TDengine client driver, then you need to config `firstEP` in `taos.cfg` only, and there is no need to configure `FQDN`; - 2. If you encounter the "Unable to resolve FQDN" error, please make sure the FQDN in the `/etc/hosts` file of the current computer is correctly configured, or the DNS service is correctly configured. ::: diff --git a/docs/en/14-reference/03-connector/_preparition.mdx b/docs/en/14-reference/03-connector/_preparition.mdx new file mode 100644 index 0000000000000000000000000000000000000000..87538ebfd8c60507aec90ee86e427d85979dbc4a --- /dev/null +++ b/docs/en/14-reference/03-connector/_preparition.mdx @@ -0,0 +1,10 @@ +- 已安装客户端驱动（使用原生连接必须安装，使用 REST 连接无需安装） + +:::info + +由于 TDengine 的客户端驱动使用 C 语言编写，使用原生连接时需要加载系统对应安装在本地的客户端驱动共享库文件，通常包含在 TDengine 安装包。TDengine Linux 服务端安装包附带了 TDengine 客户端，也可以单独安装 [Linux 客户端](/get-started/) 。在 Windows 环境开发时需要安装 TDengine 对应的 [Windows 客户端](https://www.taosdata.com/cn/all-downloads/#TDengine-Windows-Client) 。 + +- libtaos.so: 在 Linux 系统中成功安装 TDengine 后，依赖的 Linux 版客户端驱动 libtaos.so 文件会被自动拷贝至 /usr/lib/libtaos.so，该目录包含在 Linux 自动扫描路径上，无需单独指定。 +- taos.dll: 在 Windows 系统中安装完客户端之后，依赖的 Windows 版客户端驱动 taos.dll 文件会自动拷贝到系统默认搜索路径 C:/Windows/System32 下，同样无需要单独指定。 + +::: diff --git a/docs/en/14-reference/03-connector/_verify_linux.mdx b/docs/en/14-reference/03-connector/_verify_linux.mdx index 875c9e132b7acbbe95d394ae6cee6f2bd95ddbe0..3438b1578eaa1df38115d80fd67f491c071d619a 100644 --- a/docs/en/14-reference/03-connector/_verify_linux.mdx +++ b/docs/en/14-reference/03-connector/_verify_linux.mdx @@ -2,10 +2,6 @@ Execute TDengine CLI program `taos` directly from the Linux shell to connect to ```text $ taos -Welcome to the TDengine shell from Linux, Client Version:3.0.0.0 -Copyright (c) 2022 by TAOS Data, Inc. All rights reserved. - -Server is Community Edition. taos> show databases; name | create_time | vgroups | ntables | replica | strict | duration | keep | buffer | pagesize | pages | minrows | maxrows | comp | precision | status | retention | single_stable | cachemodel | cachesize | wal_level | wal_fsync_period | wal_retention_period | wal_retention_size | wal_roll_period | wal_seg_size | diff --git a/docs/en/14-reference/03-connector/_verify_windows.mdx b/docs/en/14-reference/03-connector/_verify_windows.mdx index 4813bd24c3b4a2a4bf04af3c397bdadd22e9e399..402b1705110295f4d465c9ae7c734eb3e0ad0b5f 100644 --- a/docs/en/14-reference/03-connector/_verify_windows.mdx +++ b/docs/en/14-reference/03-connector/_verify_windows.mdx @@ -1,11 +1,6 @@ Go to the `C:\TDengine` directory from `cmd` and execute TDengine CLI program `taos.exe` directly to connect to the TDengine service and enter the TDengine CLI interface, for example, as follows: ```text -Welcome to the TDengine shell from Windows, Client Version:3.0.0.0 -Copyright (c) 2022 by TAOS Data, Inc. All rights reserved. - -Server is Community Edition. - taos> show databases; name | create_time | vgroups | ntables | replica | strict | duration | keep | buffer | pagesize | pages | minrows | maxrows | comp | precision | status | retention | single_stable | cachemodel | cachesize | wal_level | wal_fsync_period | wal_retention_period | wal_retention_size | wal_roll_period | wal_seg_size | ========================================================================================================================================================================================================================================================================================================================================================================================================================================================================= diff --git a/docs/en/14-reference/03-connector/_windows_install.mdx b/docs/en/14-reference/03-connector/_windows_install.mdx index 2819be615ee0a80da9f0324d8d41e9b247e8a7f6..ea638ed1ed6c64c3ec4ceaea436f65dd1f09a27e 100644 --- a/docs/en/14-reference/03-connector/_windows_install.mdx +++ b/docs/en/14-reference/03-connector/_windows_install.mdx @@ -1,11 +1,10 @@ -import PkgList from "/components/PkgList"; +import PkgListV3 from "/components/PkgListV3"; 1. Download the client installation package - - - [All downloads](https://www.taosdata.com/cn/all-downloads/) + + [All Downloads](../../releases) 2. Execute the installer, select the default value as prompted, and complete the installation 3. Installation path diff --git a/docs/en/14-reference/03-connector/cpp.mdx b/docs/en/14-reference/03-connector/cpp.mdx index e0cdf2bf2ce7eed06cacaf71a5b9baf56a3aee2b..5839ed4af89723dcee5e80c186af25a90ae59972 100644 --- a/docs/en/14-reference/03-connector/cpp.mdx +++ b/docs/en/14-reference/03-connector/cpp.mdx @@ -4,7 +4,7 @@ sidebar_label: C/C++ title: C/C++ Connector --- -C/C++ developers can use TDengine's client driver and the C/C++ connector, to develop their applications to connect to TDengine clusters for data writing, querying, and other functions. To use the C/C++ connector you must include the TDengine header file _taos.h_, which lists the function prototypes of the provided APIs. The application also needs to link to the corresponding dynamic libraries on the platform where it is located. +C/C++ developers can use TDengine's client driver and the C/C++ connector, to develop their applications to connect to TDengine clusters for data writing, querying, and other functions. To use the C/C++ connector you must include the TDengine header file _taos.h_, which lists the function prototypes of the provided APIs. The application also needs to link to the corresponding dynamic libraries on the platform where it is located. ```c #include @@ -12,8 +12,8 @@ C/C++ developers can use TDengine's client driver and the C/C++ connector, to de After TDengine server or client installation, `taos.h` is located at -- Linux: `/usr/local/taos/include` -- Windows: `C:\TDengine\include` +- Linux：`/usr/local/taos/include` +- Windows：`C:\TDengine\include` The dynamic libraries for the TDengine client driver are located in. @@ -28,7 +28,7 @@ Please refer to [list of supported platforms](/reference/connector#supported-pla The version number of the TDengine client driver and the version number of the TDengine server should be same. A lower version of the client driver is compatible with a higher version of the server, if the first three version numbers are the same (i.e., only the fourth version number is different). For e.g. if the client version is x.y.z.1 and the server version is x.y.z.2 the client and server are compatible. But in general we do not recommend using a lower client version with a newer server version. It is also strongly discouraged to use a higher version of the client driver to access a lower version of the TDengine server. -## Installation steps +## Installation Steps Please refer to the [Installation Steps](/reference/connector#installation-steps) for TDengine client driver installation @@ -45,7 +45,7 @@ The following is sample code for establishing a connection, which omits the quer exit(1); } - /* put your code here for query and writing */ + /* put your code here for read and write */ taos_close(taos); taos_cleanup(); @@ -60,7 +60,7 @@ In the above example code, `taos_connect()` establishes a connection to port 603 ::: -## Example program +## Sample program This section shows sample code for standard access methods to TDengine clusters using the client driver. @@ -125,7 +125,7 @@ You can find it in the installation directory under the `examples/c` path. This ::: -## API reference +## API Reference The following describes the basic API, synchronous API, asynchronous API, subscription API, and schemaless write API of TDengine client driver, respectively. @@ -141,10 +141,9 @@ The base API is used to do things like create database connections and provide a Cleans up the runtime environment and should be called before the application exits. -- ` int taos_options(TSDB_OPTION option, const void * arg, ...) ` +- `int taos_options(TSDB_OPTION option, const void * arg, ...)` - Set client options, currently supports region setting (`TSDB_OPTION_LOCALE`), character set -(`TSDB_OPTION_CHARSET`), time zone (`TSDB_OPTION_TIMEZONE`), configuration file path (`TSDB_OPTION_CONFIGDIR`). The region setting, character set, and time zone default to the current settings of the operating system. + Set client options, currently supports region setting (`TSDB_OPTION_LOCALE`), character set (`TSDB_OPTION_CHARSET`), time zone (`TSDB_OPTION_TIMEZONE`), configuration file path (`TSDB_OPTION_CONFIGDIR`). The region setting, character set, and time zone default to the current settings of the operating system. - `char *taos_get_client_info()` @@ -157,7 +156,7 @@ The base API is used to do things like create database connections and provide a - host: FQDN of any node in the TDengine cluster - user: user name - pass: password - - db: the database name. Even if the user does not provide this, the connection will still work correctly. The user can create a new database through this connection. If the user provides the database name, it means that the database has already been created and the connection can be used for regular operations on the database. + - db: the database name. Even if the user does not provide this, the connection will still work correctly. The user can create a new database through this connection. If the user provides the database name, it means that the database has already been created and the connection can be used for regular operations on the database. - port: the port the taosd program is listening on NULL indicates a failure. The application needs to save the returned parameters for subsequent use. @@ -171,7 +170,7 @@ The base API is used to do things like create database connections and provide a Get server-side version information. -- ` int taos_select_db(TAOS *taos, const char *db)` +- `int taos_select_db(TAOS *taos, const char *db)` Set the current default database to `db`. @@ -211,15 +210,15 @@ The APIs described in this subsection are all synchronous interfaces. After bein Get the number of rows affected by the executed SQL statement. -- ` TAOS_FIELD *taos_fetch_fields(TAOS_RES *res)` +- `TAOS_FIELD *taos_fetch_fields(TAOS_RES *res)` Gets the properties of each column of the query result set (column name, column data type, column length), used in conjunction with `taos_num_fields()` to parse a tuple (one row) of data returned by `taos_fetch_row()`. The structure of `TAOS_FIELD` is as follows. ```c typedef struct taosField { - char name[65]; // column name - uint8_t type; // data type - int16_t bytes; // length, in bytes + char name[65]; // column name + uint8_t type; // data type + int16_t bytes; // length, in bytes } TAOS_FIELD; ``` @@ -227,7 +226,7 @@ typedef struct taosField { Stops the execution of the current query. -- ` void taos_free_result(TAOS_RES *res)` +- `void taos_free_result(TAOS_RES *res)` Frees the query result set and the associated resources. Be sure to call this API to free the resources after the query is completed. Failing to call this, may lead to a memory leak in the application. However, note that the application will crash if you call a function like `taos_consume()` to get the query results after freeing the resources. @@ -235,7 +234,7 @@ typedef struct taosField { Get the reason for the failure of the last API call. The return value is an error message identified by a string. -- 'int taos_errno(TAOS_RES *res)` +- `int taos_errno(TAOS_RES *res)` Get the reason for the last API call failure. The return value is the error code. @@ -294,7 +293,7 @@ The specific functions related to the interface are as follows (see also the [pr Creates a TAOS_STMT object for subsequent calls. -- ` int taos_stmt_prepare(TAOS_STMT *stmt, const char *sql, unsigned long length)` +- `int taos_stmt_prepare(TAOS_STMT *stmt, const char *sql, unsigned long length)` Parse a SQL command, and bind the parsed result and parameter information to `stmt`. If the parameter length is greater than 0, use this parameter as the length of the SQL command. If it is equal to 0, the length of the SQL command will be determined automatically. @@ -332,16 +331,16 @@ The specific functions related to the interface are as follows (see also the [pr ```c typedef struct TAOS_MULTI_BIND { - int buffer_type; - void * buffer; - uintptr_t buffer_length; - uintptr_t * length; - char * is_null; - int num; // the number of columns + int buffer_type; + void * buffer; + uintptr_t buffer_length; + uintptr_t * length; + char * is_null; + int num; // the number of columns } TAOS_MULTI_BIND; ``` -- ` int taos_stmt_add_batch(TAOS_STMT *stmt)` +- `int taos_stmt_add_batch(TAOS_STMT *stmt)` Adds the currently bound parameter to the batch. After calling this function, you can call `taos_stmt_bind_param()` or `taos_stmt_bind_param_batch()` again to bind a new parameter. Note that this function only supports INSERT/IMPORT statements. Other SQL command such as SELECT will return an error. @@ -349,7 +348,7 @@ The specific functions related to the interface are as follows (see also the [pr Execute the prepared statement. Currently, a statement can only be executed once. -- ` TAOS_RES* taos_stmt_use_result(TAOS_STMT *stmt)` +- `TAOS_RES* taos_stmt_use_result(TAOS_STMT *stmt)` Gets the result set of a statement. Use the result set in the same way as in the non-parametric call. When finished, `taos_free_result()` should be called on this result set to free resources. @@ -357,7 +356,7 @@ The specific functions related to the interface are as follows (see also the [pr Finish execution and release all resources. -- ` char * taos_stmt_errstr(TAOS_STMT *stmt)` +- `char * taos_stmt_errstr(TAOS_STMT *stmt)` (Available in 2.1.3.0 and later versions) Used to get error information if other STMT APIs return errors (return error codes or null pointers). @@ -405,46 +404,3 @@ In addition to writing data using the SQL method or the parameter binding API, w **Supported Versions** This feature interface is supported from version 2.3.0.0. - -### Subscription and Consumption API - -The Subscription API currently supports subscribing to one or more tables and continuously fetching the latest data written to them by polling periodically. - -- `TAOS_SUB *taos_subscribe(TAOS* taos, int restart, const char* topic, const char *sql, TAOS_SUBSCRIBE_CALLBACK fp, void *param, int interval)` - - This function is responsible for starting the subscription service, returning the subscription object on success and `NULL` on failure, with the following parameters. - - - taos: the database connection that has been established. - - restart: if the subscription already exists, whether to restart or continue the previous subscription. - - topic: the topic of the subscription (i.e., the name). This parameter is the unique identifier of the subscription. - - sql: the query statement of the subscription which can only be a _select_ statement. Only the original data should be queried, and data can only be queried in temporal order. - - fp: the callback function when the query result is received only used when called asynchronously. This parameter should be passed `NULL` when called synchronously. The function prototype is described below. - - param: additional parameter when calling the callback function. The system API will pass it to the callback function as is, without any processing. - - interval: polling period in milliseconds. The callback function will be called periodically according to this parameter when called asynchronously. The interval should not be too small to avoid impact on system performance when called synchronously. If the interval between two calls to `taos_consume()` is less than this period, the API will block until the interval exceeds this period. - -- ` typedef void (*TAOS_SUBSCRIBE_CALLBACK)(TAOS_SUB* tsub, TAOS_RES *res, void* param, int code)` - - The prototype of the callback function in asynchronous mode with the following parameters - - - tsub: subscription object - - res: query result set, note that there may be no records in the result set - - param: additional parameters provided by the client program when calling `taos_subscribe()` - - code: error code - - :::note - The callback function should not take too long to process, especially if the returned result set has a lot of data. Otherwise, it may lead to an abnormal state, such as client blocking. If you must perform complex calculations, we recommend handling them in a separate thread. - - ::: - -- `TAOS_RES *taos_consume(TAOS_SUB *tsub)` - - In synchronous mode, this function is used to fetch the results of a subscription. The user application places it in a loop. If the interval between two calls to `taos_consume()` is less than the polling period of the subscription, the API will block until the interval exceeds this period. If a new record arrives in the database, the API returns that latest record. Otherwise, it returns an empty result set with no records. If the return value is `NULL`, there is a system error. This API should not be called by user programs in asynchronous mode. - - :::note - After calling `taos_consume()`, the user application should make sure to call `taos_fetch_row()` or `taos_fetch_block()` to process the subscription results as soon as possible. Otherwise, the server-side will keep caching the query result data waiting to be read by the client, which in extreme cases will cause the server side to run out of memory and affect the stability of the service. - - ::: - -- `void taos_unsubscribe(TAOS_SUB *tsub, int keepProgress)` - - Unsubscribe. If the parameter `keepProgress` is not 0, the API will keep the progress information of the subscription, and subsequent calls to `taos_subscribe()` will continue based on this progress; otherwise, the progress information will be deleted, and subsequent readings will have to be restarted. diff --git a/docs/en/14-reference/03-connector/csharp.mdx b/docs/en/14-reference/03-connector/csharp.mdx index 2d1b62fe89c542280c4264dd478538fa00634c79..388ae49d09e1ee8a7e0f012432d9bbb98da3fc45 100644 --- a/docs/en/14-reference/03-connector/csharp.mdx +++ b/docs/en/14-reference/03-connector/csharp.mdx @@ -8,7 +8,7 @@ title: C# Connector import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; -import Preparation from "./_preparation.mdx" +import Preparition from "./_preparition.mdx" import CSInsert from "../../07-develop/03-insert-data/_cs_sql.mdx" import CSInfluxLine from "../../07-develop/03-insert-data/_cs_line.mdx" import CSOpenTSDBTelnet from "../../07-develop/03-insert-data/_cs_opts_telnet.mdx" @@ -16,16 +16,17 @@ import CSOpenTSDBJson from "../../07-develop/03-insert-data/_cs_opts_json.mdx" import CSQuery from "../../07-develop/04-query-data/_cs.mdx" import CSAsyncQuery from "../../07-develop/04-query-data/_cs_async.mdx" - `TDengine.Connector` is a C# language connector provided by TDengine that allows C# developers to develop C# applications that access TDengine cluster data. The `TDengine.Connector` connector supports connect to TDengine instances via the TDengine client driver (taosc), providing data writing, querying, subscription, schemaless writing, bind interface, etc. The `TDengine.Connector` currently does not provide a REST connection interface. Developers can write their RESTful application by referring to the [REST API](/reference/rest-api/) documentation. This article describes how to install `TDengine.Connector` in a Linux or Windows environment and connect to TDengine clusters via `TDengine.Connector` to perform basic operations such as data writing and querying. -The source code of `TDengine.Connector` is hosted on [GitHub](https://github.com/taosdata/taos-connector-dotnet). +Note: TDengine Connector 3.x is not compatible with TDengine 2.x. In an environment with TDengine 2.x, you must use TDengine.Connector 1.x for the C# connector. + +The source code of `TDengine.Connector` is hosted on [GitHub](https://github.com/taosdata/taos-connector-dotnet/tree/3.0). -## Supported Platforms +## Supported platforms The supported platforms are the same as those supported by the TDengine client driver. @@ -57,29 +58,29 @@ Please refer to [version support list](/reference/connector#version-support) You can reference the `TDengine.Connector` published in Nuget to the current project via the `dotnet` command under the path of the existing .NET project. -``` +``` bash dotnet add package TDengine.Connector ``` -You can download TDengine's source code and directly reference the latest version of the TDengine.Connector library +You can [download the source code](https://github.com/taosdata/taos-connector-dotnet/tree/3.0) and directly reference the latest version of the TDengine.Connector library. -``` -git clone https://github.com/taosdata/TDengine.git -cd TDengine/src/connector/C#/src/ -cp -r TDengineDriver/ myProject +```bash +git clone -b 3.0 https://github.com/taosdata/taos-connector-dotnet.git +cd taos-connector-dotnet +cp -r src/ myProject cd myProject -dotnet add TDengineDriver/TDengineDriver.csproj +dotnet add exmaple.csproj reference src/TDengine.csproj ``` -## Create a connection +## Establish a Connection -```csharp +``` C# using TDengineDriver; namespace TDengineExample @@ -146,25 +147,24 @@ namespace TDengineExample |Sample program |Sample program description | |--------------------------------------------------------------------------------------------------------------------|------------ --------------------------------| -| [C#checker](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/C%23checker) | Using TDengine.Connector, you can test C# Driver's synchronous writes and queries | -| [TDengineTest](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/TDengineTest) | A simple example of writing and querying using TDengine. -| [insertCn](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/insertCn) | Example of writing and querying Chinese characters using TDengine. -| [jsonTag](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/jsonTag) | Example of writing and querying JSON tag type data using TDengine. -| [stmt](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/stmt) | Example of parameter binding using TDengine. -| [schemaless](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/schemaless) | Example of writing with schemaless implemented using TDengine. |schemaless -| [benchmark](https://github.com/taosdata/TDengine/tree/develop/examples/C%23/taosdemo) | A simple benchmark implemented using TDengine. -| [async query](https://github.com/taosdata/taos-connector-dotnet/blob/develop/examples/QueryAsyncSample.cs) | Example of an asynchronous query implemented using TDengine. Example of an asynchronous query -| [subscribe](https://github.com/taosdata/taos-connector-dotnet/blob/develop/examples/SubscribeSample.cs) | Example of subscribing to data using TDengine. Data example +| [CURD](https://github.com/taosdata/taos-connector-dotnet/blob/3.0/examples/Query/Query.cs) | Table creation, data insertion, and query examples with TDengine.Connector | +| [JSON Tag](https://github.com/taosdata/taos-connector-dotnet/blob/3.0/examples/JSONTag) | Writing and querying JSON tag data with TDengine Connector | +| [stmt](https://github.com/taosdata/taos-connector-dotnet/tree/3.0/examples/Stmt) | Parameter binding with TDengine Connector | +| [schemaless](https://github.com/taosdata/taos-connector-dotnet/blob/3.0/examples/schemaless) | Schemaless writes with TDengine Connector | +| [async query](https://github.com/taosdata/taos-connector-dotnet/blob/3.0/examples/AsyncQuery/QueryAsync.cs) | Asynchronous queries with TDengine Connector | +| [TMQ](https://github.com/taosdata/taos-connector-dotnet/blob/3.0/examples/TMQ/TMQ.cs) | Data subscription with TDengine Connector | ## Important update records | TDengine.Connector | Description | |--------------------|--------------------------------| +| 3.0.0 | Supports TDengine 3.0.0.0. TDengine 2.x is not supported. Added `TDengine.Impl.GetData()` interface to deserialize query results. | +| 1.0.7 | Fixed TDengine.Query() memory leak. | | 1.0.6 | Fix schemaless bug in 1.0.4 and 1.0.5. | -| 1.0.5 | Fix Windows sync query Chinese error bug. | 1.0.4 | Fix schemaless bug. -| 1.0.4 | Add asynchronous query, subscription, and other functions. Fix the binding parameter bug. -| 1.0.3 | Add parameter binding, schemaless, JSON tag, etc. | new -| 1.0.2 | Add connection management, synchronous query, error messages, etc. ## Other +| 1.0.5 | Fix Windows sync query Chinese error bug. | 1.0.4 | Fix schemaless bug. | +| 1.0.4 | Add asynchronous query, subscription, and other functions. Fix the binding parameter bug. | +| 1.0.3 | Add parameter binding, schemaless, JSON tag, etc. | +| 1.0.2 | Add connection management, synchronous query, error messages, etc. | ## Other descriptions @@ -179,7 +179,7 @@ namespace TDengineExample 1. "Unable to establish connection", "Unable to resolve FQDN" - Usually, it's caused by an incorrect FQDN configuration. Please refer to this section in the [FAQ](https://docs.tdengine.com/2.4/train-faq/faq/#2-how-to-handle-unable-to-establish-connection) to troubleshoot. + Usually, it's caused by an incorrect FQDN configuration. Please refer to this section in the [FAQ](https://docs.tdengine.com/2.4/train-faq/faq/#2-how-to-handle-unable-to-establish-connection) to troubleshoot. 2. Unhandled exception. System.DllNotFoundException: Unable to load DLL 'taos' or one of its dependencies: The specified module cannot be found. diff --git a/docs/en/14-reference/03-connector/go.mdx b/docs/en/14-reference/03-connector/go.mdx index 8a05f2d841bbcdbab2bdb7471691ca0ae49a4f6b..29263550403e71614296e52285c956040b04387f 100644 --- a/docs/en/14-reference/03-connector/go.mdx +++ b/docs/en/14-reference/03-connector/go.mdx @@ -8,7 +8,7 @@ title: TDengine Go Connector import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; -import Preparation from "./_preparation.mdx" +import Preparition from "./_preparition.mdx" import GoInsert from "../../07-develop/03-insert-data/_go_sql.mdx" import GoInfluxLine from "../../07-develop/03-insert-data/_go_line.mdx" import GoOpenTSDBTelnet from "../../07-develop/03-insert-data/_go_opts_telnet.mdx" @@ -23,7 +23,7 @@ This article describes how to install `driver-go` and connect to TDengine cluste The source code of `driver-go` is hosted on [GitHub](https://github.com/taosdata/driver-go). -## Supported Platforms +## Supported platforms Native connections are supported on the same platforms as the TDengine client driver. REST connections are supported on all platforms that can run Go. @@ -41,33 +41,31 @@ A "native connection" is established by the connector directly to the TDengine i * Normal queries * Continuous queries * Subscriptions -* schemaless interface -* parameter binding interface +* Schemaless interface +* Parameter binding interface ### REST connection A "REST connection" is a connection between the application and the TDengine instance via the REST API provided by the taosAdapter component. The following features are supported: -* General queries +* Normal queries * Continuous queries -## Installation steps +## Installation Steps -### Pre-installation +### Pre-installation preparation -- Install Go development environment (Go 1.14 and above, GCC 4.8.5 and above) +* Install Go development environment (Go 1.14 and above, GCC 4.8.5 and above) - If you use the native connector, please install the TDengine client driver. Please refer to [Install Client Driver](/reference/connector/#install-client-driver) for specific steps Configure the environment variables and check the command. -* `go env` -* `gcc -v` +* ```go env``` +* ```gcc -v``` ### Use go get to install -``` -go get -u github.com/taosdata/driver-go/v2@develop -``` +`go get -u github.com/taosdata/driver-go/v3@latest` ### Manage with go mod @@ -75,14 +73,14 @@ go get -u github.com/taosdata/driver-go/v2@develop ```text go mod init taos-demo - ``` + ``` 2. Introduce taosSql ```go import ( "database/sql" - _ "github.com/taosdata/driver-go/v2/taosSql" + _ "github.com/taosdata/driver-go/v3/taosSql" ) ``` @@ -90,7 +88,7 @@ go get -u github.com/taosdata/driver-go/v2@develop ```text go mod tidy - ``` + ``` 4. Run the program with `go run taos-demo` or compile the binary with the `go build` command. @@ -99,14 +97,14 @@ go get -u github.com/taosdata/driver-go/v2@develop go build ``` -## Create a connection +## Establishing a connection ### Data source name (DSN) Data source names have a standard format, e.g. [PEAR DB](http://pear.php.net/manual/en/package.database.db.intro-dsn.php), but no type prefix (square brackets indicate optionally): ``` text -[username[:password]@][protocol[(address)]]/[dbname][?param1=value1&... ¶mN=valueN] +[username[:password]@][protocol[(address)]]/[dbname][?param1=value1&...¶mN=valueN] ``` DSN in full form. @@ -114,7 +112,6 @@ DSN in full form. ```text username:password@protocol(address)/dbname?param=value ``` - ### Connecting via connector @@ -126,7 +123,7 @@ Use `taosSql` as `driverName` and use a correct [DSN](#DSN) as `dataSourceName`, * configPath specifies the `taos.cfg` directory -Example. +For example: ```go package main @@ -135,13 +132,13 @@ import ( "database/sql" "fmt" - _ "github.com/taosdata/driver-go/v2/taosSql" + _ "github.com/taosdata/driver-go/v3/taosSql" ) func main() { var taosUri = "root:taosdata@tcp(localhost:6030)/" taos, err := sql.Open("taosSql", taosUri) - if err ! = nil { + if err != nil { fmt.Println("failed to connect TDengine, err:", err) return } @@ -158,7 +155,7 @@ Use `taosRestful` as `driverName` and use a correct [DSN](#DSN) as `dataSourceNa * `disableCompression` whether to accept compressed data, default is true do not accept compressed data, set to false if transferring data using gzip compression. * `readBufferSize` The default size of the buffer for reading data is 4K (4096), which can be adjusted upwards when the query result has a lot of data. -Example. +For example: ```go package main @@ -167,13 +164,13 @@ import ( "database/sql" "fmt" - _ "github.com/taosdata/driver-go/v2/taosRestful" + _ "github.com/taosdata/driver-go/v3/taosRestful" ) func main() { var taosUri = "root:taosdata@http(localhost:6041)/" taos, err := sql.Open("taosRestful", taosUri) - if err ! = nil { + if err != nil { fmt.Println("failed to connect TDengine, err:", err) return } @@ -208,14 +205,14 @@ func main() { ### More sample programs -* [sample program](https://github.com/taosdata/TDengine/tree/develop/examples/go) -* [Video tutorial](https://www.taosdata.com/blog/2020/11/11/1951.html). +* [sample program](https://github.com/taosdata/driver-go/tree/3.0/examples) + ## Usage limitations Since the REST interface is stateless, the `use db` syntax will not work. You need to put the db name into the SQL command, e.g. `create table if not exists tb1 (ts timestamp, a int)` to `create table if not exists test.tb1 (ts timestamp, a int)` otherwise it will report the error `[0x217] Database not specified or available`. -You can also put the db name in the DSN by changing `root:taosdata@http(localhost:6041)/` to `root:taosdata@http(localhost:6041)/test`. This method is supported by taosAdapter since TDengine 2.4.0.5. Executing the `create database` statement when the specified db does not exist will not report an error while executing other queries or writing against that db will report an error. +You can also put the db name in the DSN by changing `root:taosdata@http(localhost:6041)/` to `root:taosdata@http(localhost:6041)/test`. Executing the `create database` statement when the specified db does not exist will not report an error while executing other queries or writing against that db will report an error. The complete example is as follows. @@ -227,7 +224,7 @@ import ( "fmt" "time" - _ "github.com/taosdata/driver-go/v2/taosRestful" + _ "github.com/taosdata/driver-go/v3/taosRestful" ) func main() { @@ -269,35 +266,27 @@ func main() { ## Frequently Asked Questions -1. Cannot find the package `github.com/taosdata/driver-go/v2/taosRestful` - - Change the `github.com/taosdata/driver-go/v2` line in the require block of the `go.mod` file to `github.com/taosdata/driver-go/v2 develop`, then execute `go mod tidy`. - -2. bind interface in database/sql crashes +1. bind interface in database/sql crashes REST does not support parameter binding related interface. It is recommended to use `db.Exec` and `db.Query`. -3. error `[0x217] Database not specified or available` after executing other statements with `use db` statement +2. error `[0x217] Database not specified or available` after executing other statements with `use db` statement The execution of SQL command in the REST interface is not contextual, so using `use db` statement will not work, see the usage restrictions section above. -4. use `taosSql` without error but use `taosRestful` with error `[0x217] Database not specified or available` +3. use `taosSql` without error but use `taosRestful` with error `[0x217] Database not specified or available` Because the REST interface is stateless, using the `use db` statement will not take effect. See the usage restrictions section above. -5. Upgrade `github.com/taosdata/driver-go/v2/taosRestful` - - Change the `github.com/taosdata/driver-go/v2` line in the `go.mod` file to `github.com/taosdata/driver-go/v2 develop`, then execute `go mod tidy`. - -6. `readBufferSize` parameter has no significant effect after being increased +4. `readBufferSize` parameter has no significant effect after being increased Increasing `readBufferSize` will reduce the number of `syscall` calls when fetching results. If the query result is smaller, modifying this parameter will not improve performance significantly. If you increase the parameter value too much, the bottleneck will be parsing JSON data. If you need to optimize the query speed, you must adjust the value based on the actual situation to achieve the best query performance. -7. `disableCompression` parameter is set to `false` when the query efficiency is reduced +5. `disableCompression` parameter is set to `false` when the query efficiency is reduced When set `disableCompression` parameter to `false`, the query result will be compressed by `gzip` and then transmitted, so you have to decompress the data by `gzip` after getting it. -8. `go get` command can't get the package, or timeout to get the package +6. `go get` command can't get the package, or timeout to get the package Set Go proxy `go env -w GOPROXY=https://goproxy.cn,direct`. @@ -311,14 +300,13 @@ func main() { :::info This API is created successfully without checking permissions, but only when you execute a Query or Exec, and check if user/password/host/port is legal. - ::: -* `func (db *DB) Exec(query string, args . .interface{}) (Result, error)` +* `func (db *DB) Exec(query string, args ...interface{}) (Result, error)` `sql.Open` built-in method to execute non-query related SQL. -* `func (db *DB) Query(query string, args ... . interface{}) (*Rows, error)` +* `func (db *DB) Query(query string, args ...interface{}) (*Rows, error)` `sql.Open` Built-in method to execute query statements. @@ -338,17 +326,33 @@ The `af` package encapsulates TDengine advanced functions such as connection man #### Subscribe to -* `func (conn *Connector) Subscribe(restart bool, topic string, sql string, interval time.Duration) (Subscriber, error)` +* `func NewConsumer(conf *Config) (*Consumer, error)` + +Creates consumer group. + +* `func (c *Consumer) Subscribe(topics []string) error` + +Subscribes to a topic. + +* `func (c *Consumer) Poll(timeout time.Duration) (*Result, error)` - Subscribe to data. +Polling information. -* `func (s *taosSubscriber) Consume() (driver.Rows, error)` +* `func (c *Consumer) Commit(ctx context.Context, message unsafe.Pointer) error` - Consume the subscription data, returning the `Rows` structure of the `database/sql/driver` package. +Commit information. -* `func (s *taosSubscriber) Unsubscribe(keepProgress bool)` +* `func (c *Consumer) FreeMessage(message unsafe.Pointer)` - Unsubscribe from data. +Free information. + +* `func (c *Consumer) Unsubscribe() error` + +Unsubscribe. + +* `func (c *Consumer) Close() error` + +Close consumer. #### schemaless @@ -370,11 +374,7 @@ The `af` package encapsulates TDengine advanced functions such as connection man Parameter bound single row insert. -* `func (conn *Connector) StmtQuery(sql string, params *param.Param) (rows driver.Rows, err error)` - - Parameter bound query that returns the `Rows` structure of the `database/sql/driver` package. - -* `func (conn *Connector) InsertStmt() *insertstmt. +* `func (conn *Connector) InsertStmt() *insertstmt.InsertStmt` Initialize the parameters. @@ -412,4 +412,4 @@ The `af` package encapsulates TDengine advanced functions such as connection man ## API Reference -Full API see [driver-go documentation](https://pkg.go.dev/github.com/taosdata/driver-go/v2) +Full API see [driver-go documentation](https://pkg.go.dev/github.com/taosdata/driver-go/v3) diff --git a/docs/en/14-reference/03-connector/java.mdx b/docs/en/14-reference/03-connector/java.mdx index 1ea033eab4c7f9afb2cd4bf1eb82f5fd5b31d31a..39514c37ebf45974ad90b1b7b1e548c8cd4ea672 100644 --- a/docs/en/14-reference/03-connector/java.mdx +++ b/docs/en/14-reference/03-connector/java.mdx @@ -3,17 +3,17 @@ toc_max_heading_level: 4 sidebar_position: 2 sidebar_label: Java title: TDengine Java Connector -description: TDengine Java based on JDBC API and provide both native and REST connections +description: The TDengine Java Connector is implemented on the standard JDBC API and provides native and REST connectors. --- import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; -'taos-jdbcdriver' is TDengine's official Java language connector, which allows Java developers to develop applications that access the TDengine database. 'taos-jdbcdriver' implements the interface of the JDBC driver standard and provides two forms of connectors. One is to connect to a TDengine instance natively through the TDengine client driver (taosc), which supports functions including data writing, querying, subscription, schemaless writing, and bind interface. And the other is to connect to a TDengine instance through the REST interface provided by taosAdapter (2.4.0.0 and later). The implementation of the REST connection and those of the native connections have slight differences in features. +`taos-jdbcdriver` is the official Java connector for TDengine. Java developers can use it to develop applications that access data in TDengine. `taos-jdbcdriver` implements standard JDBC driver interfaces and two connection methods: One is **native connection**, which connects to TDengine instances natively through the TDengine client driver (taosc), supporting data writing, querying, subscriptions, schemaless writing, and bind interface. The second is **REST connection** which is implemented through taosAdapter. The set of features implemented by the REST connection differs slightly from those implemented by the native connection. -![TDengine Database tdengine-connector](tdengine-jdbc-connector.webp) +![TDengine Database Connector Java](tdengine-jdbc-connector.webp) -The preceding diagram shows two ways for a Java app to access TDengine via connector: +The preceding figure shows the two ways in which a Java application can access TDengine. - JDBC native connection: Java applications use TSDBDriver on physical node 1 (pnode1) to call client-driven directly (`libtaos.so` or `taos.dll`) APIs to send writing and query requests to taosd instances located on physical node 2 (pnode2). - JDBC REST connection: The Java application encapsulates the SQL as a REST request via RestfulDriver, sends it to the REST server (taosAdapter) on physical node 2. taosAdapter forwards the request to TDengine server and returns the result. @@ -30,18 +30,17 @@ TDengine's JDBC driver implementation is as consistent as possible with the rela ## Supported platforms -Native connection supports the same platform as TDengine client-driven support. +Native connections are supported on the same platforms as the TDengine client driver. REST connection supports all platforms that can run Java. ## Version support -Please refer to [Version Support List](/reference/connector#version-support). +Please refer to [version support list](/reference/connector#version-support) ## TDengine DataType vs. Java DataType TDengine currently supports timestamp, number, character, Boolean type, and the corresponding type conversion with Java is as follows: - | TDengine DataType | JDBCType | | ----------------- | ---------------------------------- | | TIMESTAMP | java.sql.Timestamp | @@ -58,7 +57,7 @@ TDengine currently supports timestamp, number, character, Boolean type, and the **Note**: Only TAG supports JSON types -## Installation steps +## Installation Steps ### Pre-installation preparation @@ -72,13 +71,15 @@ Before using Java Connector to connect to the database, the following conditions +taos-jdbcdriver has been published on the [Sonatype Repository](https://search.maven.org/artifact/com.taosdata.jdbc/taos-jdbcdriver) and synchronized to other major repositories. + - [sonatype](https://search.maven.org/artifact/com.taosdata.jdbc/taos-jdbcdriver) - [mvnrepository](https://mvnrepository.com/artifact/com.taosdata.jdbc/taos-jdbcdriver) - [maven.aliyun](https://maven.aliyun.com/mvn/search) Add following dependency in the `pom.xml` file of your Maven project: -```xml +```xml-dtd com.taosdata.jdbc taos-jdbcdriver @@ -91,26 +92,26 @@ Add following dependency in the `pom.xml` file of your Maven project: You can build Java connector from source code after cloning the TDengine project: -``` -git clone https://github.com/taosdata/taos-connector-jdbc.git --branch 2.0 +```shell +git clone https://github.com/taosdata/taos-connector-jdbc.git cd taos-connector-jdbc mvn clean install -Dmaven.test.skip=true ``` -After compilation, a jar package named taos-jdbcdriver-2.0.XX-dist.jar is generated in the target directory, and the compiled jar file is automatically placed in the local Maven repository. +After you have compiled taos-jdbcdriver, the `taos-jdbcdriver-3.0.*-dist.jar` file is created in the target directory. The compiled JAR file is automatically stored in your local Maven repository. -## Establish a connection +## Establishing a connection TDengine's JDBC URL specification format is: -`jdbc:[TAOS| TAOS-RS]://[host_name]:[port]/[database_name]? [user={user}|&password={password}|&charset={charset}|&cfgdir={config_dir}|&locale={locale}|&timezone={timezone}]` +`jdbc:[TAOS|TAOS-RS]://[host_name]:[port]/[database_name]?[user={user}|&password={password}|&charset={charset}|&cfgdir={config_dir}|&locale={locale}|&timezone={timezone}]` For establishing connections, native connections differ slightly from REST connections. - + ```java Class.forName("com.taosdata.jdbc.TSDBDriver"); @@ -130,7 +131,7 @@ The configuration parameters in the URL are as follows: - charset: The character set used by the client, the default value is the system character set. - locale: Client locale, by default, use the system's current locale. - timezone: The time zone used by the client, the default value is the system's current time zone. -- batchfetch: true: pulls result sets in batches when executing queries; false: pulls result sets row by row. The default value is: true. Enabling batch pulling and obtaining a batch of data can improve query performance when the query data volume is large. +- batchfetch: true: pulls result sets in batches when executing queries; false: pulls result sets row by row. The default value is true. Enabling batch pulling and obtaining a batch of data can improve query performance when the query data volume is large. - batchErrorIgnore:true: When executing statement executeBatch, if there is a SQL execution failure in the middle, the following SQL will continue to be executed. false: No more statements after the failed SQL are executed. The default value is: false. For more information about JDBC native connections, see [Video Tutorial](https://www.taosdata.com/blog/2020/11/11/1955.html). @@ -141,44 +142,41 @@ When you use a JDBC native connection to connect to a TDengine cluster, you can 1. Do not specify hostname and port in Java applications. - ```java - public Connection getConn() throws Exception{ - Class.forName("com.taosdata.jdbc.TSDBDriver"); - String jdbcUrl = "jdbc:TAOS://:/test?user=root&password=taosdata"; - Properties connProps = new Properties(); - connProps.setProperty(TSDBDriver.PROPERTY_KEY_CHARSET, "UTF-8"); - connProps.setProperty(TSDBDriver.PROPERTY_KEY_LOCALE, "en_US.UTF-8"); - connProps.setProperty(TSDBDriver.PROPERTY_KEY_TIME_ZONE, "UTC-8"); - Connection conn = DriverManager.getConnection(jdbcUrl, connProps); - return conn; - } - ``` +```java +public Connection getConn() throws Exception{ + Class.forName("com.taosdata.jdbc.TSDBDriver"); + String jdbcUrl = "jdbc:TAOS://:/test?user=root&password=taosdata"; + Properties connProps = new Properties(); + connProps.setProperty(TSDBDriver.PROPERTY_KEY_CHARSET, "UTF-8"); + connProps.setProperty(TSDBDriver.PROPERTY_KEY_LOCALE, "en_US.UTF-8"); + connProps.setProperty(TSDBDriver.PROPERTY_KEY_TIME_ZONE, "UTC-8"); + Connection conn = DriverManager.getConnection(jdbcUrl, connProps); + return conn; +} +``` 2. specify the firstEp and the secondEp in the configuration file taos.cfg - ```shell - # first fully qualified domain name (FQDN) for TDengine system - firstEp cluster_node1:6030 +```shell +# first fully qualified domain name (FQDN) for TDengine system +firstEp cluster_node1:6030 - # second fully qualified domain name (FQDN) for TDengine system, for cluster only - secondEp cluster_node2:6030 +# second fully qualified domain name (FQDN) for TDengine system, for cluster only +secondEp cluster_node2:6030 - # default system charset - # charset UTF-8 +# default system charset +# charset UTF-8 - # system locale - # locale en_US.UTF-8 - ``` +# system locale +# locale en_US.UTF-8 +``` In the above example, JDBC uses the client's configuration file to establish a connection to a hostname `cluster_node1`, port 6030, and a database named `test`. When the firstEp node in the cluster fails, JDBC attempts to connect to the cluster using secondEp. In TDengine, as long as one node in firstEp and secondEp is valid, the connection to the cluster can be established normally. -:::note The configuration file here refers to the configuration file on the machine where the application that calls the JDBC Connector is located, the default path is `/etc/taos/taos.cfg` on Linux, and the default path is `C://TDengine/cfg/taos.cfg` on Windows. -::: - @@ -196,11 +194,11 @@ There is no dependency on the client driver when Using a JDBC REST connection. C 2. jdbcUrl starting with "jdbc:TAOS-RS://". 3. use 6041 as the connection port. -The configuration parameters in the URL are as follows. +The configuration parameters in the URL are as follows: -- user: Login TDengine user name, default value 'root'. -- password: user login password, default value 'taosdata'. -- batchfetch: true: pull the result set in batch when executing the query; false: pull the result set row by row. The default value is false. batchfetch uses HTTP for data transfer. The JDBC REST connection supports bulk data pulling function in taos-jdbcdriver-2.0.38 and TDengine 2.4.0.12 and later versions. taos-jdbcdriver and TDengine transfer data via WebSocket connection. Compared with HTTP, WebSocket enables JDBC REST connection to support large data volume querying and improve query performance. +- user: Log in to the TDengine username. The default value is 'root'. +- password: User login password, the default value is 'taosdata'. +- batchfetch: true: pulls result sets in batches when executing queries; false: pulls result sets row by row. The default value is: false. batchfetch uses HTTP for data transfer. JDBC REST supports batch pulls. taos-jdbcdriver and TDengine transfer data via WebSocket connection. Compared with HTTP, WebSocket enables JDBC REST connection to support large data volume querying and improve query performance. - charset: specify the charset to parse the string, this parameter is valid only when set batchfetch to true. - batchErrorIgnore: true: when executing executeBatch of Statement, if one SQL execution fails in the middle, continue to execute the following SQL. false: no longer execute any statement after the failed SQL. The default value is: false. - httpConnectTimeout: REST connection timeout in milliseconds, the default value is 5000 ms. @@ -212,13 +210,13 @@ The configuration parameters in the URL are as follows. :::note -- Unlike the native connection method, the REST interface is stateless. When using the JDBC REST connection, you need to specify the database name of the table and super table in SQL. For example. +- Unlike the native connection method, the REST interface is stateless. When using the JDBC REST connection, you need to specify the database name of the table and super table in SQL. For example: ```sql INSERT INTO test.t1 USING test.weather (ts, temperature) TAGS('California.SanFrancisco') VALUES(now, 24.6); ``` -- Starting from taos-jdbcdriver-2.0.36 and TDengine 2.2.0.0, if dbname is specified in the URL, JDBC REST connections will use `/rest/sql/dbname` as the URL for REST requests by default, and there is no need to specify dbname in SQL. For example, if the URL is `jdbc:TAOS-RS://127.0.0.1:6041/test`, then the SQL can be executed: insert into test using weather(ts, temperature) tags('California.SanFrancisco') values(now, 24.6); +- If the dbname is specified in the URL, the JDBC REST connection uses /rest/sql/dbname as the default URL for RESTful requests. In this case, it is not necessary to specify the dbname in SQL. For example, if the URL is `jdbc:TAOS-RS://127.0.0.1:6041/test`, then the SQL can be executed: insert into test using weather(ts, temperature) tags('California.SanFrancisco') values(now, 24.6); ::: @@ -229,10 +227,10 @@ INSERT INTO test.t1 USING test.weather (ts, temperature) TAGS('California.SanFra In addition to getting the connection from the specified URL, you can use Properties to specify parameters when the connection is established. -**Note**: +Note: - The client parameter set in the application is process-level. If you want to update the parameters of the client, you need to restart the application. This is because the client parameter is a global parameter that takes effect only the first time the application is set. -- The following sample code is based on taos-jdbcdriver-2.0.36. +- The following sample code is based on taos-jdbcdriver-3.0.0. ```java public Connection getConn() throws Exception{ @@ -258,14 +256,14 @@ public Connection getRestConn() throws Exception{ } ``` -In the above example, a connection is established to `taosdemo.com`, port is 6030/6041, and database named `test`. The connection specifies the user name as `root` and the password as `taosdata` in the URL and specifies the character set, language environment, time zone, and whether to enable bulk fetching in the connProps. +In the above example, a connection is established to `taosdemo.com`, port is 6030/6041, and database named `test`. The connection specifies the user name as `root` and the password as `taosdata` in the URL and specifies the character set, language environment, time zone, and whether to enable bulk fetching in the connProps.The url specifies the user name as `root` and the password as `taosdata`. The configuration parameters in properties are as follows. - TSDBDriver.PROPERTY_KEY_USER: login TDengine user name, default value 'root'. - TSDBDriver.PROPERTY_KEY_PASSWORD: user login password, default value 'taosdata'. - TSDBDriver.PROPERTY_KEY_BATCH_LOAD: true: pull the result set in batch when executing query; false: pull the result set row by row. The default value is: false. -- TSDBDriver.PROPERTY_KEY_BATCH_ERROR_IGNORE: true: when executing executeBatch of Statement, if there is a SQL execution failure in the middle, continue to execute the following sq. false: no longer execute any statement after the failed SQL. The default value is: false. +- TSDBDriver.PROPERTY_KEY_BATCH_ERROR_IGNORE: true: when executing executeBatch of Statement, if there is a SQL execution failure in the middle, continue to execute the following sql. false: no longer execute any statement after the failed SQL. The default value is: false. - TSDBDriver.PROPERTY_KEY_CONFIG_DIR: only works when using JDBC native connection. Client configuration file directory path, default value `/etc/taos` on Linux OS, default value `C:/TDengine/cfg` on Windows OS. - TSDBDriver.PROPERTY_KEY_CHARSET: In the character set used by the client, the default value is the system character set. - TSDBDriver.PROPERTY_KEY_LOCALE: this only takes effect when using JDBC native connection. Client language environment, the default value is system current locale. @@ -273,7 +271,7 @@ The configuration parameters in properties are as follows. - TSDBDriver.HTTP_CONNECT_TIMEOUT: REST connection timeout in milliseconds, the default value is 5000 ms. It only takes effect when using JDBC REST connection. - TSDBDriver.HTTP_SOCKET_TIMEOUT: socket timeout in milliseconds, the default value is 5000 ms. It only takes effect when using JDBC REST connection and batchfetch is false. - TSDBDriver.PROPERTY_KEY_MESSAGE_WAIT_TIMEOUT: message transmission timeout in milliseconds, the default value is 3000 ms. It only takes effect when using JDBC REST connection and batchfetch is true. -- TSDBDriver.PROPERTY_KEY_USE_SSL: connecting Securely Using SSL. true: using SSL conneciton, false: not using SSL connection. It only takes effect when using using JDBC REST connection. +- TSDBDriver.PROPERTY_KEY_USE_SSL: connecting Securely Using SSL. true: using SSL conneciton, false: not using SSL connection. It only takes effect when using JDBC REST connection. For JDBC native connections, you can specify other parameters, such as log level, SQL length, etc., by specifying URL and Properties. For more detailed configuration, please refer to [Client Configuration](/reference/config/#Client-Only). ### Priority of configuration parameters @@ -305,7 +303,7 @@ stmt.executeUpdate("create table if not exists tb (ts timestamp, temperature int > **Note**: If you do not use `use db` to specify the database, all subsequent operations on the table need to add the database name as a prefix, such as db.tb. -### Insert data +### 插入数据 ```java // insert data @@ -320,7 +318,7 @@ System.out.println("insert " + affectedRows + " rows."); ### Querying data ```java -// query data +// insert data ResultSet resultSet = stmt.executeQuery("select * from tb"); Timestamp ts = null; @@ -355,25 +353,21 @@ try (Statement statement = connection.createStatement()) { } ``` -There are three types of error codes that the JDBC connector can report: - -- Error code of the JDBC driver itself (error code between 0x2301 and 0x2350) -- Error code of the native connection method (error code between 0x2351 and 0x2400) -- Error code of other TDengine function modules +There are three types of error codes that the JDBC connector can report: - Error code of the JDBC driver itself (error code between 0x2301 and 0x2350), - Error code of the native connection method (error code between 0x2351 and 0x2400), and - Error code of other TDengine function modules. For specific error codes, please refer to. - [TDengine Java Connector](https://github.com/taosdata/taos-connector-jdbc/blob/main/src/main/java/com/taosdata/jdbc/TSDBErrorNumbers.java) -- [TDengine_ERROR_CODE](https://github.com/taosdata/TDengine/blob/develop/src/inc/taoserror.h) + ### Writing data via parameter binding -TDengine's native JDBC connection implementation has significantly improved its support for data writing (INSERT) scenarios via bind interface with version 2.1.2.0 and later versions. Writing data in this way avoids the resource consumption of SQL syntax parsing, resulting in significant write performance improvements in many cases. +TDengine has significantly improved the bind APIs to support data writing (INSERT) scenarios. Writing data in this way avoids the resource consumption of SQL syntax parsing, resulting in significant write performance improvements in many cases. -**Note**. +**Note:** - JDBC REST connections do not currently support bind interface -- The following sample code is based on taos-jdbcdriver-2.0.36 +- The following sample code is based on taos-jdbcdriver-3.0.0 - The setString method should be called for binary type data, and the setNString method should be called for nchar type data - both setString and setNString require the user to declare the width of the corresponding column in the size parameter of the table definition @@ -578,7 +572,7 @@ public class ParameterBindingDemo { // set table name pstmt.setTableName("t5_" + i); // set tags - pstmt.setTagNString(0, "California-abc"); + pstmt.setTagNString(0, "California.SanFrancisco"); // set columns ArrayList tsList = new ArrayList<>(); @@ -589,7 +583,7 @@ public class ParameterBindingDemo { ArrayList f1List = new ArrayList<>(); for (int j = 0; j < numOfRow; j++) { - f1List.add("California-abc"); + f1List.add("California.LosAngeles"); } pstmt.setNString(1, f1List, BINARY_COLUMN_SIZE); @@ -636,12 +630,12 @@ public void setNString(int columnIndex, ArrayList list, int size) throws ### Schemaless Writing -Starting with version 2.2.0.0, TDengine has added the ability to perform schemaless writing. It is compatible with InfluxDB's Line Protocol, OpenTSDB's telnet line protocol, and OpenTSDB's JSON format protocol. See [schemaless writing](/reference/schemaless/) for details. +TDengine supports schemaless writing. It is compatible with InfluxDB's Line Protocol, OpenTSDB's telnet line protocol, and OpenTSDB's JSON format protocol. For more information, see [Schemaless Writing](../../schemaless). -**Note**. +Note: - JDBC REST connections do not currently support schemaless writes -- The following sample code is based on taos-jdbcdriver-2.0.36 +- The following sample code is based on taos-jdbcdriver-3.0.0 ```java public class SchemalessInsertTest { @@ -672,59 +666,137 @@ public class SchemalessInsertTest { } ``` -### Subscriptions +### Data Subscription The TDengine Java Connector supports subscription functionality with the following application API. -#### Create subscriptions +#### Create a Topic ```java -TSDBSubscribe sub = ((TSDBConnection)conn).subscribe("topicname", "select * from meters", false); +Connection connection = DriverManager.getConnection(url, properties); +Statement statement = connection.createStatement(); +statement.executeUpdate("create topic if not exists topic_speed as select ts, speed from speed_table"); ``` The three parameters of the `subscribe()` method have the following meanings. -- topicname: the name of the subscribed topic. This parameter is the unique identifier of the subscription. -- sql: the query statement of the subscription. This statement can only be a `select` statement. Only original data can be queried, and you can query the data only temporal order. -- restart: if the subscription already exists, whether to restart or continue the previous subscription +- topic_speed: the subscribed topic (name). This is the unique identifier of the subscription. +- sql: the query statement of the subscription which can only be a _select_ statement. Only the original data should be queried, and data can only be queried in temporal order.. + +The preceding example uses the SQL statement `select ts, speed from speed_table` and creates a subscription named `topic_speed`. -The above example will use the SQL command `select * from meters` to create a subscription named `topicname`. If the subscription exists, it will continue the progress of the previous query instead of consuming all the data from the beginning. +#### Create a Consumer + +```java +Properties config = new Properties(); +config.setProperty("enable.auto.commit", "true"); +config.setProperty("group.id", "group1"); +config.setProperty("value.deserializer", "com.taosdata.jdbc.tmq.ConsumerTest.ResultDeserializer"); + +TaosConsumer consumer = new TaosConsumer<>(config); +``` + +- enable.auto.commit: Specifies whether to commit automatically. +- group.id: consumer: Specifies the group that the consumer is in. +- value.deserializer: To deserialize the results, you can inherit `com.taosdata.jdbc.tmq.ReferenceDeserializer` and specify the result set bean. You can also inherit `com.taosdata.jdbc.tmq.Deserializer` and perform custom deserialization based on the SQL result set. +- For more information, see [Consumer Parameters](../../../develop/tmq). #### Subscribe to consume data ```java -int total = 0; while(true) { - TSDBResultSet rs = sub.consume(); - int count = 0; - while(rs.next()) { - count++; - } - total += count; - System.out.printf("%d rows consumed, total %d\n", count, total); - Thread.sleep(1000); + ConsumerRecords records = consumer.poll(Duration.ofMillis(100)); + for (ResultBean record : records) { + process(record); + } } ``` -The `consume()` method returns a result set containing all new data from the last `consume()`. Be sure to choose a reasonable frequency for calling `consume()` as needed (e.g. `Thread.sleep(1000)` in the example). Otherwise, it will cause unnecessary stress on the server-side. +`poll` obtains one message each time it is run. #### Close subscriptions ```java -sub.close(true); +// Unsubscribe +consumer.unsubscribe(); +// Close consumer +consumer.close() ``` -The `close()` method closes a subscription. If its argument is `true` it means that the subscription progress information is retained, and the subscription with the same name can be created to continue consuming data; if it is `false` it does not retain the subscription progress. +For more information, see [Data Subscription](../../../develop/tmq). -### Closing resources +### Usage examples ```java -resultSet.close(); -stmt.close(); -conn.close(); -``` +public abstract class ConsumerLoop { + private final TaosConsumer consumer; + private final List topics; + private final AtomicBoolean shutdown; + private final CountDownLatch shutdownLatch; + + public ConsumerLoop() throws SQLException { + Properties config = new Properties(); + config.setProperty("msg.with.table.name", "true"); + config.setProperty("enable.auto.commit", "true"); + config.setProperty("group.id", "group1"); + config.setProperty("value.deserializer", "com.taosdata.jdbc.tmq.ConsumerTest.ConsumerLoop$ResultDeserializer"); + + this.consumer = new TaosConsumer<>(config); + this.topics = Collections.singletonList("topic_speed"); + this.shutdown = new AtomicBoolean(false); + this.shutdownLatch = new CountDownLatch(1); + } + + public abstract void process(ResultBean result); + + public void pollData() throws SQLException { + try { + consumer.subscribe(topics); + + while (!shutdown.get()) { + ConsumerRecords records = consumer.poll(Duration.ofMillis(100)); + for (ResultBean record : records) { + process(record); + } + } + consumer.unsubscribe(); + } finally { + consumer.close(); + shutdownLatch.countDown(); + } + } + + public void shutdown() throws InterruptedException { + shutdown.set(true); + shutdownLatch.await(); + } + + public static class ResultDeserializer extends ReferenceDeserializer { -> **Be sure to close the connection**, otherwise, there will be a connection leak. + } + + public static class ResultBean { + private Timestamp ts; + private int speed; + + public Timestamp getTs() { + return ts; + } + + public void setTs(Timestamp ts) { + this.ts = ts; + } + + public int getSpeed() { + return speed; + } + + public void setSpeed(int speed) { + this.speed = speed; + } + } +} +``` ### Use with connection pool @@ -755,7 +827,7 @@ Example usage is as follows. //query or insert // ... - connection.close(); // put back to connection pool + connection.close(); // put back to conneciton pool } ``` @@ -787,26 +859,12 @@ public static void main(String[] args) throws Exception { //query or insert // ... - connection.close(); // put back to connection pool + connection.close(); // put back to conneciton pool } ``` > For more questions about using druid, please see [Official Instructions](https://github.com/alibaba/druid). -**Caution:** - -- TDengine `v1.6.4.1` provides a special function `select server_status()` for heartbeat detection, so it is recommended to use `select server_status()` for Validation Query when using connection pooling. - -As you can see below, `select server_status()` returns `1` on successful execution. - -```sql -taos> select server_status(); -server_status()| -================ -1 | -Query OK, 1 row(s) in set (0.000141s) -``` - ### More sample programs The source code of the sample application is under `TDengine/examples/JDBC`: @@ -817,12 +875,13 @@ The source code of the sample application is under `TDengine/examples/JDBC`: - SpringJdbcTemplate: using taos-jdbcdriver in Spring JdbcTemplate. - mybatisplus-demo: using taos-jdbcdriver in Springboot + Mybatis. -Please refer to: [JDBC example](https://github.com/taosdata/TDengine/tree/develop/examples/JDBC) +[JDBC example](https://github.com/taosdata/TDengine/tree/3.0/examples/JDBC) ## Recent update logs | taos-jdbcdriver version | major changes | | :---------------------: | :--------------------------------------------: | +| 3.0.0 | Support for TDengine 3.0 | | 2.0.39 - 2.0.40 | Add REST connection/request timeout parameters | | 2.0.38 | JDBC REST connections add bulk pull function | | 2.0.37 | Support json tags | @@ -842,7 +901,7 @@ Please refer to: [JDBC example](https://github.com/taosdata/TDengine/tree/develo **Solution**: On Windows you can copy `C:\TDengine\driver\taos.dll` to the `C:\Windows\System32` directory, on Linux the following soft link will be created `ln -s /usr/local/taos/driver/libtaos.so.x.x.x.x /usr/lib/libtaos.so` will work. -3. java.lang.UnsatisfiedLinkError: taos.dll Can't load AMD 64 bit on an IA 32-bit platform +3. java.lang.UnsatisfiedLinkError: taos.dll Can't load AMD 64 bit on a IA 32-bit platform **Cause**: Currently, TDengine only supports 64-bit JDK. @@ -850,11 +909,11 @@ Please refer to: [JDBC example](https://github.com/taosdata/TDengine/tree/develo 4. java.lang.NoSuchMethodError: setByteArray - **Cause**: taos-jdbcdriver version 3.* only supports TDengine 3.0 or above. + **Cause**: taos-jbdcdriver 3.* only supports TDengine 3.0 and later. - **Solution**: connect TDengine 2.* using taos-jdbcdriver 2.* version. + **Solution**: Use taos-jdbcdriver 2.* with your TDengine 2.* deployment. -For other questions, please refer to [FAQ](/train-faq/faq) +For additional troubleshooting, see [FAQ](../../../train-faq/faq). ## API Reference diff --git a/docs/en/14-reference/03-connector/node.mdx b/docs/en/14-reference/03-connector/node.mdx index 8f586acde4848af71efcb23358be1f8486cedb8e..d1700444351d6f54f799a1c84674735800959c3c 100644 --- a/docs/en/14-reference/03-connector/node.mdx +++ b/docs/en/14-reference/03-connector/node.mdx @@ -8,20 +8,20 @@ title: TDengine Node.js Connector import Tabs from "@theme/Tabs"; import TabItem from "@theme/TabItem"; -import Preparation from "./_preparation.mdx"; +import Preparition from "./_preparition.mdx"; import NodeInsert from "../../07-develop/03-insert-data/_js_sql.mdx"; import NodeInfluxLine from "../../07-develop/03-insert-data/_js_line.mdx"; import NodeOpenTSDBTelnet from "../../07-develop/03-insert-data/_js_opts_telnet.mdx"; import NodeOpenTSDBJson from "../../07-develop/03-insert-data/_js_opts_json.mdx"; import NodeQuery from "../../07-develop/04-query-data/_js.mdx"; -`td2.0-connector` and `td2.0-rest-connector` are the official Node.js language connectors for TDengine. Node.js developers can develop applications to access TDengine instance data. +`@tdengine/client` and `@tdengine/rest` are the official Node.js connectors. Node.js developers can develop applications to access TDengine instance data. Note: The connectors for TDengine 3.0 are different than those for TDengine 2.x. The new connectors do not support TDengine 2.x. -`td2.0-connector` is a **native connector** that connects to TDengine instances via the TDengine client driver (taosc) and supports data writing, querying, subscriptions, schemaless writing, and bind interface. The `td2.0-rest-connector` is a **REST connector** that connects to TDengine instances via the REST interface provided by taosAdapter. The REST connector can run on any platform, but performance is slightly degraded, and the interface implements a somewhat different set of functional features than the native interface. +`@tdengine/client` is **native connection**, which connects to TDengine instances natively through the TDengine client driver (taosc), supporting data writing, querying, subscriptions, schemaless writing, and bind interface. `@tdengine/rest` is the **REST connection**, which connects to TDengine instances via the REST interface provided by taosAdapter. The REST connector can run on any platform, but performance is slightly degraded, and the interface implements a somewhat different set of functional features than the native interface. -The Node.js connector source code is hosted on [GitHub](https://github.com/taosdata/taos-connector-node). +The source code for the Node.js connectors is located on [GitHub](https://github.com/taosdata/taos-connector-node/tree/3.0). -## Supported Platforms +## Supported platforms The platforms supported by the native connector are the same as those supported by the TDengine client driver. The REST connector supports all platforms that can run Node.js. @@ -34,22 +34,22 @@ Please refer to [version support list](/reference/connector#version-support) ### Native connectors -1. connection management -2. general query -3. continuous query -4. parameter binding -5. subscription function +1. Connection Management +2. General Query +3. Continuous Query +4. Parameter Binding +5. Subscription 6. Schemaless ### REST Connector -1. connection management -2. general queries -3. Continuous query +1. Connection Management +2. General Query +3. Continuous Query -## Installation steps +## Installation Steps -### Pre-installation +### Pre-installation preparation - Install the Node.js development environment - If you are using the REST connector, skip this step. However, if you use the native connector, please install the TDengine client driver. Please refer to [Install Client Driver](/reference/connector#Install-Client-Driver) for more details. We use [node-gyp](https://github.com/nodejs/node-gyp) to interact with TDengine instances and also need to install some dependencies mentioned below depending on the specific OS. @@ -58,7 +58,7 @@ Please refer to [version support list](/reference/connector#version-support) - `python` (recommended for `v2.7` , `v3.x.x` currently not supported) -- `td2.0-connector` 2.0.6 supports Node.js LTS v10.9.0 or later, Node.js LTS v12.8.0 or later; 2.0.5 and earlier support Node.js LTS v10.x versions. Other versions may have package compatibility issues +- `@tdengine/client` 3.0.0 supports Node.js LTS v10.9.0 or later and Node.js LTS v12.8.0 or later. Older versions may be incompatible. - `make` - C compiler, [GCC](https://gcc.gnu.org) v4.8.5 or higher @@ -90,32 +90,32 @@ If using ARM64 Node.js on Windows 10 ARM, you must add "Visual C++ compilers and ```bash -npm install td2.0-connector +npm install @tdengine/client ``` ```bash -npm i td2.0-rest-connector +npm install @tdengine/rest ``` -### Installation verification +### Verify After installing the TDengine client, use the `nodejsChecker.js` program to verify that the current environment supports Node.js access to TDengine. Verification in details: -- Create a new installation verification directory, e.g. `~/tdengine-test`, and download the [nodejsChecker.js source code](https://github.com/taosdata/TDengine/tree/develop/examples/nodejs/) from GitHub. to the work directory. +- Create an installation test folder such as `~/tdengine-test`. Download the [nodejsChecker.js source code](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/nodejsChecker.js) to your local machine. - Execute the following command from the command-line. ```bash npm init -y -npm install td2.0-connector +npm install @tdengine/client node nodejsChecker.js host=localhost ``` @@ -126,13 +126,13 @@ node nodejsChecker.js host=localhost Please choose to use one of the connectors. - + -Install and refer to `td2.0-connector` package: +Install and import the `@tdengine/client` package. ```javascript //A cursor also needs to be initialized in order to interact with TDengine from Node.js. -const taos = require("td2.0-connector"); +const taos = require("@tdengine/client"); var conn = taos.connect({ host: "127.0.0.1", user: "root", @@ -149,12 +149,12 @@ conn.close(); -Install and refer to `td2.0-rest-connector`package: +Install and import the `@tdengine/rest` package. ```javascript //A cursor also needs to be initialized in order to interact with TDengine from Node.js. -import { options, connect } from "td2.0-rest-connector"; -options.path = "/rest/sqlt"; +import { options, connect } from "@tdengine/rest"; +options.path = "/rest/sql"; // set host options.host = "localhost"; // set other options like user/passwd @@ -170,50 +170,47 @@ let cursor = conn.cursor(); ### Write data -#### SQL Writing +#### SQL Write -#### InfluxDB line protocol writing +#### InfluxDB line protocol write -#### OpenTSDB Telnet line protocol writing +#### OpenTSDB Telnet line protocol write -#### OpenTSDB JSON line protocol writing +#### OpenTSDB JSON line protocol write -### Query data +### Querying data -## More Sample Programs + +## More sample programs | Sample Programs | Sample Program Description | | --------------------------------------------------------------------------------------------------------------------------------- --------- | -------------------------------------- | -| [connection](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/cursorClose.js) | Example of establishing a connection. | -| [stmtBindBatch](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/stmtBindParamBatchSample.js) | Example of binding a multi-line parameter Example of inserting. | -| [stmtBind](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/stmtBindParamSample.js) | Example of a line-by-line bind parameter insertion. | -| [stmtBindSingleParamBatch](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/) stmtBindSingleParamBatchSample.js) | Example of binding parameters by column. | -| [stmtUseResult](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/stmtUseResultSample.js) | Example of a bound parameter query. | -| [json tag](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/testJsonTag.js) | Example of using Json tag. | -| [Nanosecond](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/testNanoseconds.js) | An example of using timestamps with nanosecond precision. | -| [Microsecond](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/testMicroseconds.js) | Example of using microsecond timestamp. | -| [schemless insert](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/testSchemalessInsert.js) | schemless Example of a schemless insert. | -| [subscribe](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/testSubscribe.js) | Example of using subscribe. | -| [asyncQuery](https://github.com/taosdata/taos-connector-node/tree/develop/nodejs/examples/tset.js) | An example of using asynchronous queries. | -| [REST](https://github.com/taosdata/taos-connector-node/blob/develop/typescript-rest/example/example.ts) | An example of using TypeScript with REST connections. | +| [basicUse](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/queryExample.js) | Basic operations such as establishing connections and running SQl commands. | +| [stmtBindBatch](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/bindParamBatch.js) | Binding multi-line parameter insertion. | | +| [stmtBindSingleParamBatch](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/bindSingleParamBatch.js) | Columnar binding parameter insertion | +| [stmtQuery](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/stmtQuery.js) | Binding parameter query | +| [schemless insert](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/schemaless.js) | Schemaless insert | +| [TMQ](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/tmq.js) | Using data subscription | +| [asyncQuery](https://github.com/taosdata/taos-connector-node/blob/3.0/nodejs/examples/asyncQueryExample.js) | Using asynchronous queries | +| [REST](https://github.com/taosdata/taos-connector-node/blob/3.0/typescript-rest/example/example.ts) | Using TypeScript with the REST connector | + +## Usage limitations + +`@tdengine/client` 3.0.0 supports Node.js LTS v12.8.0 to 12.9.1 and 10.9.0 to 10.20.0. -## Usage restrictions -Node.js Connector >= v2.0.6 currently supports node versions >=v12.8.0 <= v12.9.1 || >=v10.20.0 <= v10.9.0; v10.x versions are supported in 2.0.5 and earlier, other versions may have package compatibility issues. -## Other notes -See [video tutorial](https://www.taosdata.com/blog/2020/11/11/1957.html) for the Node.js connector usage. ## Frequently Asked Questions @@ -223,26 +220,34 @@ See [video tutorial](https://www.taosdata.com/blog/2020/11/11/1957.html) for the sudo systemctl start taosadapter ``` -2. "Unable to establish connection", "Unable to resolve FQDN" +2. Node.js versions + + `@tdengine/client` supports Node.js v10.9.0 to 10.20.0 and 12.8.0 to 12.9.1. + +3. "Unable to establish connection", "Unable to resolve FQDN" Usually, the root cause is an incorrect FQDN configuration. You can refer to this section in the [FAQ](https://docs.tdengine.com/2.4/train-faq/faq/#2-how-to-handle-unable-to-establish-connection) to troubleshoot. -## Important Updates +## Important update records ### Native connectors -| td2.0-connector version | description | -| -------------------- | ---------------------------------------------------------------- | -| 2.0.12 | Fix bug with cursor.close() error. | 2.0.12 | Fix bug with cursor.close() error. -| 2.0.11 | Support for binding parameters, json tag, schemaless interface, etc. | -| 2.0.10 | Support connection management, general query, continuous query, get system information, subscribe function, etc. | ### REST Connector - +| package name | version | TDengine version | Description | +|------------------|---------|---------------------|------------------------------------------------------------------| +| @tdengine/client | 3.0.0 | 3.0.0 | Supports TDengine 3.0. Not compatible with TDengine 2.x. | +| td2.0-connector | 2.0.12 | 2.4.x；2.5.x；2.6.x | Fixed cursor.close() bug. | +| td2.0-connector | 2.0.11 | 2.4.x；2.5.x；2.6.x | Supports parameter binding, JSON tags and schemaless interface | +| td2.0-connector | 2.0.10 | 2.4.x；2.5.x；2.6.x | Supports connection management, standard queries, connection queries, system information, and data subscription | ### REST Connector -| td2.0-rest-connector version | Description | -| ------------------------- | ---------------------------------------------------------------- | -| 1.0.3 | Support connection management, general query, get system information, error message, continuous query, etc. |# API Reference +| package name | version | TDengine version | Description | +|----------------------|---------|---------------------|---------------------------------------------------------------------------| +| @tdengine/rest | 3.0.0 | 3.0.0 | Supports TDengine 3.0. Not compatible with TDengine 2.x. | +| td2.0-rest-connector | 1.0.7 | 2.4.x；2.5.x；2.6.x | Removed default port 6041。 | +| td2.0-rest-connector | 1.0.6 | 2.4.x；2.5.x；2.6.x | Fixed affectRows bug with create, insert, update, and alter. | +| td2.0-rest-connector | 1.0.5 | 2.4.x；2.5.x；2.6.x | Support cloud token | +| td2.0-rest-connector | 1.0.3 | 2.4.x；2.5.x；2.6.x | Supports connection management, standard queries, system information, error information, and continuous queries | ## API Reference -[API Reference](https://docs.taosdata.com/api/td2.0-connector/) +[API Reference](https://docs.taosdata.com/api/td2.0-connector/) \ No newline at end of file diff --git a/docs/en/14-reference/03-connector/python.mdx b/docs/en/14-reference/03-connector/python.mdx index 04eb2e57d4455a83b62633ecb988cb64bf837fea..1a15da62d6cd94a95b62208cdf76ab6b5df1230a 100644 --- a/docs/en/14-reference/03-connector/python.mdx +++ b/docs/en/14-reference/03-connector/python.mdx @@ -8,14 +8,14 @@ description: "taospy is the official Python connector for TDengine. taospy provi import Tabs from "@theme/Tabs"; import TabItem from "@theme/TabItem"; -`taospy` is the official Python connector for TDengine. `taospy` provides a rich set of APIs that makes it easy for Python applications to access TDengine. `taospy` wraps both the [native interface](/reference/connector/cpp) and [REST interface](/reference/rest-api) of TDengine, which correspond to the `taos` and `taosrest` modules of the `taospy` package, respectively. +`taospy is the official Python connector for TDengine. taospy provides a rich API that makes it easy for Python applications to use TDengine. `taospy` wraps both the [native interface](/reference/connector/cpp) and [REST interface](/reference/rest-api) of TDengine, which correspond to the `taos` and `taosrest` modules of the `taospy` package, respectively. In addition to wrapping the native and REST interfaces, `taospy` also provides a set of programming interfaces that conforms to the [Python Data Access Specification (PEP 249)](https://peps.python.org/pep-0249/). It is easy to integrate `taospy` with many third-party tools, such as [SQLAlchemy](https://www.sqlalchemy.org/) and [pandas](https://pandas.pydata.org/). The direct connection to the server using the native interface provided by the client driver is referred to hereinafter as a "native connection"; the connection to the server using the REST interface provided by taosAdapter is referred to hereinafter as a "REST connection". The source code for the Python connector is hosted on [GitHub](https://github.com/taosdata/taos-connector-python). -## Supported Platforms +## Supported platforms - The [supported platforms](/reference/connector/#supported-platforms) for the native connection are the same as the ones supported by the TDengine client. - REST connections are supported on all platforms that can run Python. @@ -35,7 +35,6 @@ We recommend using the latest version of `taospy`, regardless of the version of 1. Install Python. Python >= 3.6 is recommended. If Python is not available on your system, refer to the [Python BeginnersGuide](https://wiki.python.org/moin/BeginnersGuide/Download) to install it. 2. Install [pip](https://pypi.org/project/pip/). In most cases, the Python installer comes with the pip utility. If not, please refer to [pip documentation](https://pip.pypa.io/en/stable/installation/) to install it. - If you use a native connection, you will also need to [Install Client Driver](/reference/connector#Install-Client-Driver). The client install package includes the TDengine client dynamic link library (`libtaos.so` or `taos.dll`) and the TDengine CLI. ### Install via pip @@ -80,7 +79,7 @@ pip3 install git+https://github.com/taosdata/taos-connector-python.git -### Installation verification +### Verify @@ -110,10 +109,11 @@ If you have multiple versions of Python on your system, you may have various `pi C:\> pip3 install taospy Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Requirement already satisfied: taospy in c:\users\username\appdata\local\programs\python\python310\lib\site-packages (2.3.0) +``` ::: -## Establish connection +## Establishing a connection ### Connectivity testing @@ -202,12 +202,12 @@ All arguments to the `connect()` function are optional keyword arguments. The fo - `url`： The URL of taosAdapter REST service. The default is . - `user`: TDengine user name. The default is `root`. - `password`: TDengine user password. The default is `taosdata`. -- `timeout`: HTTP request timeout in seconds. The default is `socket._GLOBAL_DEFAULT_TIMEOUT`. Usually, no configuration is needed. +- `timeout`: HTTP request timeout. Enter a value in seconds. The default is `socket._GLOBAL_DEFAULT_TIMEOUT`. Usually, no configuration is needed. -## Sample program +## Example program ### Basic Usage @@ -255,12 +255,12 @@ The TaosCursor class uses native connections for write and query operations. In ##### Use of TaosRestCursor class -The ``TaosRestCursor`` class is an implementation of the PEP249 Cursor interface. +The `TaosRestCursor` class is an implementation of the PEP249 Cursor interface. ```python title="Use of TaosRestCursor" {{#include docs/examples/python/connect_rest_examples.py:basic}} ``` -- `cursor.execute` : Used to execute arbitrary SQL statements. +- `cursor.execute`: Used to execute arbitrary SQL statements. - `cursor.rowcount` : For write operations, returns the number of successful rows written. For query operations, returns the number of rows in the result set. - `cursor.description` : Returns the description of the field. Please refer to [TaosRestCursor](https://docs.taosdata.com/api/taospy/taosrest/cursor.html) for the specific format of the description information. @@ -274,6 +274,8 @@ The `RestClient` class is a direct wrapper for the [REST API](/reference/rest-ap For a more detailed description of the `sql()` method, please refer to [RestClient](https://docs.taosdata.com/api/taospy/taosrest/restclient.html). + + @@ -304,8 +306,7 @@ For a more detailed description of the `sql()` method, please refer to [RestClie | [bind_row.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/bind-row.py) | bind_row.py | [insert_lines.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/insert-lines.py) | InfluxDB line protocol writing | | [json_tag.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/json-tag.py) | Use JSON type tags | -| [subscribe-async.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/subscribe-async.py) | Asynchronous subscription | -| [subscribe-sync.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/subscribe-sync.py) | synchronous-subscribe | +| [tmq.py](https://github.com/taosdata/taos-connector-python/blob/main/examples/tmq.py) | TMQ subscription | ## Other notes @@ -324,22 +325,15 @@ Due to the current imperfection of Python's nanosecond support (see link below), 1. https://stackoverflow.com/questions/10611328/parsing-datetime-strings-containing-nanoseconds 2. https://www.python.org/dev/peps/pep-0564/ - -## Frequently Asked Questions - -Welcome to [ask questions or report questions](https://github.com/taosdata/taos-connector-python/issues). - ## Important Update -| Connector version | Important Update | Release date | -| ---------- | --------------------------------------------------------------------------------- | ---------- | -| 2.3.1 | 1. support TDengine REST API
    2. remove support for Python version below 3.6 | 2022-04-28 | -| 2.2.5 | support timezone option when connect | 2022-04-13 | -| 2.2.2 | support sqlalchemy dialect plugin | 2022-03-28 | - [**Release Notes**] (https://github.com/taosdata/taos-connector-python/releases) ## API Reference - [taos](https://docs.taosdata.com/api/taospy/taos/) - [taosrest](https://docs.taosdata.com/api/taospy/taosrest) + +## Frequently Asked Questions + +Welcome to [ask questions or report questions](https://github.com/taosdata/taos-connector-python/issues). diff --git a/docs/en/14-reference/03-connector/rust.mdx b/docs/en/14-reference/03-connector/rust.mdx index ab06f72069e29361a033f724308d950afe6e8d42..e9b16ba94d1db27d0571aad24d04492aeea32fb8 100644 --- a/docs/en/14-reference/03-connector/rust.mdx +++ b/docs/en/14-reference/03-connector/rust.mdx @@ -8,43 +8,45 @@ title: TDengine Rust Connector import Tabs from '@theme/Tabs'; import TabItem from '@theme/TabItem'; -import Preparation from "./_preparation.mdx" +import Preparition from "./_preparition.mdx" import RustInsert from "../../07-develop/03-insert-data/_rust_sql.mdx" import RustBind from "../../07-develop/03-insert-data/_rust_stmt.mdx" import RustQuery from "../../07-develop/04-query-data/_rust.mdx" -[`taos`][taos] is the official Rust language connector for TDengine. Rust developers can develop applications to access the TDengine instance data. +[![Crates.io](https://img.shields.io/crates/v/taos)](https://crates.io/crates/taos) ![Crates.io](https://img.shields.io/crates/d/taos) [![docs.rs](https://img.shields.io/docsrs/taos)](https://docs.rs/taos) -Rust connector provides two ways to establish connections. One is the **Native Connection**, which connects to TDengine instances via the TDengine client driver (taosc). The other is **Websocket connection**, which connects to TDengine instances via taosAdapter service. +`taos` is the official Rust connector for TDengine. Rust developers can develop applications to access the TDengine instance data. -The source code is hosted on [taosdata/taos-connector-rust](https://github.com/taosdata/taos-connector-rust). +`taos` provides two ways to establish connections. One is the **Native Connection**, which connects to TDengine instances via the TDengine client driver (taosc). The other is the **WebSocket connection**, which connects to TDengine instances via the WebSocket interface provided by taosAdapter. You can specify a connection type with Cargo features. By default, both types are supported. The Websocket connection can be used on any platform. The native connection can be used on any platform that the TDengine Client supports. + +The source code for the Rust connectors is located on [GitHub](https://github.com/taosdata/taos-connector-rust). ## Supported platforms -The platforms supported by native connections are the same as those supported by the TDengine client driver. -REST connections are supported on all platforms that can run Rust. +Native connections are supported on the same platforms as the TDengine client driver. +Websocket connections are supported on all platforms that can run Go. ## Version support -Please refer to [version support list](/reference/connector#version-support). +Please refer to [version support list](/reference/connector#version-support) The Rust Connector is still under rapid development and is not guaranteed to be backward compatible before 1.0. We recommend using TDengine version 3.0 or higher to avoid known issues. ## Installation -### Pre-installation +### Pre-installation preparation * Install the Rust development toolchain * If using the native connection, please install the TDengine client driver. Please refer to [install client driver](/reference/connector#install-client-driver) -### Add dependencies +# Add taos dependency -Add the dependency to the [Rust](https://rust-lang.org) project as follows, depending on the connection method selected. +Depending on the connection method, add the [taos][taos] dependency in your Rust project as follows: - + -Add [taos] to the `Cargo.toml` file. +In `cargo.toml`, add [taos][taos]: ```toml [dependencies] @@ -53,9 +55,10 @@ taos = "*" ``` - -Add [taos] to the `Cargo.toml` file. + + +In `cargo.toml`, add [taos][taos] and enable the native feature: ```toml [dependencies] @@ -65,7 +68,7 @@ taos = { version = "*", default-features = false, features = ["native"] } -Add [taos] to the `Cargo.toml` file and enable the `ws` feature. +In `cargo.toml`, add [taos][taos] and enable the ws feature: ```toml [dependencies] @@ -75,15 +78,15 @@ taos = { version = "*", default-features = false, features = ["ws"] } -## Create a connection +## Establishing a connection -In rust connector, we use a DSN connection string as a connection builder. For example, +[TaosBuilder] creates a connection constructor through the DSN connection description string. ```rust let builder = TaosBuilder::from_dsn("taos://")?; ``` -You can now use connection client to create the connection. +You can now use this object to create the connection. ```rust let conn = builder.build()?; @@ -96,9 +99,7 @@ let conn1 = builder.build()?; let conn2 = builder.build()?; ``` -DSN is short for **D**ata **S**ource **N**ame string - [a data structure used to describe a connection to a data source](https://en.wikipedia.org/wiki/Data_source_name). - -A common DSN is basically constructed as this: +The structure of the DSN description string is as follows: ```text [+]://[[:@]:][/][?=[&=]] @@ -106,31 +107,31 @@ A common DSN is basically constructed as this: |driver| protocol | | username | password | host | port | database | params | ``` -- **Driver**: the main entrypoint to a processer. **Required**. In Rust connector, the supported driver names are listed here: - - **taos**: the legacy TDengine connection data source. - - **tmq**: subscription data source from TDengine. - - **http/ws**: use websocket protocol via `ws://` scheme. - - **https/wss**: use websocket protocol via `wss://` scheme. -- **Protocol**: the additional information appended to driver, which can be be used to support different kind of data sources. By default, leave it empty for native driver(only under feature "native"), and `ws/wss` for websocket driver (only under feature "ws"). **Optional**. -- **Username**: as its definition, is the username to the connection. **Optional**. -- **Password**: the password of the username. **Optional**. -- **Host**: address host to the datasource. **Optional**. -- **Port**: address port to the datasource. **Optional**. -- **Database**: database name or collection name in the datasource. **Optional**. -- **Params**: a key-value map for any other informations to the datasource. **Optional**. - -Here is a simple DSN connection string example: +The parameters are described as follows: + +- **driver**: Specify a driver name so that the connector can choose which method to use to establish the connection. Supported driver names are as follows: + - **taos**: Table names use the TDengine connector driver. + - **tmq**: Use the TMQ to subscribe to data. + - **http/ws**: Use Websocket to establish connections. + - **https/wss**: Use Websocket to establish connections, and enable SSL/TLS. +- **protocol**: Specify which connection method to use. For example, `taos+ws://localhost:6041` uses Websocket to establish connections. +- **username/password**: Username and password used to create connections. +- **host/port**: Specifies the server and port to establish a connection. If you do not specify a hostname or port, native connections default to `localhost:6030` and Websocket connections default to `localhost:6041`. +- **database**: Specify the default database to connect to. +- **params**：Optional parameters. + +A sample DSN description string is as follows: ```text taos+ws://localhost:6041/test ``` -which means connect `localhost` with port `6041` via `ws` protocol, and make `test` as the default database. +This indicates that the Websocket connection method is used on port 6041 to connect to the server localhost and use the database `test` by default. -So that you can use DSN to specify connection protocol at runtime: +You can create DSNs to connect to servers in your environment. ```rust -use taos::*; // use it like a `prelude` mod, we need some traits at next. +use taos::*; // use native protocol. let builder = TaosBuilder::from_dsn("taos://localhost:6030")?; @@ -140,7 +141,7 @@ let conn1 = builder.build(); let conn2 = TaosBuilder::from_dsn("taos+ws://localhost:6041")?; ``` -After connected, you can perform the following operations on the database. +After the connection is established, you can perform operations on your database. ```rust async fn demo(taos: &Taos, db: &str) -> Result<(), Error> { @@ -179,7 +180,7 @@ async fn demo(taos: &Taos, db: &str) -> Result<(), Error> { } ``` -Rust connector provides two kinds of ways to fetch data: +There are two ways to query data: Using built-in types or the [serde](https://serde.rs) deserialization framework. ```rust // Query option 1, use rows stream. @@ -225,41 +226,41 @@ Rust connector provides two kinds of ways to fetch data: -#### Stmt bind +#### STMT Write ### Query data -| + ## API Reference -### Connector builder +### Connector Constructor -Use DSN to directly construct a TaosBuilder object. +You create a connector constructor by using a DSN. ```rust -let builder = TaosBuilder::from_dsn("")? ; +let cfg = TaosBuilder::default().build()?; ``` -Use `builder` to create many connections: +You use the builder object to create multiple connections. ```rust let conn: Taos = cfg.build(); ``` -### Connection pool +### Connection pooling -In complex applications, we recommend enabling connection pools. Connection pool for [taos] is implemented using [r2d2] by enabling "r2d2" feature. +In complex applications, we recommend enabling connection pools. [taos] implements connection pools based on [r2d2]. -Basically, a connection pool with default parameters can be generated as: +As follows, a connection pool with default parameters can be generated. ```rust let pool = TaosBuilder::from_dsn(dsn)?.pool()?; ``` -You can set the connection pool parameters using the `PoolBuilder`. +You can set the same connection pool parameters using the connection pool's constructor. ```rust let dsn = "taos://localhost:6030"; @@ -273,23 +274,23 @@ let opts = PoolBuilder::new() let pool = TaosBuilder::from_dsn(dsn)?.with_pool_builder(opts)?; ``` -In the application code, use `pool.get()?` to get a connection object [Taos]. +In the application code, use `pool.get()? ` to get a connection object [Taos]. ```rust -let taos = pool.get()? ; +let taos = pool.get()?; ``` -### Connection methods +# Connectors -The [Taos] connection struct provides several APIs for convenient use. +The [Taos][struct.Taos] object provides an API to perform operations on multiple databases. -1. `exec`: Execute some non-query SQL statements, such as `CREATE`, `ALTER`, `INSERT` etc. and return affected rows (only meaningful to `INSERT`). +1. `exec`: Execute some non-query SQL statements, such as `CREATE`, `ALTER`, `INSERT`, etc. ```rust let affected_rows = taos.exec("INSERT INTO tb1 VALUES(now, NULL)").await?; ``` -2. `exec_many`: You can execute many SQL statements in order with `exec_many` method. +2. `exec_many`: Run multiple SQL statements simultaneously or in order. ```rust taos.exec_many([ @@ -299,15 +300,15 @@ The [Taos] connection struct provides several APIs for convenient use. ]).await?; ``` -3. `query`: Execute the query statement and return the [ResultSet] object. +3. `query`: Run a query statement and return a [ResultSet] object. ```rust - let mut q = taos.query("select * from log.logs").await? + let mut q = taos.query("select * from log.logs").await?; ``` - The [ResultSet] object stores the query result data and basic information about the returned columns (column name, type, length). + The [ResultSet] object stores query result data and the names, types, and lengths of returned columns - Get filed information with `fields` method. + You can obtain column information by using [.fields()]. ```rust let cols = q.fields(); @@ -316,7 +317,7 @@ The [Taos] connection struct provides several APIs for convenient use. } ``` - Users could fetch data by rows. + It fetches data line by line. ```rust let mut rows = result.rows(); @@ -332,7 +333,7 @@ The [Taos] connection struct provides several APIs for convenient use. } ``` - Or use it with [serde](https://serde.rs) deserialization. + Or use the [serde](https://serde.rs) deserialization framework. ```rust #[derive(Debug, Deserialize)] @@ -359,15 +360,17 @@ The [Taos] connection struct provides several APIs for convenient use. Note that Rust asynchronous functions and an asynchronous runtime are required. -[Taos] provides a few Rust methods that encapsulate SQL to reduce the frequency of `format!` code blocks. +[Taos][struct.Taos] provides Rust methods for some SQL statements to reduce the number of `format!`s. - `.describe(table: &str)`: Executes `DESCRIBE` and returns a Rust data structure. - `.create_database(database: &str)`: Executes the `CREATE DATABASE` statement. - `.use_database(database: &str)`: Executes the `USE` statement. -### Bind API +In addition, this structure is also the entry point for [Parameter Binding](#Parameter Binding Interface) and [Line Protocol Interface](#Line Protocol Interface). Please refer to the specific API descriptions for usage. + +### Bind Interface -Similar to the C interface, Rust provides the bind interface's wrapping. First, create a bind object [Stmt] for a SQL command with the [Taos] object. +Similar to the C interface, Rust provides the bind interface's wrapping. First, the [Taos][struct.taos] object creates a parameter binding object [Stmt] for an SQL statement. ```rust let mut stmt = Stmt::init(&taos).await?; @@ -387,17 +390,17 @@ stmt.set_tbname("d0")?; #### `.set_tags(&[tag])` -Bind tag values when the SQL statement uses a super table. +Bind sub-table table names and tag values when the SQL statement uses a super table. ```rust let mut stmt = taos.stmt("insert into ? using stb0 tags(?) values(? ,?)")?; stmt.set_tbname("d0")?; -stmt.set_tags(&[Value::VarChar("涛思".to_string())])?; +stmt.set_tags(&[Value::VarChar("taos".to_string())])?; ``` #### `.bind(&[column])` -Bind value types. Use the [ColumnView] structure to construct the desired type and bind. +Bind value types. Use the [ColumnView] structure to create and bind the required types. ```rust let params = vec![ @@ -421,42 +424,42 @@ let rows = stmt.bind(¶ms)?.add_batch()?.execute()?; #### `.execute()` -Execute to insert all bind records. [Stmt] objects can be reused, re-bind, and executed after execution. Remember to call `add_batch` before `execute`. +Execute SQL. [Stmt] objects can be reused, re-binded, and executed after execution. Before execution, ensure that all data has been added to the queue with `.add_batch`. ```rust -stmt.add_batch()?.execute()?; +stmt.execute()?; // next bind cycle. -// stmt.set_tbname()? ; -//stmt.bind()? ; -//stmt.add_batch().execute()? ; +//stmt.set_tbname()?; +//stmt.bind()?; +//stmt.execute()?; ``` -A runnable example for bind can be found [here](https://github.com/taosdata/taos-connector-rust/blob/main/examples/bind.rs). +For a working example, see [GitHub](https://github.com/taosdata/taos-connector-rust/blob/main/examples/bind.rs). -### Subscription API +### Subscriptions -Users can subscribe a [TOPIC](../../../taos-sql/tmq/) with TMQ(the TDengine Message Queue) API. +TDengine starts subscriptions through [TMQ](../../../taos-sql/tmq/). -Start from a TMQ builder: +You create a TMQ connector by using a DSN. ```rust let tmq = TmqBuilder::from_dsn("taos://localhost:6030/?group.id=test")?; ``` -Build a consumer: +Create a consumer: ```rust let mut consumer = tmq.build()?; ``` -Subscribe a topic: +A single consumer can subscribe to one or more topics. ```rust consumer.subscribe(["tmq_meters"]).await?; ``` -Consume messages, and commit the offset for each message. +The TMQ is of [futures::Stream](https://docs.rs/futures/latest/futures/stream/index.html) type. You can use the corresponding API to consume each message in the queue and then use `.commit` to mark them as consumed. ```rust { @@ -495,22 +498,21 @@ Unsubscribe: consumer.unsubscribe().await; ``` -In TMQ DSN, you must choose to subscribe with a group id. Also, there's several options could be set: - -- `group.id`: **Required**, a group id is any visible string you set. -- `client.id`: a optional client description string. -- `auto.offset.reset`: choose to subscribe from *earliest* or *latest*, default is *none* which means 'earliest'. -- `enable.auto.commit`: automatically commit with specified time interval. By default - in the recommended way _ you must use `commit` to ensure that you've consumed the messages correctly, otherwise, consumers will received repeated messages when re-subscribe. -- `auto.commit.interval.ms`: the auto commit interval in milliseconds. +The following parameters can be configured for the TMQ DSN. Only `group.id` is mandatory. -Check the whole subscription example at [GitHub](https://github.com/taosdata/taos-connector-rust/blob/main/examples/subscribe.rs). +- `group.id`: Within a consumer group, load balancing is implemented by consuming messages on an at-least-once basis. +- `client.id`: Subscriber client ID. +- `auto.offset.reset`: Initial point of subscription. *earliest* subscribes from the beginning, and *latest* subscribes from the newest message. The default is earliest. Note: This parameter is set per consumer group. +- `enable.auto.commit`: Automatically commits. This can be enabled when data consistency is not essential. +- `auto.commit.interval.ms`: Interval for automatic commits. + +For more information, see [GitHub sample file](https://github.com/taosdata/taos-connector-rust/blob/main/examples/subscribe.rs). -Please move to the Rust documentation hosting page for other related structure API usage instructions: . +For information about other structure APIs, see the [Rust documentation](https://docs.rs/taos). -[TDengine]: https://github.com/taosdata/TDengine +[taos]: https://github.com/taosdata/rust-connector-taos [r2d2]: https://crates.io/crates/r2d2 -[Taos]: https://docs.rs/taos/latest/taos/struct.Taos.html -[ResultSet]: https://docs.rs/taos/latest/taos/struct.ResultSet.html -[Value]: https://docs.rs/taos/latest/taos/enum.Value.html -[Stmt]: https://docs.rs/taos/latest/taos/stmt/struct.Stmt.html -[taos]: https://crates.io/crates/taos +[TaosBuilder]: https://docs.rs/taos/latest/taos/struct.TaosBuilder.html +[TaosCfg]: https://docs.rs/taos/latest/taos/struct.TaosCfg.html +[struct.Taos]: https://docs.rs/taos/latest/taos/struct.Taos.html +[Stmt]: https://docs.rs/taos/latest/taos/struct.Stmt.html diff --git a/docs/en/14-reference/04-taosadapter.md b/docs/en/14-reference/04-taosadapter.md index cad229c32d602e8fc595ec06f72a1a486e2af77b..dc47246e2088deb33cda94d5b2bd72e11fa2b52c 100644 --- a/docs/en/14-reference/04-taosadapter.md +++ b/docs/en/14-reference/04-taosadapter.md @@ -30,7 +30,7 @@ taosAdapter provides the following features. ### Install taosAdapter -taosAdapter has been part of TDengine server software since TDengine v2.4.0.0. If you use the TDengine server, you don't need additional steps to install taosAdapter. You can download taosAdapter from [TDengine official website](https://tdengine.com/all-downloads/) to download the TDengine server installation package (taosAdapter is included in v2.4.0.0 and later version). If you need to deploy taosAdapter separately on another server other than the TDengine server, you should install the full TDengine server package on that server to install taosAdapter. If you need to build taosAdapter from source code, you can refer to the [Building taosAdapter]( https://github.com/taosdata/taosadapter/blob/develop/BUILD.md) documentation. +taosAdapter has been part of TDengine server software since TDengine v2.4.0.0. If you use the TDengine server, you don't need additional steps to install taosAdapter. You can download taosAdapter from [TDengine official website](https://tdengine.com/all-downloads/) to download the TDengine server installation package (taosAdapter is included in v2.4.0.0 and later version). If you need to deploy taosAdapter separately on another server other than the TDengine server, you should install the full TDengine server package on that server to install taosAdapter. If you need to build taosAdapter from source code, you can refer to the [Building taosAdapter]( https://github.com/taosdata/taosadapter/blob/3.0/BUILD.md) documentation. ### Start/Stop taosAdapter @@ -69,20 +69,23 @@ Usage of taosAdapter: --debug enable debug mode. Env "TAOS_ADAPTER_DEBUG" --help Print this help message and exit --influxdb.enable enable influxdb. Env "TAOS_ADAPTER_INFLUXDB_ENABLE" (default true) + --log.enableRecordHttpSql whether to record http sql. Env "TAOS_ADAPTER_LOG_ENABLE_RECORD_HTTP_SQL" --log.path string log path. Env "TAOS_ADAPTER_LOG_PATH" (default "/var/log/taos") --log.rotationCount uint log rotation count. Env "TAOS_ADAPTER_LOG_ROTATION_COUNT" (default 30) --log.rotationSize string log rotation size(KB MB GB), must be a positive integer. Env "TAOS_ADAPTER_LOG_ROTATION_SIZE" (default "1GB") --log.rotationTime duration log rotation time. Env "TAOS_ADAPTER_LOG_ROTATION_TIME" (default 24h0m0s) + --log.sqlRotationCount uint record sql log rotation count. Env "TAOS_ADAPTER_LOG_SQL_ROTATION_COUNT" (default 2) + --log.sqlRotationSize string record sql log rotation size(KB MB GB), must be a positive integer. Env "TAOS_ADAPTER_LOG_SQL_ROTATION_SIZE" (default "1GB") + --log.sqlRotationTime duration record sql log rotation time. Env "TAOS_ADAPTER_LOG_SQL_ROTATION_TIME" (default 24h0m0s) --logLevel string log level (panic fatal error warn warning info debug trace). Env "TAOS_ADAPTER_LOG_LEVEL" (default "info") --monitor.collectDuration duration Set monitor duration. Env "TAOS_MONITOR_COLLECT_DURATION" (default 3s) --monitor.identity string The identity of the current instance, or 'hostname:port' if it is empty. Env "TAOS_MONITOR_IDENTITY" --monitor.incgroup Whether running in cgroup. Env "TAOS_MONITOR_INCGROUP" - --monitor.password string TDengine password. Env "TAOS_MONITOR_PASSWORD" (default "taosdata") - --monitor.pauseAllMemoryThreshold float Memory percentage threshold for pause all. Env "TAOS_MONITOR_PAUSE_ALL_MEMORY_THRESHOLD" (default 80) + --monitor.password string TDengine password. Env "TAOS_MONITOR_PASSWORD" (default "taosdata") --monitor.pauseAllMemoryThreshold float Memory percentage threshold for pause all. Env "TAOS_MONITOR_PAUSE_ALL_MEMORY_THRESHOLD" (default 80) --monitor.pauseQueryMemoryThreshold float Memory percentage threshold for pause query. Env "TAOS_MONITOR_PAUSE_QUERY_MEMORY_THRESHOLD" (default 70) --monitor.user string TDengine user. Env "TAOS_MONITOR_USER" (default "root") --monitor.writeInterval duration Set write to TDengine interval. Env "TAOS_MONITOR_WRITE_INTERVAL" (default 30s) - --monitor.writeToTD Whether write metrics to TDengine. Env "TAOS_MONITOR_WRITE_TO_TD" (default true) + --monitor.writeToTD Whether write metrics to TDengine. Env "TAOS_MONITOR_WRITE_TO_TD" --node_exporter.caCertFile string node_exporter ca cert file path. Env "TAOS_ADAPTER_NODE_EXPORTER_CA_CERT_FILE" --node_exporter.certFile string node_exporter cert file path. Env "TAOS_ADAPTER_NODE_EXPORTER_CERT_FILE" --node_exporter.db string node_exporter db name. Env "TAOS_ADAPTER_NODE_EXPORTER_DB" (default "node_exporter") @@ -98,8 +101,10 @@ Usage of taosAdapter: --node_exporter.urls strings node_exporter urls. Env "TAOS_ADAPTER_NODE_EXPORTER_URLS" (default [http://localhost:9100]) --node_exporter.user string node_exporter user. Env "TAOS_ADAPTER_NODE_EXPORTER_USER" (default "root") --opentsdb.enable enable opentsdb. Env "TAOS_ADAPTER_OPENTSDB_ENABLE" (default true) + --opentsdb_telnet.batchSize int opentsdb_telnet batch size. Env "TAOS_ADAPTER_OPENTSDB_TELNET_BATCH_SIZE" (default 1) --opentsdb_telnet.dbs strings opentsdb_telnet db names. Env "TAOS_ADAPTER_OPENTSDB_TELNET_DBS" (default [opentsdb_telnet,collectd_tsdb,icinga2_tsdb,tcollector_tsdb]) --opentsdb_telnet.enable enable opentsdb telnet,warning: without auth info(default false). Env "TAOS_ADAPTER_OPENTSDB_TELNET_ENABLE" + --opentsdb_telnet.flushInterval duration opentsdb_telnet flush interval (0s means not valid) . Env "TAOS_ADAPTER_OPENTSDB_TELNET_FLUSH_INTERVAL" --opentsdb_telnet.maxTCPConnections int max tcp connections. Env "TAOS_ADAPTER_OPENTSDB_TELNET_MAX_TCP_CONNECTIONS" (default 250) --opentsdb_telnet.password string opentsdb_telnet password. Env "TAOS_ADAPTER_OPENTSDB_TELNET_PASSWORD" (default "taosdata") --opentsdb_telnet.ports ints opentsdb telnet tcp port. Env "TAOS_ADAPTER_OPENTSDB_TELNET_PORTS" (default [6046,6047,6048,6049]) @@ -111,9 +116,6 @@ Usage of taosAdapter: -P, --port int http port. Env "TAOS_ADAPTER_PORT" (default 6041) --prometheus.enable enable prometheus. Env "TAOS_ADAPTER_PROMETHEUS_ENABLE" (default true) --restfulRowLimit int restful returns the maximum number of rows (-1 means no limit). Env "TAOS_ADAPTER_RESTFUL_ROW_LIMIT" (default -1) - --ssl.certFile string ssl cert file path. Env "TAOS_ADAPTER_SSL_CERT_FILE" - --ssl.enable enable ssl. Env "TAOS_ADAPTER_SSL_ENABLE" - --ssl.keyFile string ssl key file path. Env "TAOS_ADAPTER_SSL_KEY_FILE" --statsd.allowPendingMessages int statsd allow pending messages. Env "TAOS_ADAPTER_STATSD_ALLOW_PENDING_MESSAGES" (default 50000) --statsd.db string statsd db name. Env "TAOS_ADAPTER_STATSD_DB" (default "statsd") --statsd.deleteCounters statsd delete counter cache after gather. Env "TAOS_ADAPTER_STATSD_DELETE_COUNTERS" (default true) @@ -149,11 +151,12 @@ You do not need to care about these configurations if you do not make interface For details on the CORS protocol, please refer to: [https://www.w3.org/wiki/CORS_Enabled](https://www.w3.org/wiki/CORS_Enabled) or [https://developer.mozilla.org/zh-CN/docs/Web/HTTP/CORS](https://developer.mozilla.org/zh-CN/docs/Web/HTTP/CORS). -See [example/config/taosadapter.toml](https://github.com/taosdata/taosadapter/blob/develop/example/config/taosadapter.toml) for sample configuration files. +See [example/config/taosadapter.toml](https://github.com/taosdata/taosadapter/blob/3.0/example/config/taosadapter.toml) for sample configuration files. ## Feature List -- Compatible with RESTful interfaces [REST API](/reference/rest-api/) +- RESTful interface + [https://docs.tdengine.com/reference/rest-api/](https://docs.tdengine.com/reference/rest-api/) - Compatible with InfluxDB v1 write interface [https://docs.influxdata.com/influxdb/v2.0/reference/api/influxdb-1x/write/](https://docs.influxdata.com/influxdb/v2.0/reference/api/influxdb-1x/write/) - Compatible with OpenTSDB JSON and telnet format writes @@ -176,13 +179,7 @@ See [example/config/taosadapter.toml](https://github.com/taosdata/taosadapter/bl ### TDengine RESTful interface -You can use any client that supports the http protocol to write data to or query data from TDengine by accessing the REST interface address `http://:6041/`. See the [official documentation](/reference/connector#restful) for details. The following EndPoint is supported. - -```text -/rest/sql -/rest/sqlt -/rest/sqlutc -``` +You can use any client that supports the http protocol to write data to or query data from TDengine by accessing the REST interface address `http://:6041/rest/sql`. See the [official documentation](/reference/rest-api/) for details. ### InfluxDB @@ -203,7 +200,7 @@ Note: InfluxDB token authorization is not supported at present. Only Basic autho ### OpenTSDB -You can use any client that supports the http protocol to access the RESTful interface address `http://:6041/` to write data in OpenTSDB compatible format to TDengine. +You can use any client that supports the http protocol to access the RESTful interface address `http://:6041/` to write data in OpenTSDB compatible format to TDengine. The EndPoint is as follows: ```text /opentsdb/v1/put/json/ @@ -254,7 +251,7 @@ HTTP response content. Stops processing all write and query requests when the `pauseAllMemoryThreshold` threshold is exceeded. -HTTP response: code 503 +HTTP response content. - code 503 - body "memory exceeds threshold" @@ -269,7 +266,7 @@ Status check interface `http://:6041/-/ping` Corresponding configuration parameter -``text +```text monitor.collectDuration monitoring interval environment variable `TAOS_MONITOR_COLLECT_DURATION` (default value 3s) monitor.incgroup whether to run in cgroup (set to true for running in container) environment variable `TAOS_MONITOR_INCGROUP` monitor.pauseAllMemoryThreshold memory threshold for no more inserts and queries environment variable `TAOS_MONITOR_PAUSE_ALL_MEMORY_THRESHOLD` (default 80) @@ -297,11 +294,11 @@ taosAdapter supports writing the metrics of HTTP monitoring, CPU percentage, and For configuration parameters | **Configuration items** | **Description** | **Default values** | -| ----------------------- | --------------------------------------------------------- | ---------- | +|-------------------------|--------------------------------------------|----------| | monitor.collectDuration | CPU and memory collection interval | 3s | | monitor.identity | The current taosadapter identifier will be used if not set to `hostname:port` | | | monitor.incgroup | whether it is running in a cgroup (set to true for running in a container) | false | -| monitor.writeToTD | Whether to write to TDengine | true | +| monitor.writeToTD | Whether to write to TDengine | false | | monitor.user | TDengine connection username | root | | monitor.password | TDengine connection password | taosdata | | monitor.writeInterval | Write to TDengine interval | 30s | @@ -313,9 +310,7 @@ taosAdapter controls the number of results returned by the parameter `restfulRow This parameter controls the number of results returned by the following interfaces: - `http://:6041/rest/sql` -- `http://:6041/rest/sqlt` -- `http://:6041/rest/sqlutc` -- ` http://:6041/prometheus/v1/remote_read/:db` +- `http://:6041/prometheus/v1/remote_read/:db` ## Troubleshooting @@ -328,10 +323,10 @@ You can also adjust the level of the taosAdapter log output by setting the `--lo In TDengine server 2.2.x.x or earlier, the TDengine server process (taosd) contains an embedded HTTP service. As mentioned earlier, taosAdapter is a standalone software managed using `systemd` and has its own process ID. There are some configuration parameters and behaviors that are different between the two. See the following table for details. | **#** | **embedded httpd** | **taosAdapter** | **comment** | -| ----- | ------------------- | ------------------------------------ | ------------------------------------------------------------------ ------------------------------------------------------------------------ | +|-------|---------------------|-------------------------------|------------------------------------------------------------------------------------------------| | 1 | httpEnableRecordSql | --logLevel=debug | | | 2 | httpMaxThreads | n/a | taosAdapter Automatically manages thread pools without this parameter | | 3 | telegrafUseFieldNum | See the taosAdapter telegraf configuration method | | | 4 | restfulRowLimit | restfulRowLimit | Embedded httpd outputs 10240 rows of data by default, the maximum allowed is 102400. taosAdapter also provides restfulRowLimit but it is not limited by default. You can configure it according to the actual scenario. | 5 | httpDebugFlag | Not applicable | httpdDebugFlag does not work for taosAdapter | -| 6 | httpDBNameMandatory | N/A | taosAdapter requires the database name to be specified in the URL | +| 6 | httpDBNameMandatory | N/A | taosAdapter requires the database name to be specified in the URL | \ No newline at end of file diff --git a/docs/en/14-reference/05-taosbenchmark.md b/docs/en/14-reference/05-taosbenchmark.md index 5d47cc06e8034d8c49669d71c3f98c1f587acb33..8f63dddfb711a98840cb423d1a4b0c1556a7b5fd 100644 --- a/docs/en/14-reference/05-taosbenchmark.md +++ b/docs/en/14-reference/05-taosbenchmark.md @@ -23,11 +23,7 @@ There are two ways to install taosBenchmark: TaosBenchmark needs to be executed on the terminal of the operating system, it supports two configuration methods: [Command-line arguments](#command-line-arguments-in-detail) and [JSON configuration file](#configuration-file-parameters-in-detail). These two methods are mutually exclusive. Users can use `-f ` to specify a configuration file. When running taosBenchmark with command-line arguments to control its behavior, users should use other parameters for configuration, but not the `-f` parameter. In addition, taosBenchmark offers a special way of running without parameters. -<<<<<<< HEAD -taosBenchmark supports complete performance testing of TDengine. taosBenchmark supports the TDengine functions in three categories: write, query, and subscribe. These three functions are mutually exclusive, and users can select only one of them each time taosBenchmark runs. It is important to note that the type of functionality to be tested is not configurable when using the command-line configuration method, which can only test writing performance. To test the query and subscription performance of the TDengine, you must use the configuration file method and specify the function type to test via the parameter `filetype` in the configuration file. -======= -taosBenchmark supports the complete performance testing of TDengine by providing functionally to write, query, and subscribe. These three functions are mutually exclusive, users can only select one of them each time taosBenchmark runs. The query and subscribe functionalities are only configurable using a json configuration file by specifying the parameter `filetype`, while write can be performed through both the command-line and a configuration file. ->>>>>>> 108548b4d6 (docs: typo) +taosBenchmark supports the complete performance testing of TDengine by providing functionally to write, query, and subscribe. These three functions are mutually exclusive, users can only select one of them each time taosBenchmark runs. The query and subscribe functionalities are only configurable using a json configuration file by specifying the parameter `filetype`, while write can be performed through both the command-line and a configuration file. If you want to test the performance of queries or data subscriptionm configure taosBenchmark with the configuration file. You can modify the value of the `filetype` parameter to specify the function that you want to test. **Make sure that the TDengine cluster is running correctly before running taosBenchmark. ** @@ -61,8 +57,9 @@ Use the following command-line to run taosBenchmark and control its behavior via taosBenchmark -f ``` +**Sample configuration files** + #### Configuration file examples -##### Example of inserting a scenario JSON configuration file
    insert.json @@ -73,7 +70,7 @@ taosBenchmark -f
    -##### Query Scenario JSON Profile Example +#### Query Scenario JSON Profile Example
    query.json @@ -84,7 +81,7 @@ taosBenchmark -f
    -##### Subscription JSON configuration example +#### Subscription JSON configuration example
    subscribe.json @@ -128,7 +125,7 @@ taosBenchmark -f Enables interleaved insertion mode and specifies the number of rows of data to be inserted into each child table. Interleaved insertion mode means inserting the number of rows specified by this parameter into each sub-table and repeating the process until all sub-tables have been inserted. The default value is 0, i.e., data is inserted into one sub-table before the next sub-table is inserted. - **-i/--insert-interval ** : - Specify the insert interval in `ms` for interleaved insert mode. The default value is 0. It only works if `-B/--interlace-rows` is greater than 0. That means that after inserting interlaced rows for each child table, the data insertion with multiple threads will wait for the interval specified by this value before proceeding to the next round of writes. + Specify the insert interval in `ms` for interleaved insert mode. The default value is 0. It only works if `-B/--interlace-rows` is greater than 0. After inserting interlaced rows for each child table, the data insertion thread will wait for the interval specified by this value before proceeding to the next round of writes. - **-r/--rec-per-req ** : Writing the number of rows of records per request to TDengine, the default value is 30000. @@ -184,7 +181,7 @@ taosBenchmark -A INT,DOUBLE,NCHAR,BINARY\(16\) This parameter indicates writing data with random values. The default is false. If users use this parameter, taosBenchmark will generate the random values. For tag/data columns of numeric type, the value is a random value within the range of values of that type. For NCHAR and BINARY type tag columns/data columns, the value is the random string within the specified length range. - **-x/--aggr-func** : - Switch parameter to indicate query aggregation function after insertion. The default value is false. + Switch parameter to indicate query aggregation function after insertion. The default is false. - **-y/--answer-yes** : Switch parameter that requires the user to confirm at the prompt to continue. The default value is false. @@ -230,45 +227,34 @@ The parameters listed in this section apply to all function modes. #### Database related configuration parameters -The parameters related to database creation are configured in `dbinfo` in the json configuration file, as follows. These parameters correspond to the database parameters specified when `create database` in TDengine. +The parameters related to database creation are configured in `dbinfo` in the json configuration file, as follows. The other parameters correspond to the database parameters specified when `create database` in [../../taos-sql/database]. - **name**: specify the name of the database. - **drop**: indicate whether to delete the database before inserting. The default is true. -- **replica**: specify the number of replicas when creating the database. - -- **days**: specify the time span for storing data in a single data file. The default is 10. - -- **cache**: specify the size of the cache blocks in MB. The default value is 16. - -- **blocks**: specify the number of cache blocks in each vnode. The default is 6. - -- **precision**: specify the database time precision. The default value is "ms". - -- **keep**: specify the number of days to keep the data. The default value is 3650. +#### Stream processing related configuration parameters -- **minRows**: specify the minimum number of records in the file block. The default value is 100. +The parameters for creating streams are configured in `stream` in the json configuration file, as shown below. -- **maxRows**: specify the maximum number of records in the file block. The default value is 4096. +- **stream_name**: Name of the stream. Mandatory. -- **comp**: specify the file compression level. The default value is 2. +- **stream_stb**: Name of the supertable for the stream. Mandatory. -- **walLevel** : specify WAL level, default is 1. +- **stream_sql**: SQL statement for the stream to process. Mandatory. -- **cacheLast**: indicate whether to allow the last record of each table to be kept in memory. The default value is 0. The value can be 0, 1, 2, or 3. +- **trigger_mode**: Triggering mode for stream processing. Optional. -- **quorum**: specify the number of writing acknowledgments in multi-replica mode. The default value is 1. +- **watermark**: Watermark for stream processing. Optional. -- **fsync**: specify the interval of fsync in ms when users set WAL to 2. The default value is 3000. - -- **update** : indicate whether to support data update, default value is 0, optional values are 0, 1, 2. +- **drop**: Whether to create the stream. Specify yes to create the stream or no to not create the stream. #### Super table related configuration parameters The parameters for creating super tables are configured in `super_tables` in the json configuration file, as shown below. - **name**: Super table name, mandatory, no default value. + - **child_table_exists** : whether the child table already exists, default value is "no", optional value is "yes" or "no". - **child_table_count** : The number of child tables, the default value is 10. @@ -319,6 +305,22 @@ The parameters for creating super tables are configured in `super_tables` in the - **tags_file** : only works when insert_mode is taosc, rest. The final tag value is related to the childtable_count. Suppose the tag data rows in the CSV file are smaller than the given number of child tables. In that case, taosBenchmark will read the CSV file data cyclically until the number of child tables specified by childtable_count is generated. Otherwise, taosBenchmark will read the childtable_count rows of tag data only. The final number of child tables generated is the smaller of the two. +#### TSMA configuration parameters + +The configuration parameters for specifying TSMAs are in `tsmas` in `super_tables`. + +- **name**: Specifies TSMA name. Mandatory. + +- **function**: Specifies TSMA function. Mandatory. + +- **interval**: Specifies TSMA interval. Mandatory. + +- **sliding**: Specifies time offset for TSMA window. Mandatory. + +- **custom**: Specifies custom configurations to attach to the end of the TSMA creation statement. Optional. + +- **start_when_inserted**: Specifies the number of inserted rows after which TSMA is started. Optional. The default value is 0. + #### Tag and Data Column Configuration Parameters The configuration parameters for specifying super table tag columns and data columns are in `columns` and `tag` in `super_tables`, respectively. @@ -338,6 +340,8 @@ The configuration parameters for specifying super table tag columns and data col - **values**: The value field of the nchar/binary column/label, which will be chosen randomly from the values. +- **sma**: Insert the column into the BSMA. Enter `yes` or `no`. The default is `no`. + #### insertion behavior configuration parameters - **thread_count**: specify the number of threads to insert data. Default is 8. @@ -350,17 +354,17 @@ The configuration parameters for specifying super table tag columns and data col - **confirm_parameter_prompt**: The switch parameter requires the user to confirm after the prompt to continue. The default value is false. -- **interlace_rows**: Enables interleaved insertion mode and specifies the number of rows of data to be inserted into each child table at a time. Interleaved insertion mode means inserting the number of rows specified by this parameter into each sub-table and repeating the process until all sub-tables are inserted. The default value is 0, which means that data will be inserted into the following child table only after data is inserted into one child table. +- **interlace_rows**: Enables interleaved insertion mode and specifies the number of rows of data to be inserted into each child table at a time. Staggered insertion mode means inserting the number of rows specified by this parameter into each sub-table and repeating the process until all sub-tables have been inserted. The default value is 0, i.e., data is inserted into one sub-table before the next sub-table is inserted. This parameter can also be configured in `super_tables`, and if so, the configuration in `super_tables` takes precedence and overrides the global setting. - **insert_interval** : - Specifies the insertion interval in ms for interleaved insertion mode. The default value is 0. Only works if `-B/--interlace-rows` is greater than 0. It means that after inserting interlace rows for each child table, the data insertion thread will wait for the interval specified by this value before proceeding to the next round of writes. - This parameter can also be configured in `super_tables`, and if configured, the configuration in `super_tables` takes high priority, overriding the global setting. + Specify the insert interval in `ms` for interleaved insert mode. The default value is 0. It only works if `-B/--interlace-rows` is greater than 0. After inserting interlaced rows for each child table, the data insertion thread will wait for the interval specified by this value before proceeding to the next round of writes. + This parameter can also be configured in `super_tables`, and if so, the configuration in `super_tables` takes precedence and overrides the global setting. - **num_of_records_per_req** : - The number of rows of data to be written per request to TDengine, the default value is 30000. When it is set too large, the TDengine client driver will return the corresponding error message, so you need to lower the setting of this parameter to meet the writing requirements. + Writing the number of rows of records per request to TDengine, the default value is 30000. When it is set too large, the TDengine client driver will return the corresponding error message, so you need to lower the setting of this parameter to meet the writing requirements. -- **prepare_rand**: The number of unique values in the generated random data. A value of 1 means that all data are the same. The default value is 10000. +- **prepare_rand**: The number of unique values in the generated random data. A value of 1 means that all data are equal. The default value is 10000. ### Query scenario configuration parameters @@ -388,7 +392,7 @@ The configuration parameters of the super table query are set in `super_table_qu - **threads**: The number of threads to execute the query SQL, the default value is 1. -- **sqls** : The default value is 1. +- **sqls**: - **sql**: The SQL command to be executed. For the query SQL of super table, keep "xxxx" in the SQL command. The program will automatically replace it with all the sub-table names of the super table. Replace it with all the sub-table names in the super table. - **result**: The file to save the query result. If not specified, taosBenchmark will not save result. @@ -411,9 +415,9 @@ The configuration parameters for subscribing to a sub-table or a generic table a - **resubAfterConsume**: "yes" means cancel the previous subscription and then subscribe again, "no" means continue the previous subscription, and the default value is "no". -- **sqls** : The default value is "no". +- **sqls**: - **sql** : The SQL command to be executed, required. - - **result** : The file to save the query result, unspecified is not saved. + - **result**: The file to save the query result. If not specified, taosBenchmark will not save result. #### Configuration parameters for subscribing to supertables @@ -431,7 +435,7 @@ The configuration parameters for subscribing to a super table are set in `super_ - **resubAfterConsume**: "yes" means cancel the previous subscription and then subscribe again, "no" means continue the previous subscription, and the default value is "no". -- **sqls** : The default value is "no". - - **sql**: SQL command to be executed, required; for the query SQL of the super table, keep "xxxx" in the SQL command, and the program will replace it with all the sub-table names of the super table automatically. +- **sqls**: + - **sql**: The SQL command to be executed. For the query SQL of super table, keep "xxxx" in the SQL command. The program will automatically replace it with all the sub-table names of the super table. Replace it with all the sub-table names in the super table. - - **result**: The file to save the query result, if not specified, it will not be saved. + - **result**: The file to save the query result. If not specified, taosBenchmark will not save result. diff --git a/docs/en/14-reference/08-taos-shell.md b/docs/en/14-reference/08-taos-shell.md index c947e86d1ce298f32e82b51c991892a9448dc88b..656db1f481250ed4e34e068a02a93b75f0ac0b81 100644 --- a/docs/en/14-reference/08-taos-shell.md +++ b/docs/en/14-reference/08-taos-shell.md @@ -8,7 +8,7 @@ The TDengine command-line interface (hereafter referred to as `TDengine CLI`) is ## Installation -If executed on the TDengine server-side, there is no need for additional installation steps to install TDengine CLI as it is already included and installed automatically. To run TDengine CLI in an environment where no TDengine server is running, the TDengine client installation package needs to be installed first. For details, please refer to [connector](/reference/connector/). +If executed on the TDengine server-side, there is no need for additional installation steps to install TDengine CLI as it is already included and installed automatically. To run TDengine CLI in an environment where no TDengine server is running, the TDengine client installation package needs to be installed first. For details, please refer to [Connector](/reference/connector/). ## Execution @@ -23,6 +23,7 @@ TDengine CLI will display a welcome message and version information if it succes ```cmd taos> ``` + After entering the TDengine CLI, you can execute various SQL commands, including inserts, queries, or administrative commands. ## Execute SQL script file @@ -51,32 +52,33 @@ You can change the behavior of TDengine CLI by specifying command-line parameter - -P PORT: Specify the port number to be used by the server. Default is `6030` - -u USER: the user name to use when connecting. Default is `root` - -p PASSWORD: the password to use when connecting to the server. Default is `taosdata` -- -?, --help: print out all command-line arguments +- -?, --help: print out all command-line arguments And many more parameters. -- -a AUTHSTR: The auth string to use when connecting to the server -- -A: Generate auth string from password +- -a AUTHSTR: Authorization information to connect to the server. +- -A: Obtain authorization information from username and password. - -c CONFIGDIR: Specify the directory where configuration file exists. The default is `/etc/taos`, and the default name of the configuration file in this directory is `taos.cfg` - -C: Print the configuration parameters of `taos.cfg` in the default directory or specified by -c - -d DATABASE: Specify the database to use when connecting to the server -- -f FILE: Execute the SQL script file in non-interactive mode -- -k: Check the service status, 0: unavailable，1: network ok，2: service ok，3: service degraded，4: exiting -- -l PKTLEN: Test package length to be used for network testing -- -n NETROLE: test scope for network connection test, default is `client`. The value can be `client`, `server` -- -N PKTNUM: Test package numbers to be used for network testing +- -f FILE: Execute the SQL script file in non-interactive mode Note that each SQL statement in the script file must be only one line. +- -k: Test the operational status of the server. 0: unavailable; 1: network ok; 2: service ok; 3: service degraded; 4: exiting +- -l PKTLEN: Test package size to be used for network testing +- -n NETROLE: test scope for network connection test, default is `client`. The value can be `client` or `server`. +- -N PKTNUM: Number of packets used for network testing - -r: output the timestamp format as unsigned 64-bits integer (uint64_t in C language) - -s COMMAND: execute SQL commands in non-interactive mode -- -t: Check the details of the service status，status same as -k -- -w DISPLAYWIDTH: 客户端列显示宽度 -- -z, --timezone=TIMEZONE: Specify time zone. Default is the value of current configuration file -- -V, --version: Print out the current version number +- -t: Test the boot status of the server. The statuses of -k apply. +- -w DISPLAYWIDTH: Specify the number of columns of the server display. +- -z TIMEZONE: Specify time zone. Default is the value of current configuration file +- -V: Print out the current version number -Example. +For example: ```bash taos -h h1.taos.com -s "use db; show tables;" ``` + ## TDengine CLI tips - You can use the up and down keys to iterate the history of commands entered diff --git a/docs/en/14-reference/09-support-platform/index.md b/docs/en/14-reference/09-support-platform/index.md index 0bb269f2329097975f665b9318b2ec3f320fca5b..19c984898d9037d4414fe13157060fe744b8e179 100644 --- a/docs/en/14-reference/09-support-platform/index.md +++ b/docs/en/14-reference/09-support-platform/index.md @@ -5,11 +5,10 @@ description: "List of platforms supported by TDengine server, client, and connec ## List of supported platforms for TDengine server -| | **Windows server 2016/2019** | **Windows 10/11** | **CentOS 7.9/8** | **Ubuntu 18/20** | **UOS** | **kylin** | **Ningsi V60/V80** | -| ------------------ | ---------------------------- | ----------------- | ---------------- | ---------------- | ------- | --------- | ------------------ | -| X64 | ● | ● | ● | ● | ● | ● | ● | -| Raspberry Pi ARM64 | | | ● | | | | | -| HUAWEI Cloud ARM64 | | | | ● | | | | +| | **Windows Server 2016/2019** | **Windows 10/11** | **CentOS 7.9/8** | **Ubuntu 18/20** | +| ------------ | ---------------------------- | ----------------- | ---------------- | ---------------- | +| X64 | ● | ● | ● | ● | +| ARM64 | | | ● | | Note: ● means officially tested and verified, ○ means unofficially tested and verified. diff --git a/docs/en/14-reference/11-docker/index.md b/docs/en/14-reference/11-docker/index.md index b7e60ab3e7f04a6078950977a563382a3524ebaa..3528fd194b0dd40b2bf67dd3e05a42d93a47ab05 100644 --- a/docs/en/14-reference/11-docker/index.md +++ b/docs/en/14-reference/11-docker/index.md @@ -1,6 +1,6 @@ --- title: Deploying TDengine with Docker -Description: "This chapter focuses on starting the TDengine service in a container and accessing it." +description: "This chapter focuses on starting the TDengine service in a container and accessing it." --- This chapter describes how to start the TDengine service in a container and access it. Users can control the behavior of the service in the container by using environment variables on the docker run command-line or in the docker-compose file. @@ -24,13 +24,10 @@ The TDengine client taos can be executed in this container to access TDengine us ```shell $ docker exec -it tdengine taos -Welcome to the TDengine shell from Linux, Client Version:2.4.0.0 -Copyright (c) 2020 by TAOS Data, Inc. - taos> show databases; - name | created_time | ntables | vgroups | replica | quorum | days | keep | cache(MB) | blocks | minrows | maxrows | wallevel | fsync | comp | cachelast | precision | update | status | status precision | update | status | -================================================================================================================================== ================================================================================================================================== ================ - log | 2022-01-17 13:57:22.270 | 10 | 1 | 1 | 1 | 10 | 30 | 1 | 3 | 100 | 4096 | 1 | 3000 | 2 | 0 | us | 0 | ready | + name | created_time | ntables | vgroups | replica | quorum | days | keep | cache(MB) | blocks | minrows | maxrows | wallevel | fsync | comp | cachelast | precision | update | status | +==================================================================================================================================================================================================================================================================================== + log | 2022-01-17 13:57:22.270 | 10 | 1 | 1 | 1 | 10 | 30 | 1 | 3 | 100 | 4096 | 1 | 3000 | 2 | 0 | us | 0 | ready | Query OK, 1 row(s) in set (0.002843s) ``` @@ -47,13 +44,10 @@ The above command starts TDengine on the host network and uses the host's FQDN t ```shell $ taos -Welcome to the TDengine shell from Linux, Client Version:2.4.0.0 -Copyright (c) 2020 by TAOS Data, Inc. - taos> show dnodes; - id | end_point | vnodes | cores | status | role | create_time | offline reason | -================================================================================================================================== ==== - 1 | myhost:6030 | 1 | 8 | ready | any | 2022-01-17 22:10:32.619 | | + id | end_point | vnodes | cores | status | role | create_time | offline reason | +====================================================================================================================================== + 1 | myhost:6030 | 1 | 8 | ready | any | 2022-01-17 22:10:32.619 | | Query OK, 1 row(s) in set (0.003233s) ``` @@ -88,13 +82,13 @@ If set `TAOS_FQDN` to the same hostname, the effect is the same as "Start TDengi ## Start TDengine on the specified network -You can also start TDengine on a specific network. +You can also start TDengine on a specific network. Perform the following steps: 1. First, create a docker network named `td-net` ```shell docker network create td-net - ``` Create td-net + ``` 2. Start TDengine @@ -111,7 +105,7 @@ You can also start TDengine on a specific network. ```shell docker run --rm -it --network td-net -e TAOS_FIRST_EP=tdengine tdengine/tdengine taos # or - # docker run --rm -it --network td-net -e tdengine/tdengine taos -h tdengine + #docker run --rm -it --network td-net -e tdengine/tdengine taos -h tdengine ``` ## Launching a client application in a container @@ -147,7 +141,7 @@ import ( "fmt" "time" - _ "github.com/taosdata/driver-go/v2/taosSql" + _ "github.com/taosdata/driver-go/v3/taosSql" ) type config struct { @@ -316,12 +310,12 @@ password: taosdata ``` :::note + - The `VERSION` environment variable is used to set the tdengine image tag - - `TAOS_FIRST_EP` must be set on the newly created instance so that it can join the TDengine cluster; if there is a high availability requirement, `TAOS_SECOND_EP` needs to be used at the same time - - `TAOS_REPLICA` is used to set the default number of database replicas. Its value range is [1,3] - We recommend setting it with `TAOS_ARBITRATOR` to use arbitrator in a two-nodes environment. - - ::: +- `TAOS_FIRST_EP` must be set on the newly created instance so that it can join the TDengine cluster; if there is a high availability requirement, `TAOS_SECOND_EP` needs to be used at the same time +- `TAOS_REPLICA` is used to set the default number of database replicas. Its value range is [1,3] + We recommend setting it with `TAOS_ARBITRATOR` to use arbitrator in a two-nodes environment. + ::: 2. Start the cluster @@ -353,9 +347,6 @@ password: taosdata ```shell $ docker-compose exec td-1 taos -s "show dnodes" - Welcome to the TDengine shell from Linux, Client Version:2.4.0.0 - Copyright (c) 2020 by TAOS Data, Inc. All rights reserved. - taos> show dnodes id | end_point | vnodes | cores | status | role | create_time | offline reason | ====================================================================================================================================== @@ -371,15 +362,15 @@ password: taosdata 2. At the same time, for flexible deployment, taosAdapter can be started in a separate container - ```docker - services: - # ... - adapter: - image: tdengine/tdengine:$VERSION - command: taosadapter - ```` + ```docker + services: + # ... + adapter: + image: tdengine/tdengine:$VERSION + command: taosadapter + ``` - Suppose you want to deploy multiple taosAdapters to improve throughput and provide high availability. In that case, the recommended configuration method uses a reverse proxy such as Nginx to offer a unified access entry. For specific configuration methods, please refer to the official documentation of Nginx. Here is an example: + Suppose you want to deploy multiple taosAdapters to improve throughput and provide high availability. In that case, the recommended configuration method uses a reverse proxy such as Nginx to offer a unified access entry. For specific configuration methods, please refer to the official documentation of Nginx. Here is an example: ```docker version: "3" diff --git a/docs/en/14-reference/12-directory.md b/docs/en/14-reference/12-directory.md index 118bce8037fdae5b303b45988277d10a99aa5445..0eaa7843ecdc284f2478cf5e43e9ce118bb69b3d 100644 --- a/docs/en/14-reference/12-directory.md +++ b/docs/en/14-reference/12-directory.md @@ -31,8 +31,10 @@ All executable files of TDengine are in the _/usr/local/taos/bin_ directory by d :::note taosdump after version 2.4.0.0 require taosTools as a standalone installation. A new version of taosBenchmark is include in taosTools too. + ::: :::tip You can configure different data directories and log directories by modifying the system configuration file `taos.cfg`. + ::: diff --git a/docs/en/14-reference/14-taosKeeper.md b/docs/en/14-reference/14-taosKeeper.md new file mode 100644 index 0000000000000000000000000000000000000000..476b5a1fd20b4dce4379026a6300ae8e26db6656 --- /dev/null +++ b/docs/en/14-reference/14-taosKeeper.md @@ -0,0 +1,134 @@ +--- +sidebar_label: taosKeeper +title: taosKeeper +description: Instructions and tips for using taosKeeper +--- + +## Introduction + +taosKeeper is a tool for TDengine that exports monitoring metrics. With taosKeeper, you can easily monitor the operational status of your TDengine deployment. taosKeeper uses the TDengine REST API. It is not necessary to install TDengine Client to use taosKeeper. + +## Installation + + +Methods of installing taosKeeper: + + + +- You can compile taosKeeper separately and install it. Please refer to the [taosKeeper](https://github.com/taosdata/taoskeeper) repository for details. --> +You can compile taosKeeper separately and install it. Please refer to the [taosKeeper](https://github.com/taosdata/taoskeeper) repository for details. + +## Run + +### Configuration and running methods + + +taosKeeper needs to be executed on the terminal of the operating system. To run taosKeeper, see [configuration file](#configuration-file-parameters-in-detail). + +**Make sure that the TDengine cluster is running correctly before running taosKeeper. ** Ensure that the monitoring service in TDengine has been started. For more information, see [TDengine Monitoring Configuration](../config/#monitoring). + + +### Configuration File + +You can quickly launch taosKeeper with the following commands. If you do not specify a configuration file, `/etc/taos/keeper.toml` is used by default. If this file does not specify configurations, the default values are used. + +```shell +taoskeeper -c +``` + +**Sample configuration files** +```toml +# enable debug in gin framework +debug = false + +# listen to server port, default 6043 +port = 6043 + +# set log level to panic, error, info, debug, or trace +loglevel = "info" + +# set pool size +gopoolsize = 50000 + +# query rotation period for TDengine monitoring data +RotationInterval = "15s" + +[tdengine] +host = "127.0.0.1" +port = 6041 +username = "root" +password = "taosdata" + +# set taosAdapter to monitor +[taosAdapter] +address = ["127.0.0.1:6041","192.168.1.95:6041"] + +[metrics] +# monitoring metric prefix +prefix = "taos" + +# cluster data identifier +cluster = "production" + +# database to store monitoring data +database = "log" + +# standard tables to monitor +tables = ["normal_table"] +``` + +### Obtain Monitoring Metrics + +taosKeeper records monitoring metrics generated by TDengine in a specified database and provides an interface through which you can export the data. + +#### View Monitoring Results + +```shell +$ taos +# the log database is used in this example +> use log; +> select * from cluster_info limit 1; +``` + +Example result set: + +```shell + ts | first_ep | first_ep_dnode_id | version | master_uptime | monitor_interval | dbs_total | tbs_total | stbs_total | dnodes_total | dnodes_alive | mnodes_total | mnodes_alive | vgroups_total | vgroups_alive | vnodes_total | vnodes_alive | connections_total | protocol | cluster_id | +=============================================================================================================================================================================================================================================================================================================================================================================== + 2022-08-16 17:37:01.629 | hlb:6030 | 1 | 3.0.0.0 | 0.27250 | 15 | 2 | 27 | 38 | 1 | 1 | 1 | 1 | 4 | 4 | 4 | 4 | 14 | 1 | 5981392874047724755 | +Query OK, 1 rows in database (0.036162s) +``` + +#### Export Monitoring Metrics + +```shell +curl http://127.0.0.1:6043/metrics +``` + +Sample result set (excerpt): + +```shell +# HELP taos_cluster_info_connections_total +# TYPE taos_cluster_info_connections_total counter +taos_cluster_info_connections_total{cluster_id="5981392874047724755"} 16 +# HELP taos_cluster_info_dbs_total +# TYPE taos_cluster_info_dbs_total counter +taos_cluster_info_dbs_total{cluster_id="5981392874047724755"} 2 +# HELP taos_cluster_info_dnodes_alive +# TYPE taos_cluster_info_dnodes_alive counter +taos_cluster_info_dnodes_alive{cluster_id="5981392874047724755"} 1 +# HELP taos_cluster_info_dnodes_total +# TYPE taos_cluster_info_dnodes_total counter +taos_cluster_info_dnodes_total{cluster_id="5981392874047724755"} 1 +# HELP taos_cluster_info_first_ep +# TYPE taos_cluster_info_first_ep gauge +taos_cluster_info_first_ep{cluster_id="5981392874047724755",value="hlb:6030"} 1 +``` \ No newline at end of file diff --git a/docs/en/14-reference/_category_.yml b/docs/en/14-reference/_category_.yml index 5f5466532be79153d42da0907df6336439593601..1fb42e60a7c2872dbf9f66096ea9a38c8aa4a295 100644 --- a/docs/en/14-reference/_category_.yml +++ b/docs/en/14-reference/_category_.yml @@ -1 +1 @@ -label: Reference +label: Reference \ No newline at end of file diff --git a/docs/en/14-reference/index.md b/docs/en/14-reference/index.md index f350eebfc1a1ca2feaedc18c4b4fa798742e31b4..f3a64913d065d1d8e321ce7433c9d605ef70bd13 100644 --- a/docs/en/14-reference/index.md +++ b/docs/en/14-reference/index.md @@ -2,11 +2,11 @@ title: Reference --- -The reference guide is a detailed introduction to TDengine including various TDengine connectors in different languages, and the tools that come with TDengine. +This section describes the TDengine connectors and utilities. ```mdx-code-block import DocCardList from '@theme/DocCardList'; import {useCurrentSidebarCategory} from '@docusaurus/theme-common'; -``` +``` \ No newline at end of file diff --git a/docs/en/28-releases.md b/docs/en/28-releases.md new file mode 100644 index 0000000000000000000000000000000000000000..a0c9eb119999571fb973b5e2243f237b8833b167 --- /dev/null +++ b/docs/en/28-releases.md @@ -0,0 +1,9 @@ +--- +sidebar_label: Releases +title: Released Versions +--- + +import Release from "/components/ReleaseV3"; + + + diff --git a/docs/examples/node/nativeexample/subscribe_demo.js b/docs/examples/node/nativeexample/subscribe_demo.js index 5b65e1c90758f208b4fc32359e97f3fd83a3a380..53cbe55d264a0e83b4d4b441b0b912872bbb7018 100644 --- a/docs/examples/node/nativeexample/subscribe_demo.js +++ b/docs/examples/node/nativeexample/subscribe_demo.js @@ -28,8 +28,7 @@ function runConsumer() { console.log(msg.topicPartition); console.log(msg.block); console.log(msg.fields) - // fixme(@xiaolei): commented temp, should be fixed. - //consumer.commit(msg); + consumer.commit(msg); console.log(`=======consumer ${i} done`) } diff --git a/docs/examples/node/package.json b/docs/examples/node/package.json index 36d3f016b5262472d5c63a2c98cc9704e57a59fe..d00d71d99fdff89af68f31a50416681733a08274 100644 --- a/docs/examples/node/package.json +++ b/docs/examples/node/package.json @@ -4,7 +4,7 @@ "main": "index.js", "license": "MIT", "dependencies": { - "@tdengine/client": "^3.0.0", + "@tdengine/client": "^3.0.1", "@tdengine/rest": "^3.0.0" } } diff --git a/docs/examples/rust/Cargo.toml b/docs/examples/rust/Cargo.toml deleted file mode 100644 index 136d09ffbbbd9c7bc1b876e7bfc630dea0560382..0000000000000000000000000000000000000000 --- a/docs/examples/rust/Cargo.toml +++ /dev/null @@ -1,2 +0,0 @@ -[workspace] -members = ["restexample", "nativeexample"] diff --git a/docs/zh/07-develop/02-model/index.mdx b/docs/zh/07-develop/02-model/index.mdx index be545e8813b26b3abbb22d8231590a909e935a83..1609eb5362cf40e7d134b0987968f7cc9bd31c92 100644 --- a/docs/zh/07-develop/02-model/index.mdx +++ b/docs/zh/07-develop/02-model/index.mdx @@ -11,10 +11,10 @@ TDengine 采用类关系型数据模型，需要建库、建表。因此对于 不同类型的数据采集点往往具有不同的数据特征，包括数据采集频率的高低，数据保留时间的长短，副本的数目，数据块的大小，是否允许更新数据等等。为了在各种场景下 TDengine 都能最大效率的工作，TDengine 建议将不同数据特征的表创建在不同的库里，因为每个库可以配置不同的存储策略。创建一个库时，除 SQL 标准的选项外，还可以指定保留时长、副本数、缓存大小、时间精度、文件块里最大最小记录条数、是否压缩、一个数据文件覆盖的天数等多种参数。比如： ```sql -CREATE DATABASE power KEEP 365 DURATION 10 BUFFER 16 VGROUPS 100 WAL 1; +CREATE DATABASE power KEEP 365 DURATION 10 BUFFER 16 WAL_LEVEL 1; ``` -上述语句将创建一个名为 power 的库，这个库的数据将保留 365 天（超过 365 天将被自动删除），每 10 天一个数据文件，每个 VNODE 的写入内存池的大小为 16 MB，数据库的 VGROUPS 数量，对该数据库入会写 WAL 但不执行 FSYNC。详细的语法及参数请见 [数据库管理](/taos-sql/database) 章节。 +上述语句将创建一个名为 power 的库，这个库的数据将保留 365 天（超过 365 天将被自动删除），每 10 天一个数据文件，每个 VNODE 的写入内存池的大小为 16 MB，对该数据库入会写 WAL 但不执行 FSYNC。详细的语法及参数请见 [数据库管理](/taos-sql/database) 章节。 创建库之后，需要使用 SQL 命令 `USE` 将当前库切换过来，例如： diff --git a/docs/zh/07-develop/04-query-data/index.mdx b/docs/zh/07-develop/04-query-data/index.mdx index 68f49d9f2b36fce83dc76e43e36f1049ae3de18d..2631d147a5f3e968e7153de8576e96f2c07c57cd 100644 --- a/docs/zh/07-develop/04-query-data/index.mdx +++ b/docs/zh/07-develop/04-query-data/index.mdx @@ -54,20 +54,20 @@ Query OK, 2 row(s) in set (0.001100s) 在 TAOS Shell，查找加利福尼亚州所有智能电表采集的电压平均值，并按照 location 分组。 ``` -taos> SELECT AVG(voltage) FROM meters GROUP BY location; - avg(voltage) | location | -============================================================= - 222.000000000 | California.LosAngeles | - 219.200000000 | California.SanFrancisco | -Query OK, 2 row(s) in set (0.002136s) +taos> SELECT AVG(voltage), location FROM meters GROUP BY location; + avg(voltage) | location | +=============================================================================================== + 219.200000000 | California.SanFrancisco | + 221.666666667 | California.LosAngeles | +Query OK, 2 rows in database (0.005995s) ``` ### 示例二 -在 TAOS shell, 查找 groupId 为 2 的所有智能电表过去 24 小时的记录条数，电流的最大值。 +在 TAOS shell, 查找 groupId 为 2 的所有智能电表的记录条数，电流的最大值。 ``` -taos> SELECT count(*), max(current) FROM meters where groupId = 2 and ts > now - 24h; +taos> SELECT count(*), max(current) FROM meters where groupId = 2; cunt(*) | max(current) | ================================== 5 | 13.4 | @@ -81,40 +81,41 @@ Query OK, 1 row(s) in set (0.002136s) 物联网场景里，经常需要通过降采样（down sampling）将采集的数据按时间段进行聚合。TDengine 提供了一个简便的关键词 interval 让按照时间窗口的查询操作变得极为简单。比如，将智能电表 d1001 采集的电流值每 10 秒钟求和 ``` -taos> SELECT sum(current) FROM d1001 INTERVAL(10s); - ts | sum(current) | +taos> SELECT _wstart, sum(current) FROM d1001 INTERVAL(10s); + _wstart | sum(current) | ====================================================== 2018-10-03 14:38:00.000 | 10.300000191 | 2018-10-03 14:38:10.000 | 24.900000572 | -Query OK, 2 row(s) in set (0.000883s) +Query OK, 2 rows in database (0.003139s) ``` 降采样操作也适用于超级表，比如：将加利福尼亚州所有智能电表采集的电流值每秒钟求和 ``` -taos> SELECT SUM(current) FROM meters where location like "California%" INTERVAL(1s); - ts | sum(current) | +taos> SELECT _wstart, SUM(current) FROM meters where location like "California%" INTERVAL(1s); + _wstart | sum(current) | ====================================================== 2018-10-03 14:38:04.000 | 10.199999809 | - 2018-10-03 14:38:05.000 | 32.900000572 | + 2018-10-03 14:38:05.000 | 23.699999809 | 2018-10-03 14:38:06.000 | 11.500000000 | 2018-10-03 14:38:15.000 | 12.600000381 | - 2018-10-03 14:38:16.000 | 36.000000000 | -Query OK, 5 row(s) in set (0.001538s) + 2018-10-03 14:38:16.000 | 34.400000572 | +Query OK, 5 rows in database (0.007413s) ``` 降采样操作也支持时间偏移，比如：将所有智能电表采集的电流值每秒钟求和，但要求每个时间窗口从 500 毫秒开始 ``` -taos> SELECT SUM(current) FROM meters INTERVAL(1s, 500a); - ts | sum(current) | +taos> SELECT _wstart, SUM(current) FROM meters INTERVAL(1s, 500a); + _wstart | sum(current) | ====================================================== - 2018-10-03 14:38:04.500 | 11.189999809 | - 2018-10-03 14:38:05.500 | 31.900000572 | - 2018-10-03 14:38:06.500 | 11.600000000 | - 2018-10-03 14:38:15.500 | 12.300000381 | - 2018-10-03 14:38:16.500 | 35.000000000 | -Query OK, 5 row(s) in set (0.001521s) + 2018-10-03 14:38:03.500 | 10.199999809 | + 2018-10-03 14:38:04.500 | 10.300000191 | + 2018-10-03 14:38:05.500 | 13.399999619 | + 2018-10-03 14:38:06.500 | 11.500000000 | + 2018-10-03 14:38:14.500 | 12.600000381 | + 2018-10-03 14:38:16.500 | 34.400000572 | +Query OK, 6 rows in database (0.005515s) ``` 物联网场景里，每个数据采集点采集数据的时间是难同步的，但很多分析算法(比如 FFT)需要把采集的数据严格按照时间等间隔的对齐，在很多系统里，需要应用自己写程序来处理，但使用 TDengine 的降采样操作就轻松解决。 diff --git a/docs/zh/12-taos-sql/03-table.md b/docs/zh/12-taos-sql/03-table.md index 1e20f73541b7465db76603dc16da8cd1daea0191..0e104bb7b6f09e886ab3c6cb55b1ecd68dfaf1ce 100644 --- a/docs/zh/12-taos-sql/03-table.md +++ b/docs/zh/12-taos-sql/03-table.md @@ -110,7 +110,7 @@ alter_table_option: { 对普通表可以进行如下修改操作 1. ADD COLUMN：添加列。 2. DROP COLUMN：删除列。 -3. ODIFY COLUMN：修改列定义，如果数据列的类型是可变长类型，那么可以使用此指令修改其宽度，只能改大，不能改小。 +3. MODIFY COLUMN：修改列定义，如果数据列的类型是可变长类型，那么可以使用此指令修改其宽度，只能改大，不能改小。 4. RENAME COLUMN：修改列名称。 ### 增加列 @@ -195,4 +195,4 @@ SHOW CREATE TABLE tb_name; ``` DESCRIBE [db_name.]tb_name; -``` \ No newline at end of file +``` diff --git a/docs/zh/12-taos-sql/14-stream.md b/docs/zh/12-taos-sql/14-stream.md index 1a056e278c5e3620cf1a31b9e8e358f9b05929f8..a967299e4093a4a8654d7aaf1b8c3914726aeadf 100644 --- a/docs/zh/12-taos-sql/14-stream.md +++ b/docs/zh/12-taos-sql/14-stream.md @@ -3,9 +3,6 @@ sidebar_label: 流式计算 title: 流式计算 --- -在时序数据的处理中，经常要对原始数据进行清洗、预处理，再使用时序数据库进行长久的储存。用户通常需要在时序数据库之外再搭建 Kafka、Flink、Spark 等流计算处理引擎，增加了用户的开发成本和维护成本。 - -使用 TDengine 3.0 的流式计算引擎能够最大限度的减少对这些额外中间件的依赖，真正将数据的写入、预处理、长期存储、复杂分析、实时计算、实时报警触发等功能融为一体，并且，所有这些任务只需要使用 SQL 完成，极大降低了用户的学习成本、使用成本。 ## 创建流式计算 @@ -40,6 +37,8 @@ window_clause: { 其中，SESSION 是会话窗口，tol_val 是时间间隔的最大范围。在 tol_val 时间间隔范围内的数据都属于同一个窗口，如果连续的两条数据的时间超过 tol_val，则自动开启下一个窗口。 +窗口的定义与时序数据特色查询中的定义完全相同。 + 例如，如下语句创建流式计算，同时自动创建名为 avg_vol 的超级表，此流计算以一分钟为时间窗口、30 秒为前向增量统计这些电表的平均电压，并将来自 meters 表的数据的计算结果写入 avg_vol 表，不同 partition 的数据会分别创建子表并写入不同子表。 ```sql @@ -47,10 +46,18 @@ CREATE STREAM avg_vol_s INTO avg_vol AS SELECT _wstartts, count(*), avg(voltage) FROM meters PARTITION BY tbname INTERVAL(1m) SLIDING(30s); ``` +## 流式计算的 partition + +可以使用 PARTITION BY TBNAME 或 PARTITION BY tag，对一个流进行多分区的计算，每个分区的时间线与时间窗口是独立的，会各自聚合，并写入到目的表中的不同子表。 + +不带 PARTITION BY 选项时，所有的数据将写入到一张子表。 + +流式计算创建的超级表有唯一的 tag 列 groupId，每个 partition 会被分配唯一 groupId。与 schemaless 写入一致，我们通过 MD5 计算子表名，并自动创建它。 + ## 删除流式计算 ```sql -DROP STREAM [IF NOT EXISTS] stream_name +DROP STREAM [IF NOT EXISTS] stream_name; ``` 仅删除流式计算任务，由流式计算写入的数据不会被删除。 @@ -61,6 +68,12 @@ DROP STREAM [IF NOT EXISTS] stream_name SHOW STREAMS; ``` +若要展示更详细的信息，可以使用： + +```sql +SELECT * from performance_schema.`perf_streams`; +``` + ## 流式计算的触发模式 在创建流时，可以通过 TRIGGER 指令指定流式计算的触发模式。 @@ -69,7 +82,7 @@ SHOW STREAMS; 1. AT_ONCE：写入立即触发 -2. WINDOW_CLOSE：窗口关闭时触发（窗口关闭由事件时间决定，可配合 watermark 使用，详见《流式计算的乱序数据容忍策略》） +2. WINDOW_CLOSE：窗口关闭时触发（窗口关闭由事件时间决定，可配合 watermark 使用） 3. MAX_DELAY time：若窗口关闭，则触发计算。若窗口未关闭，且未关闭时长超过 max delay 指定的时间，则触发计算。 @@ -79,21 +92,44 @@ SHOW STREAMS; MAX_DELAY 模式在窗口关闭时会立即触发计算。此外，当数据写入后，计算触发的时间超过 max delay 指定的时间，则立即触发计算 -## 流式计算的乱序数据容忍策略 +## 流式计算的窗口关闭 + +流式计算以事件时间（插入记录中的时间戳主键）为基准计算窗口关闭，而非以 TDengine 服务器的时间，以事件时间为基准，可以避免客户端与服务器时间不一致带来的问题，能够解决乱序数据写入等等问题。流式计算还提供了 watermark 来定义容忍的乱序程度。 -在创建流时，可以在 stream_option 中指定 watermark。 +在创建流时，可以在 stream_option 中指定 watermark，它定义了数据乱序的容忍上界。 流式计算通过 watermark 来度量对乱序数据的容忍程度，watermark 默认为 0。 T = 最新事件时间 - watermark -每批到来的数据都会以上述公式更新窗口关闭时间，并将窗口结束时间 < T 的所有打开的窗口关闭，若触发模式为 WINDOW_CLOSE 或 MAX_DELAY，则推送窗口聚合结果。 +每次写入的数据都会以上述公式更新窗口关闭时间，并将窗口结束时间 < T 的所有打开的窗口关闭，若触发模式为 WINDOW_CLOSE 或 MAX_DELAY，则推送窗口聚合结果。 + + +![TDengine 流式计算窗口关闭示意图](./watermark.webp) + + +图中，纵轴表示不同时刻，对于不同时刻，我们画出其对应的 TDengine 收到的数据，即为横轴。 + +横轴上的数据点表示已经收到的数据，其中蓝色的点表示事件时间(即数据中的时间戳主键)最后的数据，该数据点减去定义的 watermark 时间，得到乱序容忍的上界 T。 + +所有结束时间小于 T 的窗口都将被关闭（图中以灰色方框标记）。 + +T2 时刻，乱序数据（黄色的点）到达 TDengine，由于有 watermark 的存在，这些数据进入的窗口并未被关闭，因此可以被正确处理。 + +T3 时刻，最新事件到达，T 向后推移超过了第二个窗口关闭的时间，该窗口被关闭，乱序数据被正确处理。 + +在 window_close 或 max_delay 模式下，窗口关闭直接影响推送结果。在 at_once 模式下，窗口关闭只与内存占用有关。 + + +## 流式计算的过期数据处理策略 + +对于已关闭的窗口，再次落入该窗口中的数据被标记为过期数据. + +TDengine 对于过期数据提供两种处理方式，由 IGNORE EXPIRED 选项指定： -流式计算的过期数据处理策略 -对于已关闭的窗口，再次落入该窗口中的数据被标记为过期数据，对于过期数据，流式计算提供两种处理方式： +1. 重新计算，即 IGNORE EXPIRED 0：默认配置，从 TSDB 中重新查找对应窗口的所有数据并重新计算得到最新结果 -1. 直接丢弃：这是常见流式计算引擎提供的默认（甚至是唯一）计算模式 +2. 直接丢弃, 即 IGNORE EXPIRED 1：忽略过期数据 -2. 重新计算：从 TSDB 中重新查找对应窗口的所有数据并重新计算得到最新结果 无论在哪种模式下，watermark 都应该被妥善设置，来得到正确结果（直接丢弃模式）或避免频繁触发重算带来的性能开销（重新计算模式）。 diff --git a/docs/zh/12-taos-sql/watermark.webp b/docs/zh/12-taos-sql/watermark.webp new file mode 100644 index 0000000000000000000000000000000000000000..3307faccffdaaec6dddf5cad8b7c11016fd28bd4 Binary files /dev/null and b/docs/zh/12-taos-sql/watermark.webp differ diff --git a/docs/zh/14-reference/11-docker/index.md b/docs/zh/14-reference/11-docker/index.md index c03990ede293962b46b77e68825e68d1f3564ecc..743fc2d32f82778fb97e7879972cd23db1159c8e 100644 --- a/docs/zh/14-reference/11-docker/index.md +++ b/docs/zh/14-reference/11-docker/index.md @@ -25,10 +25,11 @@ curl -u root:taosdata -d "show databases" localhost:6041/rest/sql $ docker exec -it tdengine taos taos> show databases; - name | created_time | ntables | vgroups | replica | quorum | days | keep | cache(MB) | blocks | minrows | maxrows | wallevel | fsync | comp | cachelast | precision | update | status | -==================================================================================================================================================================================================================================================================================== - log | 2022-01-17 13:57:22.270 | 10 | 1 | 1 | 1 | 10 | 30 | 1 | 3 | 100 | 4096 | 1 | 3000 | 2 | 0 | us | 0 | ready | -Query OK, 1 row(s) in set (0.002843s) + name | +================================= + information_schema | + performance_schema | +Query OK, 2 rows in database (0.033802s) ``` 因为运行在容器中的 TDengine 服务端使用容器的 hostname 建立连接，使用 taos shell 或者各种连接器（例如 JDBC-JNI）从容器外访问容器内的 TDengine 比较复杂，所以上述方式是访问容器中 TDengine 服务的最简单的方法，适用于一些简单场景。如果在一些复杂场景下想要从容器化使用 taos shell 或者各种连接器访问容器中的 TDengine 服务，请参考下一节。 @@ -45,10 +46,11 @@ docker run -d --name tdengine --network host tdengine/tdengine $ taos taos> show dnodes; - id | end_point | vnodes | cores | status | role | create_time | offline reason | -====================================================================================================================================== - 1 | myhost:6030 | 1 | 8 | ready | any | 2022-01-17 22:10:32.619 | | -Query OK, 1 row(s) in set (0.003233s) + id | endpoint | vnodes | support_vnodes | status | create_time | note | +================================================================================================================================================= + 1 | vm98:6030 | 0 | 32 | ready | 2022-08-19 14:50:05.337 | | +Query OK, 1 rows in database (0.010654s) + ``` ## 以指定的 hostname 和 port 启动 TDengine @@ -59,12 +61,13 @@ Query OK, 1 row(s) in set (0.003233s) docker run -d \ --name tdengine \ -e TAOS_FQDN=tdengine \ - -p 6030-6049:6030-6049 \ - -p 6030-6049:6030-6049/udp \ + -p 6030:6030 \ + -p 6041-6049:6041-6049 \ + -p 6041-6049:6041-6049/udp \ tdengine/tdengine ``` -上面的命令在容器中启动一个 TDengine 服务，其所监听的 hostname 为 tdengine ，并将容器的 6030 到 6049 端口段映射到主机的 6030 到 6049 端口段 （tcp 和 udp 都需要映射)。如果主机上该端口段已经被占用，可以修改上述命令指定一个主机上空闲的端口段。如果 `rpcForceTcp` 被设置为 `1` ，可以只映射 tcp 协议。 +上面的命令在容器中启动一个 TDengine 服务，其所监听的 hostname 为 tdengine ，并将容器的 6030 端口映射到主机的 6030 端口（TCP，只能映射主机 6030 端口），6041-6049 端口段映射到主机 6041-6049 端口段（tcp 和 udp 都需要映射，如果主机上该端口段已经被占用，可以修改上述命令指定一个主机上空闲的端口段）。 接下来，要确保 "tdengine" 这个 hostname 在 `/etc/hosts` 中可解析。 @@ -103,9 +106,9 @@ taos -h tdengine -P 6030 3. 在同一网络上的另一容器中启动 TDengine 客户端 ```shell - docker run --rm -it --network td-net -e TAOS_FIRST_EP=tdengine tdengine/tdengine taos + docker run --rm -it --network td-net -e TAOS_FIRST_EP=tdengine --entrypoint=taos tdengine/tdengine # or - #docker run --rm -it --network td-net -e tdengine/tdengine taos -h tdengine + #docker run --rm -it --network td-net --entrypoint=taos tdengine/tdengine -h tdengine ``` ## 在容器中启动客户端应用 @@ -115,7 +118,7 @@ taos -h tdengine -P 6030 ```docker FROM ubuntu:20.04 RUN apt-get update && apt-get install -y wget -ENV TDENGINE_VERSION=2.4.0.0 +ENV TDENGINE_VERSION=3.0.0.0 RUN wget -c https://www.taosdata.com/assets-download/TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && tar xvf TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && cd TDengine-client-${TDENGINE_VERSION} \ @@ -129,6 +132,14 @@ RUN wget -c https://www.taosdata.com/assets-download/TDengine-client-${TDENGINE_ 以下是一个 go 应用程序的示例： +* 创建 go mod 项目： + +```bash +go mod init app +``` + +* 创建 main.go： + ```go /* * In this test program, we'll create a database and insert 4 records then select out. @@ -212,12 +223,18 @@ func checkErr(err error, prompt string) { } ``` -如下是完整版本的 dockerfile +* 更新 go mod -```docker -FROM golang:1.17.6-buster as builder -ENV TDENGINE_VERSION=2.4.0.0 -RUN wget -c https://www.taosdata.com/assets-download/TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ +```bash +go mod tidy +``` + +如下是完整版本的 dockerfile： + +```dockerfile +FROM golang:1.19.0-buster as builder +ENV TDENGINE_VERSION=3.0.0.0 +RUN wget -c https://www.taosdata.com/assets-download/3.0/TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && tar xvf TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && cd TDengine-client-${TDENGINE_VERSION} \ && ./install_client.sh \ @@ -232,8 +249,8 @@ RUN go build FROM ubuntu:20.04 RUN apt-get update && apt-get install -y wget -ENV TDENGINE_VERSION=2.4.0.0 -RUN wget -c https://www.taosdata.com/assets-download/TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ +ENV TDENGINE_VERSION=3.0.0.0 +RUN wget -c https://www.taosdata.com/assets-download/3.0/TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && tar xvf TDengine-client-${TDENGINE_VERSION}-Linux-x64.tar.gz \ && cd TDengine-client-${TDENGINE_VERSION} \ && ./install_client.sh \ @@ -248,113 +265,112 @@ CMD ["app"] 目前我们已经有了 `main.go`, `go.mod`, `go.sum`, `app.dockerfile`， 现在可以构建出这个应用程序并在 `td-net` 网络上启动它 ```shell -$ docker build -t app -f app.dockerfile -$ docker run --rm --network td-net app -h tdengine -p 6030 +$ docker build -t app -f app.dockerfile . +$ docker run --rm --network td-net app app -h tdengine -p 6030 ============= args parse result: ============= hostName: tdengine serverPort: 6030 usr: root password: taosdata ================================================ -2022-01-17 15:56:55.48 +0000 UTC 0 -2022-01-17 15:56:56.48 +0000 UTC 1 -2022-01-17 15:56:57.48 +0000 UTC 2 -2022-01-17 15:56:58.48 +0000 UTC 3 -2022-01-17 15:58:01.842 +0000 UTC 0 -2022-01-17 15:58:02.842 +0000 UTC 1 -2022-01-17 15:58:03.842 +0000 UTC 2 -2022-01-17 15:58:04.842 +0000 UTC 3 -2022-01-18 01:43:48.029 +0000 UTC 0 -2022-01-18 01:43:49.029 +0000 UTC 1 -2022-01-18 01:43:50.029 +0000 UTC 2 -2022-01-18 01:43:51.029 +0000 UTC 3 +2022-08-19 07:43:51.68 +0000 UTC 0 +2022-08-19 07:43:52.68 +0000 UTC 1 +2022-08-19 07:43:53.68 +0000 UTC 2 +2022-08-19 07:43:54.68 +0000 UTC 3 ``` ## 用 docker-compose 启动 TDengine 集群 -1. 如下 docker-compose 文件启动一个 2 副本、2 管理节点、2 数据节点以及 1 个 arbitrator 的 TDengine 集群。 - - ```docker - version: "3" - services: - arbitrator: - image: tdengine/tdengine:$VERSION - command: tarbitrator - td-1: - image: tdengine/tdengine:$VERSION - environment: - TAOS_FQDN: "td-1" - TAOS_FIRST_EP: "td-1" - TAOS_NUM_OF_MNODES: "2" - TAOS_REPLICA: "2" - TAOS_ARBITRATOR: arbitrator:6042 - volumes: - - taosdata-td1:/var/lib/taos/ - - taoslog-td1:/var/log/taos/ - td-2: - image: tdengine/tdengine:$VERSION - environment: - TAOS_FQDN: "td-2" - TAOS_FIRST_EP: "td-1" - TAOS_NUM_OF_MNODES: "2" - TAOS_REPLICA: "2" - TAOS_ARBITRATOR: arbitrator:6042 - volumes: - - taosdata-td2:/var/lib/taos/ - - taoslog-td2:/var/log/taos/ - volumes: - taosdata-td1: - taoslog-td1: - taosdata-td2: - taoslog-td2: - ``` +1. 如下 docker-compose 文件启动一个 三节点 TDengine 集群。 + +```yml +version: "3" +services: + td-1: + image: tdengine/tdengine:$VERSION + environment: + TAOS_FQDN: "td-1" + TAOS_FIRST_EP: "td-1" + volumes: + - taosdata-td1:/var/lib/taos/ + - taoslog-td1:/var/log/taos/ + td-2: + image: tdengine/tdengine:$VERSION + environment: + TAOS_FQDN: "td-2" + TAOS_FIRST_EP: "td-1" + volumes: + - taosdata-td2:/var/lib/taos/ + - taoslog-td2:/var/log/taos/ + td-3: + image: tdengine/tdengine:$VERSION + environment: + TAOS_FQDN: "td-3" + TAOS_FIRST_EP: "td-1" + volumes: + - taosdata-td3:/var/lib/taos/ + - taoslog-td3:/var/log/taos/ +volumes: + taosdata-td1: + taoslog-td1: + taosdata-td2: + taoslog-td2: + taosdata-td3: + taoslog-td3: +``` :::note -- `VERSION` 环境变量被用来设置 tdengine image tag -- 在新创建的实例上必须设置 `TAOS_FIRST_EP` 以使其能够加入 TDengine 集群；如果有高可用需求，则需要同时使用 `TAOS_SECOND_EP` -- `TAOS_REPLICA` 用来设置缺省的数据库副本数量，其取值范围为[1,3] - 在双副本环境下，推荐使用 arbitrator, 用 TAOS_ARBITRATOR 来设置 - ::: +* `VERSION` 环境变量被用来设置 tdengine image tag +* 在新创建的实例上必须设置 `TAOS_FIRST_EP` 以使其能够加入 TDengine 集群；如果有高可用需求，则需要同时使用 `TAOS_SECOND_EP` +::: 2. 启动集群 - ```shell - $ VERSION=2.4.0.0 docker-compose up -d - Creating network "test_default" with the default driver - Creating volume "test_taosdata-td1" with default driver - Creating volume "test_taoslog-td1" with default driver - Creating volume "test_taosdata-td2" with default driver - Creating volume "test_taoslog-td2" with default driver - Creating test_td-1_1 ... done - Creating test_arbitrator_1 ... done - Creating test_td-2_1 ... done - ``` +```shell +$ VERSION=3.0.0.0 docker-compose up -d +Creating network "test-docker_default" with the default driver +Creating volume "test-docker_taosdata-td1" with default driver +Creating volume "test-docker_taoslog-td1" with default driver +Creating volume "test-docker_taosdata-td2" with default driver +Creating volume "test-docker_taoslog-td2" with default driver +Creating volume "test-docker_taosdata-td3" with default driver +Creating volume "test-docker_taoslog-td3" with default driver + +Creating test-docker_td-3_1 ... done +Creating test-docker_td-1_1 ... done +Creating test-docker_td-2_1 ... done +``` 3. 查看节点状态 - ```shell - $ docker-compose ps - Name Command State Ports - --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- - test_arbitrator_1 /usr/bin/entrypoint.sh tar ... Up 6030/tcp, 6031/tcp, 6032/tcp, 6033/tcp, 6034/tcp, 6035/tcp, 6036/tcp, 6037/tcp, 6038/tcp, 6039/tcp, 6040/tcp, 6041/tcp, 6042/tcp - test_td-1_1 /usr/bin/entrypoint.sh taosd Up 6030/tcp, 6031/tcp, 6032/tcp, 6033/tcp, 6034/tcp, 6035/tcp, 6036/tcp, 6037/tcp, 6038/tcp, 6039/tcp, 6040/tcp, 6041/tcp, 6042/tcp - test_td-2_1 /usr/bin/entrypoint.sh taosd Up 6030/tcp, 6031/tcp, 6032/tcp, 6033/tcp, 6034/tcp, 6035/tcp, 6036/tcp, 6037/tcp, 6038/tcp, 6039/tcp, 6040/tcp, 6041/tcp, 6042/tcp - ``` +```shell + docker-compose ps + Name Command State Ports + +------------------------------------------------------------------- +test-docker_td-1_1 /tini -- /usr/bin/entrypoi ... Up +test-docker_td-2_1 /tini -- /usr/bin/entrypoi ... Up +test-docker_td-3_1 /tini -- /usr/bin/entrypoi ... Up +``` 4. 用 taos shell 查看 dnodes - ```shell - $ docker-compose exec td-1 taos -s "show dnodes" - - taos> show dnodes - id | end_point | vnodes | cores | status | role | create_time | offline reason | - ====================================================================================================================================== - 1 | td-1:6030 | 1 | 8 | ready | any | 2022-01-18 02:47:42.871 | | - 2 | td-2:6030 | 0 | 8 | ready | any | 2022-01-18 02:47:43.518 | | - 0 | arbitrator:6042 | 0 | 0 | ready | arb | 2022-01-18 02:47:43.633 | - | - Query OK, 3 row(s) in set (0.000811s) - ``` +```shell + +$ docker-compose exec td-1 taos -s "show dnodes" + +taos> show dnodes + + id | endpoint | vnodes | support_vnodes | status | create_time | note | +================================================================================================================================================= + + 1 | td-1:6030 | 0 | 32 | ready | 2022-08-19 07:57:29.971 | | + 2 | td-2:6030 | 0 | 32 | ready | 2022-08-19 07:57:31.415 | | + 3 | td-3:6030 | 0 | 32 | ready | 2022-08-19 07:57:31.417 | | +Query OK, 3 rows in database (0.021262s) + +``` ## taosAdapter @@ -362,93 +378,80 @@ password: taosdata 2. 同时为了部署灵活起见，可以在独立的容器中启动 taosAdapter - ```docker - services: - # ... - adapter: - image: tdengine/tdengine:$VERSION - command: taosadapter - ``` +```docker +services: + # ... + adapter: + image: tdengine/tdengine:$VERSION + command: taosadapter +``` - 如果要部署多个 taosAdapter 来提高吞吐量并提供高可用性，推荐配置方式为使用 nginx 等反向代理来提供统一的访问入口。具体配置方法请参考 nginx 的官方文档。如下是示例： - - ```docker - version: "3" - - networks: - inter: - api: - - services: - arbitrator: - image: tdengine/tdengine:$VERSION - command: tarbitrator - networks: - - inter - td-1: - image: tdengine/tdengine:$VERSION - networks: - - inter - environment: - TAOS_FQDN: "td-1" - TAOS_FIRST_EP: "td-1" - TAOS_NUM_OF_MNODES: "2" - TAOS_REPLICA: "2" - TAOS_ARBITRATOR: arbitrator:6042 - volumes: - - taosdata-td1:/var/lib/taos/ - - taoslog-td1:/var/log/taos/ - td-2: - image: tdengine/tdengine:$VERSION - networks: - - inter - environment: - TAOS_FQDN: "td-2" - TAOS_FIRST_EP: "td-1" - TAOS_NUM_OF_MNODES: "2" - TAOS_REPLICA: "2" - TAOS_ARBITRATOR: arbitrator:6042 - volumes: - - taosdata-td2:/var/lib/taos/ - - taoslog-td2:/var/log/taos/ - adapter: - image: tdengine/tdengine:$VERSION - command: taosadapter - networks: - - inter - environment: - TAOS_FIRST_EP: "td-1" - TAOS_SECOND_EP: "td-2" - deploy: - replicas: 4 - nginx: - image: nginx - depends_on: - - adapter - networks: - - inter - - api - ports: - - 6041:6041 - - 6044:6044/udp - command: [ - "sh", - "-c", - "while true; - do curl -s http://adapter:6041/-/ping >/dev/null && break; - done; - printf 'server{listen 6041;location /{proxy_pass http://adapter:6041;}}' - > /etc/nginx/conf.d/rest.conf; - printf 'stream{server{listen 6044 udp;proxy_pass adapter:6044;}}' - >> /etc/nginx/nginx.conf;cat /etc/nginx/nginx.conf; - nginx -g 'daemon off;'", - ] - volumes: - taosdata-td1: - taoslog-td1: - taosdata-td2: - taoslog-td2: - ``` +如果要部署多个 taosAdapter 来提高吞吐量并提供高可用性，推荐配置方式为使用 nginx 等反向代理来提供统一的访问入口。具体配置方法请参考 nginx 的官方文档。如下是示例： + +```yml +version: "3" + +networks: + inter: + +services: + td-1: + image: tdengine/tdengine:$VERSION + networks: + - inter + environment: + TAOS_FQDN: "td-1" + TAOS_FIRST_EP: "td-1" + volumes: + - taosdata-td1:/var/lib/taos/ + - taoslog-td1:/var/log/taos/ + td-2: + image: tdengine/tdengine:$VERSION + networks: + - inter + environment: + TAOS_FQDN: "td-2" + TAOS_FIRST_EP: "td-1" + volumes: + - taosdata-td2:/var/lib/taos/ + - taoslog-td2:/var/log/taos/ + adapter: + image: tdengine/tdengine:$VERSION + entrypoint: "taosadapter" + networks: + - inter + environment: + TAOS_FIRST_EP: "td-1" + TAOS_SECOND_EP: "td-2" + deploy: + replicas: 4 + nginx: + image: nginx + depends_on: + - adapter + networks: + - inter + ports: + - 6041:6041 + - 6044:6044/udp + command: [ + "sh", + "-c", + "while true; + do curl -s http://adapter:6041/-/ping >/dev/null && break; + done; + printf 'server{listen 6041;location /{proxy_pass http://adapter:6041;}}' + > /etc/nginx/conf.d/rest.conf; + printf 'stream{server{listen 6044 udp;proxy_pass adapter:6044;}}' + >> /etc/nginx/nginx.conf;cat /etc/nginx/nginx.conf; + nginx -g 'daemon off;'", + ] +volumes: + taosdata-td1: + taoslog-td1: + taosdata-td2: + taoslog-td2: +``` ## 使用 docker swarm 部署 @@ -457,50 +460,46 @@ password: taosdata docker-compose 文件可以参考上节。下面是使用 docker swarm 启动 TDengine 的命令： ```shell -$ VERSION=2.4.0 docker stack deploy -c docker-compose.yml taos +$ VERSION=3.0.0.0 docker stack deploy -c docker-compose.yml taos Creating network taos_inter -Creating network taos_api -Creating service taos_arbitrator +Creating service taos_nginx Creating service taos_td-1 Creating service taos_td-2 Creating service taos_adapter -Creating service taos_nginx ``` 查看和管理 ```shell $ docker stack ps taos -ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS -79ni8temw59n taos_nginx.1 nginx:latest TM1701 Running Running about a minute ago -3e94u72msiyg taos_adapter.1 tdengine/tdengine:2.4.0 TM1702 Running Running 56 seconds ago -100amjkwzsc6 taos_td-2.1 tdengine/tdengine:2.4.0 TM1703 Running Running about a minute ago -pkjehr2vvaaa taos_td-1.1 tdengine/tdengine:2.4.0 TM1704 Running Running 2 minutes ago -tpzvgpsr1qkt taos_arbitrator.1 tdengine/tdengine:2.4.0 TM1705 Running Running 2 minutes ago -rvss3g5yg6fa taos_adapter.2 tdengine/tdengine:2.4.0 TM1706 Running Running 56 seconds ago -i2augxamfllf taos_adapter.3 tdengine/tdengine:2.4.0 TM1707 Running Running 56 seconds ago -lmjyhzccpvpg taos_adapter.4 tdengine/tdengine:2.4.0 TM1708 Running Running 56 seconds ago +ID NAME IMAGE NODE DESIRED STATE CURRENT STATE ERROR PORTS +7m3sbf532bqp taos_adapter.1 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago +pj403n6ofmmh taos_adapter.2 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago +rxqfwsyk5q1h taos_adapter.3 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago +qj40lpxr40oc taos_adapter.4 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago +oe3455ulxpze taos_nginx.1 nginx:latest vm98 Running Running about a minute ago +o0tsg70nrrc6 taos_td-1.1 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago +q5m1oxs589cp taos_td-2.1 tdengine/tdengine:3.0.0.0 vm98 Running Running about a minute ago $ docker service ls -ID NAME MODE REPLICAS IMAGE PORTS -561t4lu6nfw6 taos_adapter replicated 4/4 tdengine/tdengine:2.4.0 -3hk5ct3q90sm taos_arbitrator replicated 1/1 tdengine/tdengine:2.4.0 -d8qr52envqzu taos_nginx replicated 1/1 nginx:latest *:6041->6041/tcp, *:6044->6044/udp -2isssfvjk747 taos_td-1 replicated 1/1 tdengine/tdengine:2.4.0 -9pzw7u02ichv taos_td-2 replicated 1/1 tdengine/tdengine:2.4.0 +ID NAME MODE REPLICAS IMAGE PORTS +ozuklorgl8bs taos_adapter replicated 4/4 tdengine/tdengine:3.0.0.0 +crmhdjw6vxw0 taos_nginx replicated 1/1 nginx:latest *:6041->6041/tcp, *:6044->6044/udp +o86ngy7csv5n taos_td-1 replicated 1/1 tdengine/tdengine:3.0.0.0 +rma040ny4tb0 taos_td-2 replicated 1/1 tdengine/tdengine:3.0.0.0 ``` -从上面的输出可以看到有两个 dnode， 和两个 taosAdapter，以及一个 nginx 反向代理服务。 +从上面的输出可以看到有两个 dnode， 和四个 taosAdapter，以及一个 nginx 反向代理服务。 接下来，我们可以减少 taosAdapter 服务的数量 ```shell $ docker service scale taos_adapter=1 taos_adapter scaled to 1 -overall progress: 1 out of 1 tasks -1/1: running [==================================================>] +overall progress: 1 out of 1 tasks +1/1: running [==================================================>] verify: Service converged $ docker service ls -f name=taos_adapter -ID NAME MODE REPLICAS IMAGE PORTS -561t4lu6nfw6 taos_adapter replicated 1/1 tdengine/tdengine:2.4.0 +ID NAME MODE REPLICAS IMAGE PORTS +ozuklorgl8bs taos_adapter replicated 1/1 tdengine/tdengine:3.0.0.0 ``` diff --git a/examples/JDBC/JDBCDemo/pom.xml b/examples/JDBC/JDBCDemo/pom.xml index 8cf0356721f8ffd568e87fa4a77c86eb0f90a62b..807ceb0f24644d3978274faee1bc8b47c9d7af47 100644 --- a/examples/JDBC/JDBCDemo/pom.xml +++ b/examples/JDBC/JDBCDemo/pom.xml @@ -17,7 +17,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.34 + 3.0.0 diff --git a/examples/JDBC/SpringJdbcTemplate/pom.xml b/examples/JDBC/SpringJdbcTemplate/pom.xml index eac3dec0a92a4c8aa519cd426b9c8d3895047be6..6e4941b4f1c5bb5557109d06496bff02744a3092 100644 --- a/examples/JDBC/SpringJdbcTemplate/pom.xml +++ b/examples/JDBC/SpringJdbcTemplate/pom.xml @@ -47,7 +47,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.18 + 3.0.0 diff --git a/examples/JDBC/SpringJdbcTemplate/readme.md b/examples/JDBC/SpringJdbcTemplate/readme.md index b70a6565f88d0a08b8a26a60676e729ecdb39e2e..f59bcdbeb547b0c0576b43abe4e1f2cef2175913 100644 --- a/examples/JDBC/SpringJdbcTemplate/readme.md +++ b/examples/JDBC/SpringJdbcTemplate/readme.md @@ -10,7 +10,7 @@ ```xml - + @@ -28,5 +28,5 @@ mvn clean package ``` 打包成功之后，进入 `target/` 目录下，执行以下命令就可运行测试： ```shell -java -jar SpringJdbcTemplate-1.0-SNAPSHOT-jar-with-dependencies.jar +java -jar target/SpringJdbcTemplate-1.0-SNAPSHOT-jar-with-dependencies.jar ``` \ No newline at end of file diff --git a/examples/JDBC/SpringJdbcTemplate/src/main/java/com/taosdata/example/jdbcTemplate/App.java b/examples/JDBC/SpringJdbcTemplate/src/main/java/com/taosdata/example/jdbcTemplate/App.java index 6942d62a83adafb85496a81ce93866cd0d53611d..ce26b7504ae41644032c1f59579efc310f58d527 100644 --- a/examples/JDBC/SpringJdbcTemplate/src/main/java/com/taosdata/example/jdbcTemplate/App.java +++ b/examples/JDBC/SpringJdbcTemplate/src/main/java/com/taosdata/example/jdbcTemplate/App.java @@ -28,7 +28,7 @@ public class App { //use database executor.doExecute("use test"); // create table - executor.doExecute("create table if not exists test.weather (ts timestamp, temperature int, humidity float)"); + executor.doExecute("create table if not exists test.weather (ts timestamp, temperature float, humidity int)"); WeatherDao weatherDao = ctx.getBean(WeatherDao.class); Weather weather = new Weather(new Timestamp(new Date().getTime()), random.nextFloat() * 50.0f, random.nextInt(100)); diff --git a/examples/JDBC/SpringJdbcTemplate/src/test/java/com/taosdata/example/jdbcTemplate/BatcherInsertTest.java b/examples/JDBC/SpringJdbcTemplate/src/test/java/com/taosdata/example/jdbcTemplate/BatcherInsertTest.java index 29d0f79fd4982d43078e590b4320c0df457ee44c..782fcbe0eb2020c8bcbafecb0b2d61185b139477 100644 --- a/examples/JDBC/SpringJdbcTemplate/src/test/java/com/taosdata/example/jdbcTemplate/BatcherInsertTest.java +++ b/examples/JDBC/SpringJdbcTemplate/src/test/java/com/taosdata/example/jdbcTemplate/BatcherInsertTest.java @@ -41,7 +41,7 @@ public class BatcherInsertTest { //use database executor.doExecute("use test"); // create table - executor.doExecute("create table if not exists test.weather (ts timestamp, temperature int, humidity float)"); + executor.doExecute("create table if not exists test.weather (ts timestamp, temperature float, humidity int)"); } @Test diff --git a/examples/JDBC/connectionPools/README-cn.md b/examples/JDBC/connectionPools/README-cn.md index 9b26df3c2eb2c23171a673643891a292af4c920c..6e589418b11a3d6c8c64d24b28a0ea7c65ad0830 100644 --- a/examples/JDBC/connectionPools/README-cn.md +++ b/examples/JDBC/connectionPools/README-cn.md @@ -13,13 +13,13 @@ ConnectionPoolDemo的程序逻辑： ### 如何运行这个例子： ```shell script -mvn clean package assembly:single -java -jar target/connectionPools-1.0-SNAPSHOT-jar-with-dependencies.jar -host 127.0.0.1 +mvn clean package +java -jar target/ConnectionPoolDemo-jar-with-dependencies.jar -host 127.0.0.1 ``` 使用mvn运行ConnectionPoolDemo的main方法，可以指定参数 ```shell script Usage: -java -jar target/connectionPools-1.0-SNAPSHOT-jar-with-dependencies.jar +java -jar target/ConnectionPoolDemo-jar-with-dependencies.jar -host : hostname -poolType -poolSize diff --git a/examples/JDBC/connectionPools/pom.xml b/examples/JDBC/connectionPools/pom.xml index 99a7892a250bd656479b0901682d6a86c2b27d14..61717cf1121696a97d867b5d43af75231ddd0472 100644 --- a/examples/JDBC/connectionPools/pom.xml +++ b/examples/JDBC/connectionPools/pom.xml @@ -18,7 +18,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.18 + 3.0.0 diff --git a/examples/JDBC/mybatisplus-demo/pom.xml b/examples/JDBC/mybatisplus-demo/pom.xml index ad6a63e800fb73dd3c768a8aca941f70cec235b3..5555145958de67fdf03eb744426afcfc13b6fcb3 100644 --- a/examples/JDBC/mybatisplus-demo/pom.xml +++ b/examples/JDBC/mybatisplus-demo/pom.xml @@ -47,7 +47,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.18 + 3.0.0 diff --git a/examples/JDBC/mybatisplus-demo/readme b/examples/JDBC/mybatisplus-demo/readme new file mode 100644 index 0000000000000000000000000000000000000000..b31b6c34bf1c2bd661d88fff066eb4632d456a1c --- /dev/null +++ b/examples/JDBC/mybatisplus-demo/readme @@ -0,0 +1,14 @@ +# 使用说明 + +## 创建使用db +```shell +$ taos + +> create database mp_test +``` + +## 执行测试用例 + +```shell +$ mvn clean test +``` \ No newline at end of file diff --git a/examples/JDBC/mybatisplus-demo/src/main/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapper.java b/examples/JDBC/mybatisplus-demo/src/main/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapper.java index 6733cbded9d1d180408eccaad9e8badad7d39a3d..1f0338db34019661a2d7c4a0716d953195d059a2 100644 --- a/examples/JDBC/mybatisplus-demo/src/main/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapper.java +++ b/examples/JDBC/mybatisplus-demo/src/main/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapper.java @@ -2,7 +2,17 @@ package com.taosdata.example.mybatisplusdemo.mapper; import com.baomidou.mybatisplus.core.mapper.BaseMapper; import com.taosdata.example.mybatisplusdemo.domain.Weather; +import org.apache.ibatis.annotations.Insert; +import org.apache.ibatis.annotations.Update; public interface WeatherMapper extends BaseMapper { + @Update("CREATE TABLE if not exists weather(ts timestamp, temperature float, humidity int, location nchar(100))") + int createTable(); + + @Insert("insert into weather (ts, temperature, humidity, location) values(#{ts}, #{temperature}, #{humidity}, #{location})") + int insertOne(Weather one); + + @Update("drop table if exists weather") + void dropTable(); } diff --git a/examples/JDBC/mybatisplus-demo/src/main/resources/application.yml b/examples/JDBC/mybatisplus-demo/src/main/resources/application.yml index 38180c6d75a620a63bcaab9ec350d97e65f9dd16..985ed1675ee408bad346dff2a1b7e03c5138f4df 100644 --- a/examples/JDBC/mybatisplus-demo/src/main/resources/application.yml +++ b/examples/JDBC/mybatisplus-demo/src/main/resources/application.yml @@ -2,7 +2,7 @@ spring: datasource: driver-class-name: com.taosdata.jdbc.TSDBDriver url: jdbc:TAOS://localhost:6030/mp_test?charset=UTF-8&locale=en_US.UTF-8&timezone=UTC-8 - user: root + username: root password: taosdata druid: diff --git a/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/TemperatureMapperTest.java b/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/TemperatureMapperTest.java index 4331d15d3476d3428e72a186664ed77cc59aad3e..4d9dbf8d2fb909ef46dbe23a2bb5192d4971195e 100644 --- a/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/TemperatureMapperTest.java +++ b/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/TemperatureMapperTest.java @@ -82,27 +82,15 @@ public class TemperatureMapperTest { Assert.assertEquals(1, affectRows); } - /*** - * test SelectOne - * **/ - @Test - public void testSelectOne() { - QueryWrapper wrapper = new QueryWrapper<>(); - wrapper.eq("location", "beijing"); - Temperature one = mapper.selectOne(wrapper); - System.out.println(one); - Assert.assertNotNull(one); - } - /*** * test select By map * ***/ @Test public void testSelectByMap() { Map map = new HashMap<>(); - map.put("location", "beijing"); + map.put("location", "北京"); List temperatures = mapper.selectByMap(map); - Assert.assertEquals(1, temperatures.size()); + Assert.assertTrue(temperatures.size() > 1); } /*** @@ -120,7 +108,7 @@ public class TemperatureMapperTest { @Test public void testSelectCount() { int count = mapper.selectCount(null); - Assert.assertEquals(5, count); + Assert.assertEquals(10, count); } /**** diff --git a/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapperTest.java b/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapperTest.java index 1699344552f89e1595d1317019c992dcd3820e77..dba8abd1ed006e81cf8240e66cfcc0b525af9b79 100644 --- a/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapperTest.java +++ b/examples/JDBC/mybatisplus-demo/src/test/java/com/taosdata/example/mybatisplusdemo/mapper/WeatherMapperTest.java @@ -6,6 +6,7 @@ import com.baomidou.mybatisplus.extension.plugins.pagination.Page; import com.taosdata.example.mybatisplusdemo.domain.Weather; import org.junit.Assert; import org.junit.Test; +import org.junit.Before; import org.junit.runner.RunWith; import org.springframework.beans.factory.annotation.Autowired; import org.springframework.boot.test.context.SpringBootTest; @@ -26,6 +27,18 @@ public class WeatherMapperTest { @Autowired private WeatherMapper mapper; + @Before + public void createTable(){ + mapper.dropTable(); + mapper.createTable(); + Weather one = new Weather(); + one.setTs(new Timestamp(1605024000000l)); + one.setTemperature(12.22f); + one.setLocation("望京"); + one.setHumidity(100); + mapper.insertOne(one); + } + @Test public void testSelectList() { List weathers = mapper.selectList(null); @@ -46,20 +59,20 @@ public class WeatherMapperTest { @Test public void testSelectOne() { QueryWrapper wrapper = new QueryWrapper<>(); - wrapper.eq("location", "beijing"); + wrapper.eq("location", "望京"); Weather one = mapper.selectOne(wrapper); System.out.println(one); Assert.assertEquals(12.22f, one.getTemperature(), 0.00f); - Assert.assertEquals("beijing", one.getLocation()); + Assert.assertEquals("望京", one.getLocation()); } - @Test - public void testSelectByMap() { - Map map = new HashMap<>(); - map.put("location", "beijing"); - List weathers = mapper.selectByMap(map); - Assert.assertEquals(1, weathers.size()); - } + // @Test + // public void testSelectByMap() { + // Map map = new HashMap<>(); + // map.put("location", "beijing"); + // List weathers = mapper.selectByMap(map); + // Assert.assertEquals(1, weathers.size()); + // } @Test public void testSelectObjs() { diff --git a/examples/JDBC/readme.md b/examples/JDBC/readme.md index 9a017f4feab148cb7c3fd4132360c3075c6573cb..c7d7875308d248c1abef8d47bc69a69e91374dbb 100644 --- a/examples/JDBC/readme.md +++ b/examples/JDBC/readme.md @@ -10,4 +10,4 @@ | 6 | taosdemo | This is an internal tool for testing Our JDBC-JNI, JDBC-RESTful, RESTful interfaces | -more detail: https://www.taosdata.com/cn//documentation20/connector-java/ \ No newline at end of file +more detail: https://docs.taosdata.com/reference/connector/java/ \ No newline at end of file diff --git a/examples/JDBC/springbootdemo/pom.xml b/examples/JDBC/springbootdemo/pom.xml index 9126813b67e71691692109920f891a6fb4cc5ab5..ee15f6013e4fd35bf30fb5af00b226e7c4d3d8c7 100644 --- a/examples/JDBC/springbootdemo/pom.xml +++ b/examples/JDBC/springbootdemo/pom.xml @@ -68,7 +68,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.34 + 3.0.0 diff --git a/examples/JDBC/springbootdemo/readme.md b/examples/JDBC/springbootdemo/readme.md index 67a28947d2dfb8fc069bf94fd139a7006d35a22b..a3942a6a512501b7dee1f4f4ff5ccc93da0babbb 100644 --- a/examples/JDBC/springbootdemo/readme.md +++ b/examples/JDBC/springbootdemo/readme.md @@ -1,10 +1,11 @@ ## TDengine SpringBoot + Mybatis Demo +## 需要提前创建 test 数据库 ### 配置 application.properties ```properties # datasource config spring.datasource.driver-class-name=com.taosdata.jdbc.TSDBDriver -spring.datasource.url=jdbc:TAOS://127.0.0.1:6030/log +spring.datasource.url=jdbc:TAOS://127.0.0.1:6030/test spring.datasource.username=root spring.datasource.password=taosdata diff --git a/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/controller/WeatherController.java b/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/controller/WeatherController.java index ed720fe6c02dd3a7eba6e645ea1e76d704c04d0c..3ee5b597ab08c945f6494d9a8a31da9cd3e01f25 100644 --- a/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/controller/WeatherController.java +++ b/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/controller/WeatherController.java @@ -6,7 +6,6 @@ import org.springframework.beans.factory.annotation.Autowired; import org.springframework.web.bind.annotation.*; import java.util.List; -import java.util.Map; @RequestMapping("/weather") @RestController diff --git a/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/dao/WeatherMapper.xml b/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/dao/WeatherMapper.xml index 91938ca24e3cf9c3e0f2895cf40f214d484c55d5..99d5893ec198535d9e8ef1cc6c443625d0a64ec1 100644 --- a/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/dao/WeatherMapper.xml +++ b/examples/JDBC/springbootdemo/src/main/java/com/taosdata/example/springbootdemo/dao/WeatherMapper.xml @@ -10,8 +10,7 @@ diff --git a/examples/JDBC/springbootdemo/src/main/resources/application.properties b/examples/JDBC/springbootdemo/src/main/resources/application.properties index 06daa81bbb06450d99ab3f6e640c9795c0ad5d2e..bf21047395ed534e4c7d9db919bb371fab45ec16 100644 --- a/examples/JDBC/springbootdemo/src/main/resources/application.properties +++ b/examples/JDBC/springbootdemo/src/main/resources/application.properties @@ -5,7 +5,7 @@ #spring.datasource.password=taosdata # datasource config - JDBC-RESTful spring.datasource.driver-class-name=com.taosdata.jdbc.rs.RestfulDriver -spring.datasource.url=jdbc:TAOS-RS://localhsot:6041/test?timezone=UTC-8&charset=UTF-8&locale=en_US.UTF-8 +spring.datasource.url=jdbc:TAOS-RS://localhost:6041/test?timezone=UTC-8&charset=UTF-8&locale=en_US.UTF-8 spring.datasource.username=root spring.datasource.password=taosdata spring.datasource.druid.initial-size=5 diff --git a/examples/JDBC/taosdemo/pom.xml b/examples/JDBC/taosdemo/pom.xml index 07fd4a3576243b8950ccd25515f2512226e313d6..724ecc74077c4080269c695ca50a1cf300e39d0b 100644 --- a/examples/JDBC/taosdemo/pom.xml +++ b/examples/JDBC/taosdemo/pom.xml @@ -67,7 +67,7 @@ com.taosdata.jdbc taos-jdbcdriver - 2.0.20 + 3.0.0 diff --git a/examples/JDBC/taosdemo/readme.md b/examples/JDBC/taosdemo/readme.md index 451fa2960adb98e2deb8499732aefde11f4810a1..e5f4eb132b2262990b8fa32fe3c40a617d16d247 100644 --- a/examples/JDBC/taosdemo/readme.md +++ b/examples/JDBC/taosdemo/readme.md @@ -2,9 +2,9 @@ cd tests/examples/JDBC/taosdemo mvn clean package -Dmaven.test.skip=true # 先建表，再插入的 -java -jar target/taosdemo-2.0-jar-with-dependencies.jar -host [hostname] -database [database] -doCreateTable true -superTableSQL "create table weather(ts timestamp, f1 int) tags(t1 nchar(4))" -numOfTables 1000 -numOfRowsPerTable 100000000 -numOfThreadsForInsert 10 -numOfTablesPerSQL 10 -numOfValuesPerSQL 100 +java -jar target/taosdemo-2.0.1-jar-with-dependencies.jar -host [hostname] -database [database] -doCreateTable true -superTableSQL "create table weather(ts timestamp, f1 int) tags(t1 nchar(4))" -numOfTables 1000 -numOfRowsPerTable 100000000 -numOfThreadsForInsert 10 -numOfTablesPerSQL 10 -numOfValuesPerSQL 100 # 不建表，直接插入的 -java -jar target/taosdemo-2.0-jar-with-dependencies.jar -host [hostname] -database [database] -doCreateTable false -superTableSQL "create table weather(ts timestamp, f1 int) tags(t1 nchar(4))" -numOfTables 1000 -numOfRowsPerTable 100000000 -numOfThreadsForInsert 10 -numOfTablesPerSQL 10 -numOfValuesPerSQL 100 +java -jar target/taosdemo-2.0.1-jar-with-dependencies.jar -host [hostname] -database [database] -doCreateTable false -superTableSQL "create table weather(ts timestamp, f1 int) tags(t1 nchar(4))" -numOfTables 1000 -numOfRowsPerTable 100000000 -numOfThreadsForInsert 10 -numOfTablesPerSQL 10 -numOfValuesPerSQL 100 ``` 需求： diff --git a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/TaosDemoApplication.java b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/TaosDemoApplication.java index d4f5ff26886b9f90a4235d47bfd004dae9de93f6..6854054703776da46abdbff593724bef179f5b6d 100644 --- a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/TaosDemoApplication.java +++ b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/TaosDemoApplication.java @@ -32,8 +32,10 @@ public class TaosDemoApplication { System.exit(0); } // 初始化 - final DataSource dataSource = DataSourceFactory.getInstance(config.host, config.port, config.user, config.password); - if (config.executeSql != null && !config.executeSql.isEmpty() && !config.executeSql.replaceAll("\\s", "").isEmpty()) { + final DataSource dataSource = DataSourceFactory.getInstance(config.host, config.port, config.user, + config.password); + if (config.executeSql != null && !config.executeSql.isEmpty() + && !config.executeSql.replaceAll("\\s", "").isEmpty()) { Thread task = new Thread(new SqlExecuteTask(dataSource, config.executeSql)); task.start(); try { @@ -55,7 +57,7 @@ public class TaosDemoApplication { databaseParam.put("keep", Integer.toString(config.keep)); databaseParam.put("days", Integer.toString(config.days)); databaseParam.put("replica", Integer.toString(config.replica)); - //TODO: other database parameters + // TODO: other database parameters databaseService.createDatabase(databaseParam); databaseService.useDatabase(config.database); long end = System.currentTimeMillis(); @@ -70,11 +72,13 @@ public class TaosDemoApplication { if (config.database != null && !config.database.isEmpty()) superTableMeta.setDatabase(config.database); } else if (config.numOfFields == 0) { - String sql = "create table " + config.database + "." + config.superTable + " (ts timestamp, temperature float, humidity int) tags(location nchar(64), groupId int)"; + String sql = "create table " + config.database + "." + config.superTable + + " (ts timestamp, temperature float, humidity int) tags(location nchar(64), groupId int)"; superTableMeta = SuperTableMetaGenerator.generate(sql); } else { // create super table with specified field size and tag size - superTableMeta = SuperTableMetaGenerator.generate(config.database, config.superTable, config.numOfFields, config.prefixOfFields, config.numOfTags, config.prefixOfTags); + superTableMeta = SuperTableMetaGenerator.generate(config.database, config.superTable, config.numOfFields, + config.prefixOfFields, config.numOfTags, config.prefixOfTags); } /**********************************************************************************/ // 建表 @@ -84,7 +88,8 @@ public class TaosDemoApplication { superTableService.create(superTableMeta); if (!config.autoCreateTable) { // 批量建子表 - subTableService.createSubTable(superTableMeta, config.numOfTables, config.prefixOfTable, config.numOfThreadsForCreate); + subTableService.createSubTable(superTableMeta, config.numOfTables, config.prefixOfTable, + config.numOfThreadsForCreate); } } end = System.currentTimeMillis(); @@ -93,7 +98,7 @@ public class TaosDemoApplication { // 插入 long tableSize = config.numOfTables; int threadSize = config.numOfThreadsForInsert; - long startTime = getProperStartTime(config.startTime, config.keep); + long startTime = getProperStartTime(config.startTime, config.days); if (tableSize < threadSize) threadSize = (int) tableSize; @@ -101,13 +106,13 @@ public class TaosDemoApplication { start = System.currentTimeMillis(); // multi threads to insert - int affectedRows = subTableService.insertMultiThreads(superTableMeta, threadSize, tableSize, startTime, gap, config); + int affectedRows = subTableService.insertMultiThreads(superTableMeta, threadSize, tableSize, startTime, gap, + config); end = System.currentTimeMillis(); logger.info("insert " + affectedRows + " rows, time cost: " + (end - start) + " ms"); /**********************************************************************************/ // 查询 - /**********************************************************************************/ // 删除表 if (config.dropTable) { diff --git a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/service/QueryService.java b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/service/QueryService.java index efabff6afe904516ad9682cd7197412dc02765ef..ab0a1125d2b879d7e889e4c76cdb021ec46292f7 100644 --- a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/service/QueryService.java +++ b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/service/QueryService.java @@ -1,7 +1,5 @@ package com.taosdata.taosdemo.service; -import com.taosdata.jdbc.utils.SqlSyntaxValidator; - import javax.sql.DataSource; import java.sql.*; import java.util.ArrayList; @@ -23,10 +21,6 @@ public class QueryService { Boolean[] ret = new Boolean[sqls.length]; for (int i = 0; i < sqls.length; i++) { ret[i] = true; - if (!SqlSyntaxValidator.isValidForExecuteQuery(sqls[i])) { - ret[i] = false; - continue; - } try (Connection conn = dataSource.getConnection(); Statement stmt = conn.createStatement()) { stmt.executeQuery(sqls[i]); } catch (SQLException e) { diff --git a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/utils/SqlSpeller.java b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/utils/SqlSpeller.java index a60f0641d3a4441195c3a60639fbe3a197115dc3..7651d1e31814981499eb69d669b9176c73f33acd 100644 --- a/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/utils/SqlSpeller.java +++ b/examples/JDBC/taosdemo/src/main/java/com/taosdata/taosdemo/utils/SqlSpeller.java @@ -15,9 +15,12 @@ public class SqlSpeller { StringBuilder sb = new StringBuilder(); sb.append("create database if not exists ").append(map.get("database")).append(" "); if (map.containsKey("keep")) - sb.append("keep ").append(map.get("keep")).append(" "); - if (map.containsKey("days")) - sb.append("days ").append(map.get("days")).append(" "); + sb.append("keep "); + if (map.containsKey("days")) { + sb.append(map.get("days")).append("d "); + } else { + sb.append(" "); + } if (map.containsKey("replica")) sb.append("replica ").append(map.get("replica")).append(" "); if (map.containsKey("cache")) @@ -29,7 +32,7 @@ public class SqlSpeller { if (map.containsKey("maxrows")) sb.append("maxrows ").append(map.get("maxrows")).append(" "); if (map.containsKey("precision")) - sb.append("precision ").append(map.get("precision")).append(" "); + sb.append("precision '").append(map.get("precision")).append("' "); if (map.containsKey("comp")) sb.append("comp ").append(map.get("comp")).append(" "); if (map.containsKey("walLevel")) @@ -46,11 +49,13 @@ public class SqlSpeller { // create table if not exists xx.xx using xx.xx tags(x,x,x) public static String createTableUsingSuperTable(SubTableMeta subTableMeta) { StringBuilder sb = new StringBuilder(); - sb.append("create table if not exists ").append(subTableMeta.getDatabase()).append(".").append(subTableMeta.getName()).append(" "); - sb.append("using ").append(subTableMeta.getDatabase()).append(".").append(subTableMeta.getSupertable()).append(" "); -// String tagStr = subTableMeta.getTags().stream().filter(Objects::nonNull) -// .map(tagValue -> tagValue.getName() + " '" + tagValue.getValue() + "' ") -// .collect(Collectors.joining(",", "(", ")")); + sb.append("create table if not exists ").append(subTableMeta.getDatabase()).append(".") + .append(subTableMeta.getName()).append(" "); + sb.append("using ").append(subTableMeta.getDatabase()).append(".").append(subTableMeta.getSupertable()) + .append(" "); + // String tagStr = subTableMeta.getTags().stream().filter(Objects::nonNull) + // .map(tagValue -> tagValue.getName() + " '" + tagValue.getValue() + "' ") + // .collect(Collectors.joining(",", "(", ")")); sb.append("tags ").append(tagValues(subTableMeta.getTags())); return sb.toString(); } @@ -63,7 +68,7 @@ public class SqlSpeller { return sb.toString(); } - //f1, f2, f3 + // f1, f2, f3 private static String fieldValues(List fields) { return IntStream.range(0, fields.size()).mapToObj(i -> { if (i == 0) { @@ -73,13 +78,13 @@ public class SqlSpeller { } }).collect(Collectors.joining(",", "(", ")")); -// return fields.stream() -// .filter(Objects::nonNull) -// .map(fieldValue -> "'" + fieldValue.getValue() + "'") -// .collect(Collectors.joining(",", "(", ")")); + // return fields.stream() + // .filter(Objects::nonNull) + // .map(fieldValue -> "'" + fieldValue.getValue() + "'") + // .collect(Collectors.joining(",", "(", ")")); } - //(f1, f2, f3),(f1, f2, f3) + // (f1, f2, f3),(f1, f2, f3) private static String rowValues(List rowValues) { return rowValues.stream().filter(Objects::nonNull) .map(rowValue -> fieldValues(rowValue.getFields())) @@ -89,8 +94,10 @@ public class SqlSpeller { // insert into xx.xxx using xx.xx tags(x,x,x) values(x,x,x),(x,x,x)... public static String insertOneTableMultiValuesUsingSuperTable(SubTableValue subTableValue) { StringBuilder sb = new StringBuilder(); - sb.append("insert into ").append(subTableValue.getDatabase()).append(".").append(subTableValue.getName()).append(" "); - sb.append("using ").append(subTableValue.getDatabase()).append(".").append(subTableValue.getSupertable()).append(" "); + sb.append("insert into ").append(subTableValue.getDatabase()).append(".").append(subTableValue.getName()) + .append(" "); + sb.append("using ").append(subTableValue.getDatabase()).append(".").append(subTableValue.getSupertable()) + .append(" "); sb.append("tags ").append(tagValues(subTableValue.getTags()) + " "); sb.append("values ").append(rowValues(subTableValue.getValues())); return sb.toString(); @@ -126,7 +133,8 @@ public class SqlSpeller { // create table if not exists xx.xx (f1 xx,f2 xx...) tags(t1 xx, t2 xx...) public static String createSuperTable(SuperTableMeta tableMetadata) { StringBuilder sb = new StringBuilder(); - sb.append("create table if not exists ").append(tableMetadata.getDatabase()).append(".").append(tableMetadata.getName()); + sb.append("create table if not exists ").append(tableMetadata.getDatabase()).append(".") + .append(tableMetadata.getName()); String fields = tableMetadata.getFields().stream() .filter(Objects::nonNull).map(field -> field.getName() + " " + field.getType() + " ") .collect(Collectors.joining(",", "(", ")")); @@ -139,10 +147,10 @@ public class SqlSpeller { return sb.toString(); } - public static String createTable(TableMeta tableMeta) { StringBuilder sb = new StringBuilder(); - sb.append("create table if not exists ").append(tableMeta.getDatabase()).append(".").append(tableMeta.getName()).append(" "); + sb.append("create table if not exists ").append(tableMeta.getDatabase()).append(".").append(tableMeta.getName()) + .append(" "); String fields = tableMeta.getFields().stream() .filter(Objects::nonNull).map(field -> field.getName() + " " + field.getType() + " ") .collect(Collectors.joining(",", "(", ")")); @@ -179,16 +187,17 @@ public class SqlSpeller { public static String insertMultiTableMultiValuesWithColumns(List tables) { StringBuilder sb = new StringBuilder(); sb.append("insert into ").append(tables.stream().filter(Objects::nonNull) - .map(table -> table.getDatabase() + "." + table.getName() + " " + columnNames(table.getColumns()) + " values " + rowValues(table.getValues())) + .map(table -> table.getDatabase() + "." + table.getName() + " " + columnNames(table.getColumns()) + + " values " + rowValues(table.getValues())) .collect(Collectors.joining(" "))); return sb.toString(); } public static String insertMultiTableMultiValues(List tables) { StringBuilder sb = new StringBuilder(); - sb.append("insert into ").append(tables.stream().filter(Objects::nonNull).map(table -> - table.getDatabase() + "." + table.getName() + " values " + rowValues(table.getValues()) - ).collect(Collectors.joining(" "))); + sb.append("insert into ").append(tables.stream().filter(Objects::nonNull) + .map(table -> table.getDatabase() + "." + table.getName() + " values " + rowValues(table.getValues())) + .collect(Collectors.joining(" "))); return sb.toString(); } } diff --git a/examples/JDBC/taosdemo/src/main/resources/application.properties b/examples/JDBC/taosdemo/src/main/resources/application.properties index 488185196f1d2325fd9896d30068cbb202180a3f..4f550f6523587c060bbb2ed889024e1653fb0cb6 100644 --- a/examples/JDBC/taosdemo/src/main/resources/application.properties +++ b/examples/JDBC/taosdemo/src/main/resources/application.properties @@ -1,5 +1,5 @@ -jdbc.driver=com.taosdata.jdbc.rs.RestfulDriver -#jdbc.driver=com.taosdata.jdbc.TSDBDriver +# jdbc.driver=com.taosdata.jdbc.rs.RestfulDriver +jdbc.driver=com.taosdata.jdbc.TSDBDriver hikari.maximum-pool-size=20 hikari.minimum-idle=20 hikari.max-lifetime=0 \ No newline at end of file diff --git a/examples/JDBC/taosdemo/src/test/java/com/taosdata/taosdemo/service/TableServiceTest.java b/examples/JDBC/taosdemo/src/test/java/com/taosdata/taosdemo/service/TableServiceTest.java deleted file mode 100644 index 1f52198d68823326dd81d8c419fc02d89e15ef2d..0000000000000000000000000000000000000000 --- a/examples/JDBC/taosdemo/src/test/java/com/taosdata/taosdemo/service/TableServiceTest.java +++ /dev/null @@ -1,31 +0,0 @@ -package com.taosdata.taosdemo.service; - -import com.taosdata.taosdemo.domain.TableMeta; -import org.junit.Before; -import org.junit.Test; - -import java.util.ArrayList; -import java.util.List; - -public class TableServiceTest { - private TableService tableService; - - private List tables; - - @Before - public void before() { - tables = new ArrayList<>(); - for (int i = 0; i < 1; i++) { - TableMeta tableMeta = new TableMeta(); - tableMeta.setDatabase("test"); - tableMeta.setName("weather" + (i + 1)); - tables.add(tableMeta); - } - } - - @Test - public void testCreate() { - tableService.create(tables); - } - -} \ No newline at end of file diff --git a/include/common/tdataformat.h b/include/common/tdataformat.h index 3679b3773b2c9a89147e162b568479cfb44912db..af7c88acded2e151ff730ccb1ade5fdf15f9862a 100644 --- a/include/common/tdataformat.h +++ b/include/common/tdataformat.h @@ -38,22 +38,18 @@ typedef struct STagVal STagVal; typedef struct STag STag; // bitmap -#define N1(n) ((1 << (n)) - 1) -#define BIT1_SIZE(n) (((n)-1) / 8 + 1) -#define BIT2_SIZE(n) (((n)-1) / 4 + 1) -#define SET_BIT1(p, i, v) \ - do { \ - (p)[(i) / 8] &= N1((i) % 8); \ - (p)[(i) / 8] |= (((uint8_t)(v)) << (((i) % 8))); \ - } while (0) - -#define GET_BIT1(p, i) (((p)[(i) / 8] >> ((i) % 8)) & ((uint8_t)1)) -#define SET_BIT2(p, i, v) \ - do { \ - p[(i) / 4] &= N1((i) % 4 * 2); \ - (p)[(i) / 4] |= (((uint8_t)(v)) << (((i) % 4) * 2)); \ - } while (0) -#define GET_BIT2(p, i) (((p)[(i) / 4] >> (((i) % 4) * 2)) & ((uint8_t)3)) +const static uint8_t BIT2_MAP[4][4] = {{0b00000000, 0b00000001, 0b00000010, 0}, + {0b00000000, 0b00000100, 0b00001000, 2}, + {0b00000000, 0b00010000, 0b00100000, 4}, + {0b00000000, 0b01000000, 0b10000000, 6}}; + +#define N1(n) ((((uint8_t)1) << (n)) - 1) +#define BIT1_SIZE(n) ((((n)-1) >> 3) + 1) +#define BIT2_SIZE(n) ((((n)-1) >> 2) + 1) +#define SET_BIT1(p, i, v) ((p)[(i) >> 3] = (p)[(i) >> 3] & N1((i)&7) | (((uint8_t)(v)) << ((i)&7))) +#define GET_BIT1(p, i) (((p)[(i) >> 3] >> ((i)&7)) & ((uint8_t)1)) +#define SET_BIT2(p, i, v) ((p)[(i) >> 2] = (p)[(i) >> 2] & N1(BIT2_MAP[(i)&3][3]) | BIT2_MAP[(i)&3][(v)]) +#define GET_BIT2(p, i) (((p)[(i) >> 2] >> BIT2_MAP[(i)&3][3]) & ((uint8_t)3)) // STSchema int32_t tTSchemaCreate(int32_t sver, SSchema *pSchema, int32_t nCols, STSchema **ppTSchema); @@ -171,7 +167,7 @@ struct SColVal { #pragma pack(push, 1) struct STagVal { -// char colName[TSDB_COL_NAME_LEN]; // only used for tmq_get_meta + // char colName[TSDB_COL_NAME_LEN]; // only used for tmq_get_meta union { int16_t cid; char *pKey; diff --git a/include/common/tmsg.h b/include/common/tmsg.h index f870bd161fb22f04af6ff2c1cb15056ecf4b4289..8f199c72f7284e5a1a5192fad4f0fdd7a292bab2 100644 --- a/include/common/tmsg.h +++ b/include/common/tmsg.h @@ -2667,31 +2667,6 @@ typedef struct { int32_t padding; } SRSmaExecMsg; -typedef struct { - int64_t suid; - int8_t level; -} SRSmaFetchMsg; - -static FORCE_INLINE int32_t tEncodeSRSmaFetchMsg(SEncoder* pCoder, const SRSmaFetchMsg* pReq) { - if (tStartEncode(pCoder) < 0) return -1; - - if (tEncodeI64(pCoder, pReq->suid) < 0) return -1; - if (tEncodeI8(pCoder, pReq->level) < 0) return -1; - - tEndEncode(pCoder); - return 0; -} - -static FORCE_INLINE int32_t tDecodeSRSmaFetchMsg(SDecoder* pCoder, SRSmaFetchMsg* pReq) { - if (tStartDecode(pCoder) < 0) return -1; - - if (tDecodeI64(pCoder, &pReq->suid) < 0) return -1; - if (tDecodeI8(pCoder, &pReq->level) < 0) return -1; - - tEndDecode(pCoder); - return 0; -} - typedef struct { int8_t version; // for compatibility(default 0) int8_t intervalUnit; // MACRO: TIME_UNIT_XXX diff --git a/include/common/tmsgdef.h b/include/common/tmsgdef.h index 16d596575936a6e7c74b60c4b50fc9e023597483..e2bb3e2ae16921b822c275fb1d9be7afcae29685 100644 --- a/include/common/tmsgdef.h +++ b/include/common/tmsgdef.h @@ -201,7 +201,7 @@ enum { TD_DEF_MSG_TYPE(TDMT_VND_CANCEL_SMA, "vnode-cancel-sma", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_DROP_SMA, "vnode-drop-sma", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_SUBMIT_RSMA, "vnode-submit-rsma", SSubmitReq, SSubmitRsp) - TD_DEF_MSG_TYPE(TDMT_VND_FETCH_RSMA, "vnode-fetch-rsma", SRSmaFetchMsg, NULL) + TD_DEF_MSG_TYPE(TDMT_VND_FETCH_RSMA, "vnode-fetch-rsma", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_EXEC_RSMA, "vnode-exec-rsma", NULL, NULL) TD_DEF_MSG_TYPE(TDMT_VND_DELETE, "delete-data", SVDeleteReq, SVDeleteRsp) TD_DEF_MSG_TYPE(TDMT_VND_BATCH_DEL, "batch-delete", SBatchDeleteReq, NULL) diff --git a/include/libs/nodes/nodes.h b/include/libs/nodes/nodes.h index bb75efa00ac23f163e9e2eec94df0560e78fb463..5743d3360857dab460841d89e50360ba53d36b39 100644 --- a/include/libs/nodes/nodes.h +++ b/include/libs/nodes/nodes.h @@ -105,7 +105,7 @@ typedef enum ENodeType { QUERY_NODE_COLUMN_REF, // Statement nodes are used in parser and planner module. - QUERY_NODE_SET_OPERATOR, + QUERY_NODE_SET_OPERATOR = 100, QUERY_NODE_SELECT_STMT, QUERY_NODE_VNODE_MODIF_STMT, QUERY_NODE_CREATE_DATABASE_STMT, @@ -198,7 +198,7 @@ typedef enum ENodeType { QUERY_NODE_QUERY, // logic plan node - QUERY_NODE_LOGIC_PLAN_SCAN, + QUERY_NODE_LOGIC_PLAN_SCAN = 1000, QUERY_NODE_LOGIC_PLAN_JOIN, QUERY_NODE_LOGIC_PLAN_AGG, QUERY_NODE_LOGIC_PLAN_PROJECT, @@ -215,7 +215,7 @@ typedef enum ENodeType { QUERY_NODE_LOGIC_PLAN, // physical plan node - QUERY_NODE_PHYSICAL_PLAN_TAG_SCAN, + QUERY_NODE_PHYSICAL_PLAN_TAG_SCAN = 1100, QUERY_NODE_PHYSICAL_PLAN_TABLE_SCAN, QUERY_NODE_PHYSICAL_PLAN_TABLE_SEQ_SCAN, QUERY_NODE_PHYSICAL_PLAN_TABLE_MERGE_SCAN, diff --git a/include/libs/nodes/querynodes.h b/include/libs/nodes/querynodes.h index 088da73a1ad1785322ebc069d8e932e6287c9fc6..e1f86bae58ec09600792c89f567850cf50470fa9 100644 --- a/include/libs/nodes/querynodes.h +++ b/include/libs/nodes/querynodes.h @@ -428,6 +428,9 @@ void nodesValueNodeToVariant(const SValueNode* pNode, SVariant* pVal); char* nodesGetFillModeString(EFillMode mode); int32_t nodesMergeConds(SNode** pDst, SNodeList** pSrc); +const char* operatorTypeStr(EOperatorType type); +const char* logicConditionTypeStr(ELogicConditionType type); + #ifdef __cplusplus } #endif diff --git a/include/util/tdef.h b/include/util/tdef.h index a3deb73fd4b666872204e64093565c460e7d2548..6ce157165613e4dbfded97c4c41f2060e2c4853c 100644 --- a/include/util/tdef.h +++ b/include/util/tdef.h @@ -132,15 +132,14 @@ typedef enum EOperatorType { OP_TYPE_DIV, OP_TYPE_REM, // unary arithmetic operator - OP_TYPE_MINUS, - OP_TYPE_ASSIGN, + OP_TYPE_MINUS = 20, // bitwise operator - OP_TYPE_BIT_AND, + OP_TYPE_BIT_AND = 30, OP_TYPE_BIT_OR, // binary comparison operator - OP_TYPE_GREATER_THAN, + OP_TYPE_GREATER_THAN = 40, OP_TYPE_GREATER_EQUAL, OP_TYPE_LOWER_THAN, OP_TYPE_LOWER_EQUAL, @@ -153,7 +152,7 @@ typedef enum EOperatorType { OP_TYPE_MATCH, OP_TYPE_NMATCH, // unary comparison operator - OP_TYPE_IS_NULL, + OP_TYPE_IS_NULL = 100, OP_TYPE_IS_NOT_NULL, OP_TYPE_IS_TRUE, OP_TYPE_IS_FALSE, @@ -163,8 +162,11 @@ typedef enum EOperatorType { OP_TYPE_IS_NOT_UNKNOWN, // json operator - OP_TYPE_JSON_GET_VALUE, - OP_TYPE_JSON_CONTAINS + OP_TYPE_JSON_GET_VALUE = 150, + OP_TYPE_JSON_CONTAINS, + + // internal operator + OP_TYPE_ASSIGN = 200 } EOperatorType; #define OP_TYPE_CALC_MAX OP_TYPE_BIT_OR diff --git a/include/util/tqueue.h b/include/util/tqueue.h index 0f4f1db9eee4c1b57e25464cd947c1c96218fbec..da409a90bb96c2b19ad081c4599a9fa75de1ad4e 100644 --- a/include/util/tqueue.h +++ b/include/util/tqueue.h @@ -76,6 +76,7 @@ void taosFreeQall(STaosQall *qall); int32_t taosReadAllQitems(STaosQueue *queue, STaosQall *qall); int32_t taosGetQitem(STaosQall *qall, void **ppItem); void taosResetQitems(STaosQall *qall); +int32_t taosQallItemSize(STaosQall *qall); STaosQset *taosOpenQset(); void taosCloseQset(STaosQset *qset); diff --git a/packaging/tools/make_install.bat b/packaging/tools/make_install.bat index 3c27c1beca52139679424939f8236c5f0165d9a7..d4dde391c8c2ca034c8e31ba2f5a413a3050fc1d 100644 --- a/packaging/tools/make_install.bat +++ b/packaging/tools/make_install.bat @@ -28,6 +28,13 @@ if not exist %tagert_dir%\\driver ( if not exist C:\\TDengine\\cfg\\taos.cfg ( copy %source_dir%\\packaging\\cfg\\taos.cfg %tagert_dir%\\cfg\\taos.cfg > nul ) + +if exist %binary_dir%\\test\\cfg\\taosadapter.toml ( + if not exist %tagert_dir%\\cfg\\taosadapter.toml ( + copy %binary_dir%\\test\\cfg\\taosadapter.toml %tagert_dir%\\cfg\\taosadapter.toml > nul + ) +) + copy %source_dir%\\include\\client\\taos.h %tagert_dir%\\include > nul copy %source_dir%\\include\\util\\taoserror.h %tagert_dir%\\include > nul copy %source_dir%\\include\\libs\\function\\taosudf.h %tagert_dir%\\include > nul @@ -40,6 +47,9 @@ copy %binary_dir%\\build\\bin\\udfd.exe %tagert_dir% > nul if exist %binary_dir%\\build\\bin\\taosBenchmark.exe ( copy %binary_dir%\\build\\bin\\taosBenchmark.exe %tagert_dir% > nul ) +if exist %binary_dir%\\build\\bin\\taosadapter.exe ( + copy %binary_dir%\\build\\bin\\taosadapter.exe %tagert_dir% > nul +) mshta vbscript:createobject("shell.application").shellexecute("%~s0",":hasAdmin","","runas",1)(window.close)&& echo To start/stop TDengine with administrator privileges: sc start/stop taosd &goto :eof :hasAdmin diff --git a/source/client/test/clientTests.cpp b/source/client/test/clientTests.cpp index ec270889e2b326ab6481005ece006278c686da78..4ea5443678d8b927f8df831874f32b20655cb7d0 100644 --- a/source/client/test/clientTests.cpp +++ b/source/client/test/clientTests.cpp @@ -123,7 +123,7 @@ void createNewTable(TAOS* pConn, int32_t index) { } taos_free_result(pRes); - for(int32_t i = 0; i < 100000; i += 20) { + for(int32_t i = 0; i < 3280; i += 20) { char sql[1024] = {0}; sprintf(sql, "insert into tu%d values(now+%da, %d)(now+%da, %d)(now+%da, %d)(now+%da, %d)" @@ -679,30 +679,28 @@ TEST(testCase, projection_query_tables) { TAOS_RES* pRes = taos_query(pConn, "use abc1"); taos_free_result(pRes); - pRes = taos_query(pConn, "explain verbose true select _wstart,count(*),a from st1 partition by a interval(1s)"); - printResult(pRes); -// pRes = taos_query(pConn, "create stable st1 (ts timestamp, k int) tags(a int)"); -// if (taos_errno(pRes) != 0) { -// printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); -// } -// taos_free_result(pRes); -// -// pRes = taos_query(pConn, "create stable st2 (ts timestamp, k int) tags(a int)"); -// if (taos_errno(pRes) != 0) { -// printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); -// } -// taos_free_result(pRes); -// -// pRes = taos_query(pConn, "create table tu using st1 tags(1)"); -// if (taos_errno(pRes) != 0) { -// printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); -// } -// taos_free_result(pRes); -// -// for(int32_t i = 0; i < 1; ++i) { -// printf("create table :%d\n", i); -// createNewTable(pConn, i); -// } + pRes = taos_query(pConn, "create stable st1 (ts timestamp, k int) tags(a int)"); + if (taos_errno(pRes) != 0) { + printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); + } + taos_free_result(pRes); + + pRes = taos_query(pConn, "create stable st2 (ts timestamp, k int) tags(a int)"); + if (taos_errno(pRes) != 0) { + printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); + } + taos_free_result(pRes); + + pRes = taos_query(pConn, "create table tu using st1 tags(1)"); + if (taos_errno(pRes) != 0) { + printf("failed to create table tu, reason:%s\n", taos_errstr(pRes)); + } + taos_free_result(pRes); + + for(int32_t i = 0; i < 2; ++i) { + printf("create table :%d\n", i); + createNewTable(pConn, i); + } // // pRes = taos_query(pConn, "select * from tu"); // if (taos_errno(pRes) != 0) { diff --git a/source/common/src/systable.c b/source/common/src/systable.c index 6dddcc2f7422aaa09d4e8b7691cce0c2fc107b6d..65041e1f129d6d73f9a5a13678bd1dcd5efe733b 100644 --- a/source/common/src/systable.c +++ b/source/common/src/systable.c @@ -88,7 +88,7 @@ static const SSysDbTableSchema userDBSchema[] = { {.name = "comp", .bytes = 1, .type = TSDB_DATA_TYPE_TINYINT}, {.name = "precision", .bytes = 2 + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_VARCHAR}, {.name = "status", .bytes = 10 + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_VARCHAR}, - {.name = "retention", .bytes = 60 + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_VARCHAR}, + {.name = "retentions", .bytes = 60 + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_VARCHAR}, {.name = "single_stable", .bytes = 1, .type = TSDB_DATA_TYPE_BOOL}, {.name = "cachemodel", .bytes = TSDB_CACHE_MODEL_STR_LEN + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_VARCHAR}, {.name = "cachesize", .bytes = 4, .type = TSDB_DATA_TYPE_INT}, diff --git a/source/common/src/ttypes.c b/source/common/src/ttypes.c index 156b66ae865e8a1f4b1f7562961d9f45afb688c6..fee89e2f37e3afeec0959b1d78b9a73745573edc 100644 --- a/source/common/src/ttypes.c +++ b/source/common/src/ttypes.c @@ -392,10 +392,10 @@ tDataTypeDescriptor tDataTypes[TSDB_DATA_TYPE_MAX] = { getStatics_i64}, {TSDB_DATA_TYPE_FLOAT, 5, FLOAT_BYTES, "FLOAT", 0, 0, tsCompressFloat, tsDecompressFloat, getStatics_f}, {TSDB_DATA_TYPE_DOUBLE, 6, DOUBLE_BYTES, "DOUBLE", 0, 0, tsCompressDouble, tsDecompressDouble, getStatics_d}, - {TSDB_DATA_TYPE_VARCHAR, 6, 0, "VARCHAR", 0, 0, tsCompressString, tsDecompressString, getStatics_bin}, + {TSDB_DATA_TYPE_VARCHAR, 6, 1, "VARCHAR", 0, 0, tsCompressString, tsDecompressString, getStatics_bin}, {TSDB_DATA_TYPE_TIMESTAMP, 9, LONG_BYTES, "TIMESTAMP", INT64_MIN, INT64_MAX, tsCompressTimestamp, tsDecompressTimestamp, getStatics_i64}, - {TSDB_DATA_TYPE_NCHAR, 5, 8, "NCHAR", 0, 0, tsCompressString, tsDecompressString, getStatics_nchr}, + {TSDB_DATA_TYPE_NCHAR, 5, 1, "NCHAR", 0, 0, tsCompressString, tsDecompressString, getStatics_nchr}, {TSDB_DATA_TYPE_UTINYINT, 16, CHAR_BYTES, "TINYINT UNSIGNED", 0, UINT8_MAX, tsCompressTinyint, tsDecompressTinyint, getStatics_u8}, {TSDB_DATA_TYPE_USMALLINT, 17, SHORT_BYTES, "SMALLINT UNSIGNED", 0, UINT16_MAX, tsCompressSmallint, diff --git a/source/dnode/mnode/impl/src/mndStb.c b/source/dnode/mnode/impl/src/mndStb.c index e0f2b831608fb8342a3ad31bd4c4e62152871830..ebec3d5ea686c3a976adf5d4890f2a7eb7d8be82 100644 --- a/source/dnode/mnode/impl/src/mndStb.c +++ b/source/dnode/mnode/impl/src/mndStb.c @@ -442,6 +442,8 @@ static void *mndBuildVCreateStbReq(SMnode *pMnode, SVgObj *pVgroup, SStbObj *pSt if (req.rollup) { req.rsmaParam.maxdelay[0] = pStb->maxdelay[0]; req.rsmaParam.maxdelay[1] = pStb->maxdelay[1]; + req.rsmaParam.watermark[0] = pStb->watermark[0]; + req.rsmaParam.watermark[1] = pStb->watermark[1]; if (pStb->ast1Len > 0) { if (mndConvertRsmaTask(&req.rsmaParam.qmsg[0], &req.rsmaParam.qmsgLen[0], pStb->pAst1, pStb->uid, STREAM_TRIGGER_WINDOW_CLOSE, req.rsmaParam.watermark[0]) < 0) { diff --git a/source/dnode/vnode/src/inc/meta.h b/source/dnode/vnode/src/inc/meta.h index 2efc33a8ee1350d3c2ee05536eced70b9edc3463..adfbb919206a184664655cc11746c1d25c14147b 100644 --- a/source/dnode/vnode/src/inc/meta.h +++ b/source/dnode/vnode/src/inc/meta.h @@ -66,7 +66,6 @@ int32_t metaCacheOpen(SMeta* pMeta); void metaCacheClose(SMeta* pMeta); int32_t metaCacheUpsert(SMeta* pMeta, SMetaInfo* pInfo); int32_t metaCacheDrop(SMeta* pMeta, int64_t uid); -int32_t metaCacheGet(SMeta* pMeta, int64_t uid, SMetaInfo* pInfo); struct SMeta { TdThreadRwlock lock; diff --git a/source/dnode/vnode/src/inc/sma.h b/source/dnode/vnode/src/inc/sma.h index c43772062e106ce0766f80a8f4d1a2b4fbfe62d4..ca77042bb26d72f87471e4ac80329efc92449427 100644 --- a/source/dnode/vnode/src/inc/sma.h +++ b/source/dnode/vnode/src/inc/sma.h @@ -32,7 +32,8 @@ extern "C" { #define smaTrace(...) do { if (smaDebugFlag & DEBUG_TRACE) { taosPrintLog("SMA ", DEBUG_TRACE, tsdbDebugFlag, __VA_ARGS__); }} while(0) // clang-format on -#define RSMA_TASK_INFO_HASH_SLOT 8 +#define RSMA_TASK_INFO_HASH_SLOT (8) +#define RSMA_EXECUTOR_MAX (1) typedef struct SSmaEnv SSmaEnv; typedef struct SSmaStat SSmaStat; @@ -90,14 +91,14 @@ struct SRSmaStat { SSma *pSma; int64_t commitAppliedVer; // vnode applied version for async commit int64_t refId; // shared by fetch tasks - volatile int64_t qBufSize; // queue buffer size + volatile int64_t nBufItems; // number of items in queue buffer SRWLatch lock; // r/w lock for rsma fs(e.g. qtaskinfo) + volatile int8_t nExecutor; // [1, max(half of query threads, 4)] int8_t triggerStat; // shared by fetch tasks int8_t commitStat; // 0 not in committing, 1 in committing - int8_t execStat; // 0 not in exec , 1 in exec SArray *aTaskFile; // qTaskFiles committed recently(for recovery/snapshot r/w) SHashObj *infoHash; // key: suid, value: SRSmaInfo - SHashObj *fetchHash; // key: suid, value: L1 or L2 or L1|L2 + tsem_t notEmpty; // has items in queue buffer }; struct SSmaStat { @@ -111,26 +112,28 @@ struct SSmaStat { #define SMA_STAT_TSMA(s) (&(s)->tsmaStat) #define SMA_STAT_RSMA(s) (&(s)->rsmaStat) #define RSMA_INFO_HASH(r) ((r)->infoHash) -#define RSMA_FETCH_HASH(r) ((r)->fetchHash) #define RSMA_TRIGGER_STAT(r) (&(r)->triggerStat) #define RSMA_COMMIT_STAT(r) (&(r)->commitStat) #define RSMA_REF_ID(r) ((r)->refId) #define RSMA_FS_LOCK(r) (&(r)->lock) struct SRSmaInfoItem { - int8_t level; + int8_t level : 4; + int8_t fetchLevel : 4; int8_t triggerStat; - uint16_t interval; // second - int32_t maxDelay; + uint16_t nSkipped; + int32_t maxDelay; // ms tmr_h tmrId; }; struct SRSmaInfo { STSchema *pTSchema; int64_t suid; - int64_t refId; // refId of SRSmaStat - uint64_t delFlag : 1; - uint64_t lastReceived : 63; // second + int64_t refId; // refId of SRSmaStat + int64_t lastRecv; // ms + int8_t assigned; // 0 idle, 1 assgined for exec + int8_t delFlag; + int16_t padding; T_REF_DECLARE() SRSmaInfoItem items[TSDB_RETENTION_L2]; void *taskInfo[TSDB_RETENTION_L2]; // qTaskInfo_t diff --git a/source/dnode/vnode/src/inc/tsdb.h b/source/dnode/vnode/src/inc/tsdb.h index a30f308ecd925cb8fa02f4e67f77c310fa2b404b..d1f5cfb122d6fdfee2cb8f54911a07a25cbb078c 100644 --- a/source/dnode/vnode/src/inc/tsdb.h +++ b/source/dnode/vnode/src/inc/tsdb.h @@ -45,7 +45,7 @@ typedef struct SBlockIdx SBlockIdx; typedef struct SBlock SBlock; typedef struct SBlockL SBlockL; typedef struct SColData SColData; -typedef struct SBlockDataHdr SBlockDataHdr; +typedef struct SDiskDataHdr SDiskDataHdr; typedef struct SBlockData SBlockData; typedef struct SDelFile SDelFile; typedef struct SHeadFile SHeadFile; @@ -61,7 +61,11 @@ typedef struct SRowIter SRowIter; typedef struct STsdbFS STsdbFS; typedef struct SRowMerger SRowMerger; typedef struct STsdbReadSnap STsdbReadSnap; +typedef struct SBlockInfo SBlockInfo; +typedef struct SSmaInfo SSmaInfo; +typedef struct SBlockCol SBlockCol; +#define TSDB_FILE_DLMT ((uint32_t)0xF00AFA0F) #define TSDB_MAX_SUBBLOCKS 8 #define TSDB_FHDR_SIZE 512 @@ -113,10 +117,14 @@ int32_t tPutBlock(uint8_t *p, void *ph); int32_t tGetBlock(uint8_t *p, void *ph); int32_t tBlockCmprFn(const void *p1, const void *p2); bool tBlockHasSma(SBlock *pBlock); +// SBlockL +int32_t tPutBlockL(uint8_t *p, void *ph); +int32_t tGetBlockL(uint8_t *p, void *ph); // SBlockIdx int32_t tPutBlockIdx(uint8_t *p, void *ph); int32_t tGetBlockIdx(uint8_t *p, void *ph); int32_t tCmprBlockIdx(void const *lhs, void const *rhs); +int32_t tCmprBlockL(void const *lhs, void const *rhs); // SColdata void tColDataInit(SColData *pColData, int16_t cid, int8_t type, int8_t smaOn); void tColDataReset(SColData *pColData); @@ -131,20 +139,25 @@ int32_t tGetColData(uint8_t *p, SColData *pColData); #define tBlockDataLastRow(PBLOCKDATA) tsdbRowFromBlockData(PBLOCKDATA, (PBLOCKDATA)->nRow - 1) #define tBlockDataFirstKey(PBLOCKDATA) TSDBROW_KEY(&tBlockDataFirstRow(PBLOCKDATA)) #define tBlockDataLastKey(PBLOCKDATA) TSDBROW_KEY(&tBlockDataLastRow(PBLOCKDATA)) -int32_t tBlockDataInit(SBlockData *pBlockData); + +int32_t tBlockDataCreate(SBlockData *pBlockData); +void tBlockDataDestroy(SBlockData *pBlockData, int8_t deepClear); +int32_t tBlockDataInit(SBlockData *pBlockData, int64_t suid, int64_t uid, STSchema *pTSchema); +int32_t tBlockDataInitEx(SBlockData *pBlockData, SBlockData *pBlockDataFrom); void tBlockDataReset(SBlockData *pBlockData); -int32_t tBlockDataSetSchema(SBlockData *pBlockData, STSchema *pTSchema); -int32_t tBlockDataCorrectSchema(SBlockData *pBlockData, SBlockData *pBlockDataFrom); -void tBlockDataClearData(SBlockData *pBlockData); -void tBlockDataClear(SBlockData *pBlockData, int8_t deepClear); -int32_t tBlockDataAddColData(SBlockData *pBlockData, int32_t iColData, SColData **ppColData); -int32_t tBlockDataAppendRow(SBlockData *pBlockData, TSDBROW *pRow, STSchema *pTSchema); -int32_t tBlockDataMerge(SBlockData *pBlockData1, SBlockData *pBlockData2, SBlockData *pBlockData); -int32_t tBlockDataCopy(SBlockData *pBlockDataSrc, SBlockData *pBlockDataDest); +int32_t tBlockDataAppendRow(SBlockData *pBlockData, TSDBROW *pRow, STSchema *pTSchema, int64_t uid); +void tBlockDataClear(SBlockData *pBlockData); SColData *tBlockDataGetColDataByIdx(SBlockData *pBlockData, int32_t idx); void tBlockDataGetColData(SBlockData *pBlockData, int16_t cid, SColData **ppColData); -int32_t tPutBlockData(uint8_t *p, SBlockData *pBlockData); -int32_t tGetBlockData(uint8_t *p, SBlockData *pBlockData); +int32_t tBlockDataCopy(SBlockData *pBlockDataSrc, SBlockData *pBlockDataDest); +int32_t tBlockDataMerge(SBlockData *pBlockData1, SBlockData *pBlockData2, SBlockData *pBlockData); +int32_t tBlockDataAddColData(SBlockData *pBlockData, int32_t iColData, SColData **ppColData); +int32_t tCmprBlockData(SBlockData *pBlockData, int8_t cmprAlg, uint8_t **ppOut, int32_t *szOut, uint8_t *aBuf[], + int32_t aBufN[]); +int32_t tDecmprBlockData(uint8_t *pIn, int32_t szIn, SBlockData *pBlockData, uint8_t *aBuf[]); +// SDiskDataHdr +int32_t tPutDiskDataHdr(uint8_t *p, void *ph); +int32_t tGetDiskDataHdr(uint8_t *p, void *ph); // SDelIdx int32_t tPutDelIdx(uint8_t *p, void *ph); int32_t tGetDelIdx(uint8_t *p, void *ph); @@ -168,13 +181,25 @@ void tsdbFidKeyRange(int32_t fid, int32_t minutes, int8_t precision, TSKEY *m int32_t tsdbFidLevel(int32_t fid, STsdbKeepCfg *pKeepCfg, int64_t now); int32_t tsdbBuildDeleteSkyline(SArray *aDelData, int32_t sidx, int32_t eidx, SArray *aSkyline); void tsdbCalcColDataSMA(SColData *pColData, SColumnDataAgg *pColAgg); +int32_t tPutColumnDataAgg(uint8_t *p, SColumnDataAgg *pColAgg); +int32_t tGetColumnDataAgg(uint8_t *p, SColumnDataAgg *pColAgg); +int32_t tsdbCmprData(uint8_t *pIn, int32_t szIn, int8_t type, int8_t cmprAlg, uint8_t **ppOut, int32_t nOut, + int32_t *szOut, uint8_t **ppBuf); +int32_t tsdbDecmprData(uint8_t *pIn, int32_t szIn, int8_t type, int8_t cmprAlg, uint8_t **ppOut, int32_t szOut, + uint8_t **ppBuf); +int32_t tsdbCmprColData(SColData *pColData, int8_t cmprAlg, SBlockCol *pBlockCol, uint8_t **ppOut, int32_t nOut, + uint8_t **ppBuf); +int32_t tsdbDecmprColData(uint8_t *pIn, SBlockCol *pBlockCol, int8_t cmprAlg, int32_t nVal, SColData *pColData, + uint8_t **ppBuf); +int32_t tsdbReadAndCheck(TdFilePtr pFD, int64_t offset, uint8_t **ppOut, int32_t size, int8_t toCheck); // tsdbMemTable ============================================================================================== // SMemTable -int32_t tsdbMemTableCreate(STsdb *pTsdb, SMemTable **ppMemTable); -void tsdbMemTableDestroy(SMemTable *pMemTable); -void tsdbGetTbDataFromMemTable(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid, STbData **ppTbData); -void tsdbRefMemTable(SMemTable *pMemTable); -void tsdbUnrefMemTable(SMemTable *pMemTable); +int32_t tsdbMemTableCreate(STsdb *pTsdb, SMemTable **ppMemTable); +void tsdbMemTableDestroy(SMemTable *pMemTable); +STbData *tsdbGetTbDataFromMemTable(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid); +void tsdbRefMemTable(SMemTable *pMemTable); +void tsdbUnrefMemTable(SMemTable *pMemTable); +SArray *tsdbMemTableGetTbDataArray(SMemTable *pMemTable); // STbDataIter int32_t tsdbTbDataIterCreate(STbData *pTbData, TSDBKEY *pFrom, int8_t backward, STbDataIter **ppIter); void *tsdbTbDataIterDestroy(STbDataIter *pIter); @@ -223,33 +248,33 @@ int32_t tsdbFSUpsertDelFile(STsdbFS *pFS, SDelFile *pDelFile); int32_t tsdbDataFWriterOpen(SDataFWriter **ppWriter, STsdb *pTsdb, SDFileSet *pSet); int32_t tsdbDataFWriterClose(SDataFWriter **ppWriter, int8_t sync); int32_t tsdbUpdateDFileSetHeader(SDataFWriter *pWriter); -int32_t tsdbWriteBlockIdx(SDataFWriter *pWriter, SArray *aBlockIdx, uint8_t **ppBuf); -int32_t tsdbWriteBlock(SDataFWriter *pWriter, SMapData *pMapData, uint8_t **ppBuf, SBlockIdx *pBlockIdx); -int32_t tsdbWriteBlockData(SDataFWriter *pWriter, SBlockData *pBlockData, uint8_t **ppBuf1, uint8_t **ppBuf2, - SBlockIdx *pBlockIdx, SBlock *pBlock, int8_t cmprAlg); +int32_t tsdbWriteBlockIdx(SDataFWriter *pWriter, SArray *aBlockIdx); +int32_t tsdbWriteBlock(SDataFWriter *pWriter, SMapData *pMapData, SBlockIdx *pBlockIdx); +int32_t tsdbWriteBlockL(SDataFWriter *pWriter, SArray *aBlockL); +int32_t tsdbWriteBlockData(SDataFWriter *pWriter, SBlockData *pBlockData, SBlockInfo *pBlkInfo, SSmaInfo *pSmaInfo, + int8_t cmprAlg, int8_t toLast); int32_t tsdbDFileSetCopy(STsdb *pTsdb, SDFileSet *pSetFrom, SDFileSet *pSetTo); // SDataFReader int32_t tsdbDataFReaderOpen(SDataFReader **ppReader, STsdb *pTsdb, SDFileSet *pSet); int32_t tsdbDataFReaderClose(SDataFReader **ppReader); -int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx, uint8_t **ppBuf); -int32_t tsdbReadBlock(SDataFReader *pReader, SBlockIdx *pBlockIdx, SMapData *pMapData, uint8_t **ppBuf); -int32_t tsdbReadColData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, int16_t *aColId, int32_t nCol, - SBlockData *pBlockData, uint8_t **ppBuf1, uint8_t **ppBuf2); -int32_t tsdbReadBlockData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, SBlockData *pBlockData, - uint8_t **ppBuf1, uint8_t **ppBuf2); -int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnDataAgg, uint8_t **ppBuf); +int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx); +int32_t tsdbReadBlock(SDataFReader *pReader, SBlockIdx *pBlockIdx, SMapData *pMapData); +int32_t tsdbReadBlockL(SDataFReader *pReader, SArray *aBlockL); +int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnDataAgg); +int32_t tsdbReadDataBlock(SDataFReader *pReader, SBlock *pBlock, SBlockData *pBlockData); +int32_t tsdbReadLastBlock(SDataFReader *pReader, SBlockL *pBlockL, SBlockData *pBlockData); // SDelFWriter int32_t tsdbDelFWriterOpen(SDelFWriter **ppWriter, SDelFile *pFile, STsdb *pTsdb); int32_t tsdbDelFWriterClose(SDelFWriter **ppWriter, int8_t sync); -int32_t tsdbWriteDelData(SDelFWriter *pWriter, SArray *aDelData, uint8_t **ppBuf, SDelIdx *pDelIdx); -int32_t tsdbWriteDelIdx(SDelFWriter *pWriter, SArray *aDelIdx, uint8_t **ppBuf); +int32_t tsdbWriteDelData(SDelFWriter *pWriter, SArray *aDelData, SDelIdx *pDelIdx); +int32_t tsdbWriteDelIdx(SDelFWriter *pWriter, SArray *aDelIdx); int32_t tsdbUpdateDelFileHdr(SDelFWriter *pWriter); // SDelFReader -int32_t tsdbDelFReaderOpen(SDelFReader **ppReader, SDelFile *pFile, STsdb *pTsdb, uint8_t **ppBuf); +int32_t tsdbDelFReaderOpen(SDelFReader **ppReader, SDelFile *pFile, STsdb *pTsdb); int32_t tsdbDelFReaderClose(SDelFReader **ppReader); -int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData, uint8_t **ppBuf); -int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx, uint8_t **ppBuf); +int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData); +int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx); // tsdbRead.c ============================================================================================== int32_t tsdbTakeReadSnap(STsdb *pTsdb, STsdbReadSnap **ppSnap); void tsdbUntakeReadSnap(STsdb *pTsdb, STsdbReadSnap *pSnap); @@ -277,13 +302,6 @@ size_t tsdbCacheGetCapacity(SVnode *pVnode); int32_t tsdbCacheLastArray2Row(SArray *pLastArray, STSRow **ppRow, STSchema *pSchema); // structs ======================= -typedef struct { - int minFid; - int midFid; - int maxFid; - TSKEY minKey; -} SRtn; - struct STsdbFS { SDelFile *pDelFile; SArray *aDFileSet; // SArray @@ -312,30 +330,23 @@ struct SMemSkipListNode { SMemSkipListNode *forwards[0]; }; typedef struct SMemSkipList { - uint32_t seed; int64_t size; + uint32_t seed; int8_t maxLevel; int8_t level; SMemSkipListNode *pHead; SMemSkipListNode *pTail; } SMemSkipList; -struct SDelDataInfo { - tb_uid_t suid; - tb_uid_t uid; -}; - struct STbData { tb_uid_t suid; tb_uid_t uid; TSKEY minKey; TSKEY maxKey; - int64_t minVersion; - int64_t maxVersion; - int32_t maxSkmVer; SDelData *pHead; SDelData *pTail; SMemSkipList sl; + STbData *next; }; struct SMemTable { @@ -345,11 +356,13 @@ struct SMemTable { volatile int32_t nRef; TSKEY minKey; TSKEY maxKey; - int64_t minVersion; - int64_t maxVersion; int64_t nRow; int64_t nDel; - SArray *aTbData; // SArray + struct { + int32_t nTbData; + int32_t nBucket; + STbData **aBucket; + }; }; struct TSDBROW { @@ -380,63 +393,51 @@ struct SMapData { uint8_t *pData; }; -typedef struct { +struct SBlockCol { int16_t cid; int8_t type; int8_t smaOn; - int8_t flag; // HAS_NONE|HAS_NULL|HAS_VALUE - int32_t offset; - int32_t szBitmap; // bitmap size - int32_t szOffset; // size of offset, only for variant-length data type - int32_t szValue; // compressed column value size + int8_t flag; // HAS_NONE|HAS_NULL|HAS_VALUE int32_t szOrigin; // original column value size (only save for variant data type) -} SBlockCol; - -typedef struct { - int32_t nRow; - int8_t cmprAlg; - int64_t offset; // block data offset - int32_t szBlockCol; // SBlockCol size - int32_t szVersion; // VERSION size - int32_t szTSKEY; // TSKEY size - int32_t szBlock; // total block size - int64_t sOffset; // sma offset - int32_t nSma; // sma size -} SSubBlock; + int32_t szBitmap; // bitmap size, 0 only for flag == HAS_VAL + int32_t szOffset; // offset size, 0 only for non-variant-length type + int32_t szValue; // value size, 0 when flag == (HAS_NULL | HAS_NONE) + int32_t offset; +}; + +struct SBlockInfo { + int64_t offset; // block data offset + int32_t szBlock; + int32_t szKey; +}; + +struct SSmaInfo { + int64_t offset; + int32_t size; +}; struct SBlock { - TSDBKEY minKey; - TSDBKEY maxKey; - int64_t minVersion; - int64_t maxVersion; - int32_t nRow; - int8_t last; - int8_t hasDup; - int8_t nSubBlock; - SSubBlock aSubBlock[TSDB_MAX_SUBBLOCKS]; + TSDBKEY minKey; + TSDBKEY maxKey; + int64_t minVer; + int64_t maxVer; + int32_t nRow; + int8_t hasDup; + int8_t nSubBlock; + SBlockInfo aSubBlock[TSDB_MAX_SUBBLOCKS]; + SSmaInfo smaInfo; }; struct SBlockL { - struct { - int64_t uid; - int64_t version; - TSKEY ts; - } minKey; - struct { - int64_t uid; - int64_t version; - TSKEY ts; - } maxKey; - int64_t minVer; - int64_t maxVer; - int32_t nRow; - int8_t cmprAlg; - int64_t offset; - int32_t szBlock; - int32_t szBlockCol; - int32_t szUid; - int32_t szVer; - int32_t szTSKEY; + int64_t suid; + int64_t minUid; + int64_t maxUid; + TSKEY minKey; + TSKEY maxKey; + int64_t minVer; + int64_t maxVer; + int32_t nRow; + SBlockInfo bInfo; }; struct SColData { @@ -451,10 +452,17 @@ struct SColData { uint8_t *pData; }; +// (SBlockData){.suid = 0, .uid = 0}: block data not initialized +// (SBlockData){.suid = suid, .uid = uid}: block data for ONE child table int .data file +// (SBlockData){.suid = suid, .uid = 0}: block data for N child tables int .last file +// (SBlockData){.suid = 0, .uid = uid}: block data for 1 normal table int .last/.data file struct SBlockData { - int32_t nRow; - int64_t *aVersion; - TSKEY *aTSKEY; + int64_t suid; // 0 means normal table block data, otherwise child table block data + int64_t uid; // 0 means block data in .last file, otherwise in .data file + int32_t nRow; // number of rows + int64_t *aUid; // uids of each row, only exist in block data in .last file (uid == 0) + int64_t *aVersion; // versions of each row + TSKEY *aTSKEY; // timestamp of each row SArray *aIdx; // SArray SArray *aColData; // SArray }; @@ -492,13 +500,18 @@ struct SDelIdx { int64_t size; }; -#pragma pack(push, 1) -struct SBlockDataHdr { +struct SDiskDataHdr { uint32_t delimiter; + uint32_t fmtVer; int64_t suid; int64_t uid; + int32_t szUid; + int32_t szVer; + int32_t szKey; + int32_t szBlkCol; + int32_t nRow; + int8_t cmprAlg; }; -#pragma pack(pop) struct SDelFile { volatile int32_t nRef; @@ -528,6 +541,7 @@ struct SLastFile { int64_t commitID; int64_t size; + int64_t offset; }; struct SSmaFile { @@ -562,6 +576,8 @@ struct SDelFWriter { STsdb *pTsdb; SDelFile fDel; TdFilePtr pWriteH; + + uint8_t *aBuf[1]; }; struct SDataFWriter { @@ -577,6 +593,8 @@ struct SDataFWriter { SDataFile fData; SLastFile fLast; SSmaFile fSma; + + uint8_t *aBuf[4]; }; struct STsdbReadSnap { diff --git a/source/dnode/vnode/src/inc/vnodeInt.h b/source/dnode/vnode/src/inc/vnodeInt.h index 7ac1cc4f0e5cff769307d73749e74df3f8e84e42..39c5f3873ed9884109c0dc28f66d314b12b83a99 100644 --- a/source/dnode/vnode/src/inc/vnodeInt.h +++ b/source/dnode/vnode/src/inc/vnodeInt.h @@ -189,6 +189,7 @@ SSubmitReq* tqBlockToSubmit(SVnode* pVnode, const SArray* pBlocks, const STSchem int32_t smaInit(); void smaCleanUp(); int32_t smaOpen(SVnode* pVnode); +int32_t smaPreClose(SSma* pSma); int32_t smaClose(SSma* pSma); int32_t smaBegin(SSma* pSma); int32_t smaSyncPreCommit(SSma* pSma); @@ -198,7 +199,6 @@ int32_t smaAsyncPreCommit(SSma* pSma); int32_t smaAsyncCommit(SSma* pSma); int32_t smaAsyncPostCommit(SSma* pSma); int32_t smaDoRetention(SSma* pSma, int64_t now); -int32_t smaProcessFetch(SSma* pSma, void* pMsg); int32_t smaProcessExec(SSma* pSma, void* pMsg); int32_t tdProcessTSmaCreate(SSma* pSma, int64_t version, const char* msg); @@ -323,6 +323,7 @@ struct SVnode { TdThreadMutex lock; bool blocked; bool restored; + bool inClose; tsem_t syncSem; SQHandle* pQuery; }; @@ -369,6 +370,7 @@ struct SSma { void smaHandleRes(void* pVnode, int64_t smaId, const SArray* data); enum { + SNAP_DATA_CFG = 0, SNAP_DATA_META = 1, SNAP_DATA_TSDB = 2, SNAP_DATA_DEL = 3, diff --git a/source/dnode/vnode/src/meta/metaQuery.c b/source/dnode/vnode/src/meta/metaQuery.c index 7cf365d3727e9ef0b51002398f986e8003da8a5d..805bc24d8c2824cb8e5e95df03c8b4b65ce25d6d 100644 --- a/source/dnode/vnode/src/meta/metaQuery.c +++ b/source/dnode/vnode/src/meta/metaQuery.c @@ -127,7 +127,7 @@ _err: // return 0; // } -bool metaIsTableExist(SMeta *pMeta, tb_uid_t uid) { +bool metaIsTableExist(SMeta *pMeta, tb_uid_t uid) { // query uid.idx if (tdbTbGet(pMeta->pUidIdx, &uid, sizeof(uid), NULL, NULL) < 0) { return false; @@ -512,18 +512,65 @@ STSchema *metaGetTbTSchema(SMeta *pMeta, tb_uid_t uid, int32_t sver) { } int32_t metaGetTbTSchemaEx(SMeta *pMeta, tb_uid_t suid, tb_uid_t uid, int32_t sver, STSchema **ppTSchema) { - int32_t code = 0; - STSchema *pTSchema = NULL; - SSkmDbKey skmDbKey = {.uid = suid ? suid : uid, .sver = sver}; + int32_t code = 0; + void *pData = NULL; int nData = 0; + SSkmDbKey skmDbKey; + if (sver <= 0) { + SMetaInfo info; + if (metaGetInfo(pMeta, suid ? suid : uid, &info) == 0) { + sver = info.skmVer; + } else { + TBC *pSkmDbC = NULL; + int c; + + skmDbKey.uid = suid ? suid : uid; + skmDbKey.sver = INT32_MAX; + + tdbTbcOpen(pMeta->pSkmDb, &pSkmDbC, NULL); + metaRLock(pMeta); + + if (tdbTbcMoveTo(pSkmDbC, &skmDbKey, sizeof(skmDbKey), &c) < 0) { + metaULock(pMeta); + tdbTbcClose(pSkmDbC); + code = TSDB_CODE_NOT_FOUND; + goto _exit; + } + + ASSERT(c); + + if (c < 0) { + tdbTbcMoveToPrev(pSkmDbC); + } + + const void *pKey = NULL; + int32_t nKey = 0; + tdbTbcGet(pSkmDbC, &pKey, &nKey, NULL, NULL); + + if (((SSkmDbKey *)pKey)->uid != skmDbKey.uid) { + metaULock(pMeta); + tdbTbcClose(pSkmDbC); + code = TSDB_CODE_NOT_FOUND; + goto _exit; + } + + sver = ((SSkmDbKey *)pKey)->sver; - // query + metaULock(pMeta); + tdbTbcClose(pSkmDbC); + } + } + + ASSERT(sver > 0); + + skmDbKey.uid = suid ? suid : uid; + skmDbKey.sver = sver; metaRLock(pMeta); - if (tdbTbGet(pMeta->pSkmDb, &skmDbKey, sizeof(skmDbKey), &pData, &nData) < 0) { - code = TSDB_CODE_NOT_FOUND; + if (tdbTbGet(pMeta->pSkmDb, &skmDbKey, sizeof(SSkmDbKey), &pData, &nData) < 0) { metaULock(pMeta); - goto _err; + code = TSDB_CODE_NOT_FOUND; + goto _exit; } metaULock(pMeta); @@ -545,15 +592,13 @@ int32_t metaGetTbTSchemaEx(SMeta *pMeta, tb_uid_t suid, tb_uid_t uid, int32_t sv SSchema *pSchema = pSchemaWrapper->pSchema + i; tdAddColToSchema(&sb, pSchema->type, pSchema->flags, pSchema->colId, pSchema->bytes); } - pTSchema = tdGetSchemaFromBuilder(&sb); + STSchema *pTSchema = tdGetSchemaFromBuilder(&sb); tdDestroyTSchemaBuilder(&sb); *ppTSchema = pTSchema; taosMemoryFree(pSchemaWrapper->pSchema); - return code; -_err: - *ppTSchema = NULL; +_exit: return code; } @@ -1006,6 +1051,8 @@ int32_t metaGetTableTags(SMeta *pMeta, uint64_t suid, SArray *uidList, SHashObj return TSDB_CODE_SUCCESS; } +int32_t metaCacheGet(SMeta *pMeta, int64_t uid, SMetaInfo *pInfo); + int32_t metaGetInfo(SMeta *pMeta, int64_t uid, SMetaInfo *pInfo) { int32_t code = 0; void *pData = NULL; diff --git a/source/dnode/vnode/src/meta/metaTable.c b/source/dnode/vnode/src/meta/metaTable.c index c1e59adbb0724f6339fa3136fa2cc28892fefcb5..aa107ab2532b83b40abe8b1abdc60e059ab1de34 100644 --- a/source/dnode/vnode/src/meta/metaTable.c +++ b/source/dnode/vnode/src/meta/metaTable.c @@ -357,10 +357,7 @@ int metaAlterSTable(SMeta *pMeta, int64_t version, SVCreateStbReq *pReq) { metaSaveToTbDb(pMeta, &nStbEntry); // update uid index - SMetaInfo info; - metaGetEntryInfo(&nStbEntry, &info); - tdbTbcUpsert(pUidIdxc, &pReq->suid, sizeof(tb_uid_t), - &(SUidIdxVal){.suid = info.suid, .version = info.version, .skmVer = info.skmVer}, sizeof(SUidIdxVal), 0); + metaUpdateUidIdx(pMeta, &nStbEntry); if (oStbEntry.pBuf) taosMemoryFree(oStbEntry.pBuf); metaULock(pMeta); @@ -884,7 +881,8 @@ static int metaUpdateTableTagVal(SMeta *pMeta, int64_t version, SVAlterTbReq *pA } SCtbIdxKey ctbIdxKey = {.suid = ctbEntry.ctbEntry.suid, .uid = uid}; - tdbTbUpsert(pMeta->pCtbIdx, &ctbIdxKey, sizeof(ctbIdxKey), ctbEntry.ctbEntry.pTags, ((STag*)(ctbEntry.ctbEntry.pTags))->len, &pMeta->txn); + tdbTbUpsert(pMeta->pCtbIdx, &ctbIdxKey, sizeof(ctbIdxKey), ctbEntry.ctbEntry.pTags, + ((STag *)(ctbEntry.ctbEntry.pTags))->len, &pMeta->txn); tDecoderClear(&dc1); tDecoderClear(&dc2); @@ -1091,7 +1089,8 @@ static int metaUpdateTtlIdx(SMeta *pMeta, const SMetaEntry *pME) { static int metaUpdateCtbIdx(SMeta *pMeta, const SMetaEntry *pME) { SCtbIdxKey ctbIdxKey = {.suid = pME->ctbEntry.suid, .uid = pME->uid}; - return tdbTbInsert(pMeta->pCtbIdx, &ctbIdxKey, sizeof(ctbIdxKey), pME->ctbEntry.pTags, ((STag*)(pME->ctbEntry.pTags))->len, &pMeta->txn); + return tdbTbInsert(pMeta->pCtbIdx, &ctbIdxKey, sizeof(ctbIdxKey), pME->ctbEntry.pTags, + ((STag *)(pME->ctbEntry.pTags))->len, &pMeta->txn); } int metaCreateTagIdxKey(tb_uid_t suid, int32_t cid, const void *pTagData, int32_t nTagData, int8_t type, tb_uid_t uid, diff --git a/source/dnode/vnode/src/sma/smaCommit.c b/source/dnode/vnode/src/sma/smaCommit.c index 8b924750358bde41c036118a8d030ee0e8654acf..ca5367f39714ed1f3a979068b0a9a7204d385f8c 100644 --- a/source/dnode/vnode/src/sma/smaCommit.c +++ b/source/dnode/vnode/src/sma/smaCommit.c @@ -109,7 +109,7 @@ int32_t smaBegin(SSma *pSma) { /** * @brief pre-commit for rollup sma(sync commit). * 1) set trigger stat of rsma timer TASK_TRIGGER_STAT_PAUSED. - * 2) wait all triggered fetch tasks finished + * 2) wait for all triggered fetch tasks to finish * 3) perform persist task for qTaskInfo * * @param pSma @@ -127,14 +127,14 @@ static int32_t tdProcessRSmaSyncPreCommitImpl(SSma *pSma) { // step 1: set rsma stat paused atomic_store_8(RSMA_TRIGGER_STAT(pRSmaStat), TASK_TRIGGER_STAT_PAUSED); - // step 2: wait all triggered fetch tasks finished + // step 2: wait for all triggered fetch tasks to finish int32_t nLoops = 0; while (1) { if (T_REF_VAL_GET(pStat) == 0) { - smaDebug("vgId:%d, rsma fetch tasks all finished", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma fetch tasks are all finished", SMA_VID(pSma)); break; } else { - smaDebug("vgId:%d, rsma fetch tasks not all finished yet", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma fetch tasks are not all finished yet", SMA_VID(pSma)); } ++nLoops; if (nLoops > 1000) { @@ -316,15 +316,17 @@ static int32_t tdProcessRSmaAsyncPreCommitImpl(SSma *pSma) { // step 1: set rsma stat atomic_store_8(RSMA_TRIGGER_STAT(pRSmaStat), TASK_TRIGGER_STAT_PAUSED); atomic_store_8(RSMA_COMMIT_STAT(pRSmaStat), 1); + pRSmaStat->commitAppliedVer = pSma->pVnode->state.applied; + ASSERT(pRSmaStat->commitAppliedVer > 0); - // step 2: wait all triggered fetch tasks finished + // step 2: wait for all triggered fetch tasks to finish int32_t nLoops = 0; while (1) { if (T_REF_VAL_GET(pStat) == 0) { - smaDebug("vgId:%d, rsma fetch tasks all finished", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma commit, fetch tasks are all finished", SMA_VID(pSma)); break; } else { - smaDebug("vgId:%d, rsma fetch tasks not all finished yet", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma commit, fetch tasks are not all finished yet", SMA_VID(pSma)); } ++nLoops; if (nLoops > 1000) { @@ -338,30 +340,29 @@ static int32_t tdProcessRSmaAsyncPreCommitImpl(SSma *pSma) { * 1) This is high cost task and should not put in asyncPreCommit originally. * 2) But, if put in asyncCommit, would trigger taskInfo cloning frequently. */ - nLoops = 0; - smaInfo("vgId:%d, start to wait for rsma qtask free, TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); + if (tdRSmaProcessExecImpl(pSma, RSMA_EXEC_COMMIT) < 0) { + return TSDB_CODE_FAILED; + } - int8_t old; - while (1) { - old = atomic_val_compare_exchange_8(&pRSmaStat->execStat, 0, 1); - if (old == 0) break; - if (++nLoops > 1000) { + smaInfo("vgId:%d, rsma commit, wait for all items to be consumed, TID:%p", SMA_VID(pSma), (void*)taosGetSelfPthreadId()); + nLoops = 0; + while (atomic_load_64(&pRSmaStat->nBufItems) > 0) { + ++nLoops; + if (nLoops > 1000) { sched_yield(); nLoops = 0; - smaDebug("vgId:%d, wait for rsma qtask free, TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); } } - - smaInfo("vgId:%d, end to wait for rsma qtask free, TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); - - if (tdRSmaProcessExecImpl(pSma, RSMA_EXEC_COMMIT) < 0) { - atomic_store_8(&pRSmaStat->execStat, 0); + smaInfo("vgId:%d, rsma commit, all items are consumed, TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); + if (tdRSmaPersistExecImpl(pRSmaStat, RSMA_INFO_HASH(pRSmaStat)) < 0) { return TSDB_CODE_FAILED; } + smaInfo("vgId:%d, rsma commit, operator state commited, TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); +#if 0 // consuming task of qTaskInfo clone // step 4: swap queue/qall and iQueue/iQall // lock - taosWLockLatch(SMA_ENV_LOCK(pEnv)); + // taosWLockLatch(SMA_ENV_LOCK(pEnv)); ASSERT(RSMA_INFO_HASH(pRSmaStat)); @@ -376,13 +377,9 @@ static int32_t tdProcessRSmaAsyncPreCommitImpl(SSma *pSma) { pIter = taosHashIterate(RSMA_INFO_HASH(pRSmaStat), pIter); } - atomic_store_64(&pRSmaStat->qBufSize, 0); - atomic_store_8(&pRSmaStat->execStat, 0); // unlock - taosWUnLockLatch(SMA_ENV_LOCK(pEnv)); - - // step 5: others - pRSmaStat->commitAppliedVer = pSma->pVnode->state.applied; + // taosWUnLockLatch(SMA_ENV_LOCK(pEnv)); +#endif return TSDB_CODE_SUCCESS; } @@ -398,13 +395,14 @@ static int32_t tdProcessRSmaAsyncCommitImpl(SSma *pSma) { if (!pSmaEnv) { return TSDB_CODE_SUCCESS; } - +#if 0 SRSmaStat *pRSmaStat = (SRSmaStat *)SMA_ENV_STAT(pSmaEnv); // perform persist task for qTaskInfo operator if (tdRSmaPersistExecImpl(pRSmaStat, RSMA_INFO_HASH(pRSmaStat)) < 0) { return TSDB_CODE_FAILED; } +#endif return TSDB_CODE_SUCCESS; } @@ -426,10 +424,10 @@ static int32_t tdProcessRSmaAsyncPostCommitImpl(SSma *pSma) { // step 1: merge qTaskInfo and iQTaskInfo // lock - taosWLockLatch(SMA_ENV_LOCK(pEnv)); + // taosWLockLatch(SMA_ENV_LOCK(pEnv)); - void *pIter = taosHashIterate(RSMA_INFO_HASH(pRSmaStat), NULL); - while (pIter) { + void *pIter = NULL; + while ((pIter = taosHashIterate(RSMA_INFO_HASH(pRSmaStat), pIter))) { tb_uid_t *pSuid = (tb_uid_t *)taosHashGetKey(pIter, NULL); SRSmaInfo *pRSmaInfo = *(SRSmaInfo **)pIter; if (RSMA_INFO_IS_DEL(pRSmaInfo)) { @@ -447,14 +445,13 @@ static int32_t tdProcessRSmaAsyncPostCommitImpl(SSma *pSma) { SMA_VID(pSma), refVal, *pSuid); } - pIter = taosHashIterate(RSMA_INFO_HASH(pRSmaStat), pIter); continue; } - +#if 0 if (pRSmaInfo->taskInfo[0]) { if (pRSmaInfo->iTaskInfo[0]) { SRSmaInfo *pRSmaInfo = *(SRSmaInfo **)pRSmaInfo->iTaskInfo[0]; - tdFreeRSmaInfo(pSma, pRSmaInfo, true); + tdFreeRSmaInfo(pSma, pRSmaInfo, false); pRSmaInfo->iTaskInfo[0] = NULL; } } else { @@ -463,8 +460,7 @@ static int32_t tdProcessRSmaAsyncPostCommitImpl(SSma *pSma) { taosHashPut(RSMA_INFO_HASH(pRSmaStat), pSuid, sizeof(tb_uid_t), pIter, sizeof(pIter)); smaDebug("vgId:%d, rsma async post commit, migrated from iRsmaInfoHash for table:%" PRIi64, SMA_VID(pSma), *pSuid); - - pIter = taosHashIterate(RSMA_INFO_HASH(pRSmaStat), pIter); +#endif } for (int32_t i = 0; i < taosArrayGetSize(rsmaDeleted); ++i) { @@ -480,10 +476,9 @@ static int32_t tdProcessRSmaAsyncPostCommitImpl(SSma *pSma) { taosHashRemove(RSMA_INFO_HASH(pRSmaStat), pSuid, sizeof(tb_uid_t)); } taosArrayDestroy(rsmaDeleted); - // TODO: remove suid in files? // unlock - taosWUnLockLatch(SMA_ENV_LOCK(pEnv)); + // taosWUnLockLatch(SMA_ENV_LOCK(pEnv)); // step 2: cleanup outdated qtaskinfo files tdCleanupQTaskInfoFiles(pSma, pRSmaStat); diff --git a/source/dnode/vnode/src/sma/smaEnv.c b/source/dnode/vnode/src/sma/smaEnv.c index f51aad22bd86a0e55eb013232b4f5189291b04ed..e3b83f9955faf7a8000d18974cb6ec3639948c47 100644 --- a/source/dnode/vnode/src/sma/smaEnv.c +++ b/source/dnode/vnode/src/sma/smaEnv.c @@ -209,6 +209,7 @@ static int32_t tdInitSmaStat(SSmaStat **pSmaStat, int8_t smaType, const SSma *pS SRSmaStat *pRSmaStat = (SRSmaStat *)(*pSmaStat); pRSmaStat->pSma = (SSma *)pSma; atomic_store_8(RSMA_TRIGGER_STAT(pRSmaStat), TASK_TRIGGER_STAT_INIT); + tsem_init(&pRSmaStat->notEmpty, 0, 0); // init smaMgmt smaInit(); @@ -230,12 +231,6 @@ static int32_t tdInitSmaStat(SSmaStat **pSmaStat, int8_t smaType, const SSma *pS if (!RSMA_INFO_HASH(pRSmaStat)) { return TSDB_CODE_FAILED; } - - RSMA_FETCH_HASH(pRSmaStat) = taosHashInit( - RSMA_TASK_INFO_HASH_SLOT, taosGetDefaultHashFunction(TSDB_DATA_TYPE_BIGINT), true, HASH_ENTRY_LOCK); - if (!RSMA_FETCH_HASH(pRSmaStat)) { - return TSDB_CODE_FAILED; - } } else if (smaType == TSDB_SMA_TYPE_TIME_RANGE) { // TODO } else { @@ -267,6 +262,7 @@ static void tdDestroyRSmaStat(void *pRSmaStat) { smaDebug("vgId:%d, destroy rsma stat %p", SMA_VID(pSma), pRSmaStat); // step 1: set rsma trigger stat cancelled atomic_store_8(RSMA_TRIGGER_STAT(pStat), TASK_TRIGGER_STAT_CANCELLED); + tsem_destroy(&(pStat->notEmpty)); // step 2: destroy the rsma info and associated fetch tasks if (taosHashGetSize(RSMA_INFO_HASH(pStat)) > 0) { @@ -279,17 +275,14 @@ static void tdDestroyRSmaStat(void *pRSmaStat) { } taosHashCleanup(RSMA_INFO_HASH(pStat)); - // step 3: destroy the rsma fetch hash - taosHashCleanup(RSMA_FETCH_HASH(pStat)); - - // step 4: wait all triggered fetch tasks finished + // step 3: wait for all triggered fetch tasks to finish int32_t nLoops = 0; while (1) { if (T_REF_VAL_GET((SSmaStat *)pStat) == 0) { - smaDebug("vgId:%d, rsma fetch tasks all finished", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma fetch tasks are all finished", SMA_VID(pSma)); break; } else { - smaDebug("vgId:%d, rsma fetch tasks not all finished yet", SMA_VID(pSma)); + smaDebug("vgId:%d, rsma fetch tasks are not all finished yet", SMA_VID(pSma)); } ++nLoops; if (nLoops > 1000) { diff --git a/source/dnode/vnode/src/sma/smaOpen.c b/source/dnode/vnode/src/sma/smaOpen.c index 235fb1f94161256721dcb1f87ad3a2cc6a3e98f8..e2710b26e3d3672aac1b8053b019aa0addd37920 100644 --- a/source/dnode/vnode/src/sma/smaOpen.c +++ b/source/dnode/vnode/src/sma/smaOpen.c @@ -146,6 +146,20 @@ int32_t smaClose(SSma *pSma) { return 0; } +int32_t smaPreClose(SSma *pSma) { + if (pSma && VND_IS_RSMA(pSma->pVnode)) { + SSmaEnv *pEnv = NULL; + SRSmaStat *pStat = NULL; + if (!(pEnv = SMA_RSMA_ENV(pSma)) || !(pStat = (SRSmaStat *)SMA_ENV_STAT(pEnv))) { + return 0; + } + for (int32_t i = 0; i < RSMA_EXECUTOR_MAX; ++i) { + tsem_post(&(pStat->notEmpty)); + } + } + return 0; +} + /** * @brief rsma env restore * diff --git a/source/dnode/vnode/src/sma/smaRollup.c b/source/dnode/vnode/src/sma/smaRollup.c index 9b3b0cb63d9194450d7fc898563493ebe1356216..448b8ab50862cea44390b7b0c8cbc4d27d96c20c 100644 --- a/source/dnode/vnode/src/sma/smaRollup.c +++ b/source/dnode/vnode/src/sma/smaRollup.c @@ -15,10 +15,12 @@ #include "sma.h" -#define RSMA_QTASKINFO_BUFSIZE (32768) -#define RSMA_QTASKINFO_HEAD_LEN (sizeof(int32_t) + sizeof(int8_t) + sizeof(int64_t)) // len + type + suid -#define RSMA_QTASKEXEC_BUFSIZE (1048576) -#define RSMA_SUBMIT_BATCH_SIZE (1024) +#define RSMA_QTASKINFO_BUFSIZE (32768) // size +#define RSMA_QTASKINFO_HEAD_LEN (sizeof(int32_t) + sizeof(int8_t) + sizeof(int64_t)) // len + type + suid +#define RSMA_QTASKEXEC_SMOOTH_SIZE (100) // cnt +#define RSMA_SUBMIT_BATCH_SIZE (1024) // cnt +#define RSMA_FETCH_DELAY_MAX (900000) // ms +#define RSMA_FETCH_ACTIVE_MAX (1800) // ms SSmaMgmt smaMgmt = { .inited = 0, @@ -40,11 +42,10 @@ static int32_t tdExecuteRSmaImpl(SSma *pSma, const void *pMsg, int32_t msgSiz static SRSmaInfo *tdAcquireRSmaInfoBySuid(SSma *pSma, int64_t suid); static void tdReleaseRSmaInfo(SSma *pSma, SRSmaInfo *pInfo); static void tdFreeRSmaSubmitItems(SArray *pItems); -static int32_t tdRSmaConsumeAndFetch(SSma *pSma, int64_t suid, int8_t level, SArray *pSubmitArr); -static int32_t tdRSmaFetchAndSubmitResult(SSma *pSma, qTaskInfo_t taskInfo, SRSmaInfoItem *pItem, STSchema *pTSchema, - int64_t suid); +static int32_t tdRSmaFetchAllResult(SSma *pSma, SRSmaInfo *pInfo, SArray *pSubmitArr); +static int32_t tdRSmaExecAndSubmitResult(SSma *pSma, qTaskInfo_t taskInfo, SRSmaInfoItem *pItem, STSchema *pTSchema, + int64_t suid); static void tdRSmaFetchTrigger(void *param, void *tmrId); -static int32_t tdRSmaFetchSend(SSma *pSma, SRSmaInfo *pInfo, int8_t level); static int32_t tdRSmaQTaskInfoIterInit(SRSmaQTaskInfoIter *pIter, STFile *pTFile); static int32_t tdRSmaQTaskInfoIterNextBlock(SRSmaQTaskInfoIter *pIter, bool *isFinish); static int32_t tdRSmaQTaskInfoRestore(SSma *pSma, int8_t type, SRSmaQTaskInfoIter *pIter); @@ -635,8 +636,8 @@ _end: return code; } -static int32_t tdRSmaFetchAndSubmitResult(SSma *pSma, qTaskInfo_t taskInfo, SRSmaInfoItem *pItem, STSchema *pTSchema, - int64_t suid) { +static int32_t tdRSmaExecAndSubmitResult(SSma *pSma, qTaskInfo_t taskInfo, SRSmaInfoItem *pItem, STSchema *pTSchema, + int64_t suid) { SArray *pResList = taosArrayInit(1, POINTER_BYTES); if (pResList == NULL) { terrno = TSDB_CODE_OUT_OF_MEMORY; @@ -729,22 +730,24 @@ static int32_t tdExecuteRSmaImplAsync(SSma *pSma, const void *pMsg, int32_t inpu taosWriteQitem(pInfo->queue, qItem); + pInfo->lastRecv = taosGetTimestampMs(); + SRSmaStat *pRSmaStat = SMA_RSMA_STAT(pSma); - int64_t bufSize = atomic_add_fetch_64(&pRSmaStat->qBufSize, pReq->header.contLen); + + tsem_post(&(pRSmaStat->notEmpty)); + + int64_t nItems = atomic_fetch_add_64(&pRSmaStat->nBufItems, 1); // smoothing consume - int32_t n = bufSize / RSMA_QTASKEXEC_BUFSIZE; + int32_t n = nItems / RSMA_QTASKEXEC_SMOOTH_SIZE; if (n > 1) { if (n > 10) { n = 10; } - taosMsleep(n << 4); - if (n > 2) { + taosMsleep(n << 3); + if (n > 5) { smaWarn("vgId:%d, pInfo->queue itemSize:%d, memSize:%" PRIi64 ", sleep %d ms", SMA_VID(pSma), - taosQueueItemSize(pInfo->queue), taosQueueMemorySize(pInfo->queue), n << 4); - } else { - smaDebug("vgId:%d, pInfo->queue itemSize:%d, memSize:%" PRIi64 ", sleep %d ms", SMA_VID(pSma), - taosQueueItemSize(pInfo->queue), taosQueueMemorySize(pInfo->queue), n << 4); + taosQueueItemSize(pInfo->queue), taosQueueMemorySize(pInfo->queue), n << 3); } } @@ -812,7 +815,7 @@ static int32_t tdExecuteRSmaImpl(SSma *pSma, const void *pMsg, int32_t msgSize, } SRSmaInfoItem *pItem = RSMA_INFO_ITEM(pInfo, idx); - tdRSmaFetchAndSubmitResult(pSma, qTaskInfo, pItem, pInfo->pTSchema, pInfo->suid); + tdRSmaExecAndSubmitResult(pSma, qTaskInfo, pItem, pInfo->pTSchema, pInfo->suid); return TSDB_CODE_SUCCESS; } @@ -840,25 +843,25 @@ static SRSmaInfo *tdAcquireRSmaInfoBySuid(SSma *pSma, int64_t suid) { return NULL; } - taosRLockLatch(SMA_ENV_LOCK(pEnv)); + // taosRLockLatch(SMA_ENV_LOCK(pEnv)); pRSmaInfo = taosHashGet(RSMA_INFO_HASH(pStat), &suid, sizeof(tb_uid_t)); if (pRSmaInfo && (pRSmaInfo = *(SRSmaInfo **)pRSmaInfo)) { if (RSMA_INFO_IS_DEL(pRSmaInfo)) { - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); + // taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); return NULL; } if (!pRSmaInfo->taskInfo[0]) { if (tdCloneRSmaInfo(pSma, pRSmaInfo) < 0) { - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); + // taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); return NULL; } } tdRefRSmaInfo(pSma, pRSmaInfo); - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); + // taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); ASSERT(pRSmaInfo->suid == suid); return pRSmaInfo; } - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); + // taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); return NULL; } @@ -910,22 +913,11 @@ static int32_t tdExecuteRSmaAsync(SSma *pSma, const void *pMsg, int32_t inputTyp static int32_t tdRSmaExecCheck(SSma *pSma) { SRSmaStat *pRSmaStat = SMA_RSMA_STAT(pSma); - int64_t bufSize = atomic_load_64(&pRSmaStat->qBufSize); - - if (bufSize < RSMA_QTASKEXEC_BUFSIZE) { - smaDebug("vgId:%d, bufSize is %d but has no chance to exec as less than %d", SMA_VID(pSma), bufSize, - RSMA_QTASKEXEC_BUFSIZE); - return TSDB_CODE_SUCCESS; - } - if (atomic_val_compare_exchange_8(&pRSmaStat->execStat, 0, 1) == 1) { - smaDebug("vgId:%d, bufSize is %d but has no chance to exec as qTaskInfo occupied by another task", SMA_VID(pSma), - bufSize); + if (atomic_load_8(&pRSmaStat->nExecutor) >= TMIN(RSMA_EXECUTOR_MAX, tsNumOfVnodeQueryThreads / 2)) { return TSDB_CODE_SUCCESS; } - smaDebug("vgId:%d, bufSize is %d and has chance to exec as qTaskInfo is free now", SMA_VID(pSma), bufSize); - SRSmaExecMsg fetchMsg; int32_t contLen = sizeof(SMsgHead); void *pBuf = rpcMallocCont(0 + contLen); @@ -949,7 +941,6 @@ static int32_t tdRSmaExecCheck(SSma *pSma) { return TSDB_CODE_SUCCESS; _err: - atomic_store_8(&pRSmaStat->execStat, 0); return TSDB_CODE_FAILED; } @@ -959,7 +950,7 @@ int32_t tdProcessRSmaSubmit(SSma *pSma, void *pMsg, int32_t inputType) { // only applicable when rsma env exists return TSDB_CODE_SUCCESS; } - + STbUidStore uidStore = {0}; SRetention *pRetention = SMA_RETENTION(pSma); if (!RETENTION_VALID(pRetention + 1)) { // return directly if retention level 1 is invalid @@ -967,25 +958,34 @@ int32_t tdProcessRSmaSubmit(SSma *pSma, void *pMsg, int32_t inputType) { } if (inputType == STREAM_INPUT__DATA_SUBMIT) { - STbUidStore uidStore = {0}; - tdFetchSubmitReqSuids(pMsg, &uidStore); + if (tdFetchSubmitReqSuids(pMsg, &uidStore) < 0) { + goto _err; + } if (uidStore.suid != 0) { - tdExecuteRSmaAsync(pSma, pMsg, inputType, uidStore.suid); + if (tdExecuteRSmaAsync(pSma, pMsg, inputType, uidStore.suid) < 0) { + goto _err; + } - void *pIter = taosHashIterate(uidStore.uidHash, NULL); - while (pIter) { + void *pIter = NULL; + while ((pIter = taosHashIterate(uidStore.uidHash, pIter))) { tb_uid_t *pTbSuid = (tb_uid_t *)taosHashGetKey(pIter, NULL); - tdExecuteRSmaAsync(pSma, pMsg, inputType, *pTbSuid); - pIter = taosHashIterate(uidStore.uidHash, pIter); + if (tdExecuteRSmaAsync(pSma, pMsg, inputType, *pTbSuid) < 0) { + goto _err; + } } - tdUidStoreDestory(&uidStore); - - tdRSmaExecCheck(pSma); + if (tdRSmaExecCheck(pSma) < 0) { + goto _err; + } } } + tdUidStoreDestory(&uidStore); return TSDB_CODE_SUCCESS; +_err: + tdUidStoreDestory(&uidStore); + smaError("vgId:%d, failed to process rsma submit since: %s", SMA_VID(pSma), terrstr()); + return TSDB_CODE_FAILED; } /** @@ -1416,7 +1416,10 @@ int32_t tdRSmaPersistExecImpl(SRSmaStat *pRSmaStat, SHashObj *pInfoHash) { } for (int32_t i = 0; i < TSDB_RETENTION_L2; ++i) { +#if 0 qTaskInfo_t taskInfo = RSMA_INFO_IQTASK(pRSmaInfo, i); +#endif + qTaskInfo_t taskInfo = RSMA_INFO_QTASK(pRSmaInfo, i); if (!taskInfo) { smaDebug("vgId:%d, rsma, table %" PRIi64 " level %d qTaskInfo is NULL", vid, pRSmaInfo->suid, i + 1); continue; @@ -1553,7 +1556,16 @@ static void tdRSmaFetchTrigger(void *param, void *tmrId) { smaDebug("vgId:%d, rsma fetch task started for level:%" PRIi8 " suid:%" PRIi64 " since stat is active", SMA_VID(pSma), pItem->level, pRSmaInfo->suid); // async process - tdRSmaFetchSend(pSma, pRSmaInfo, pItem->level); + pItem->fetchLevel = pItem->level; +#if 0 + SRSmaInfo *qInfo = tdAcquireRSmaInfoBySuid(pSma, pRSmaInfo->suid); + SRSmaInfoItem *qItem = RSMA_INFO_ITEM(qInfo, pItem->level - 1); + ASSERT(qItem->level == pItem->level); + ASSERT(qItem->fetchLevel == pItem->fetchLevel); +#endif + tsem_post(&(pStat->notEmpty)); + smaInfo("vgId:%d, rsma fetch task planned for level:%" PRIi8 " suid:%" PRIi64, SMA_VID(pSma), pItem->level, + pRSmaInfo->suid); } break; case TASK_TRIGGER_STAT_PAUSED: { smaDebug("vgId:%d, rsma fetch task not start for level:%" PRIi8 " suid:%" PRIi64 " since stat is paused", @@ -1568,8 +1580,8 @@ static void tdRSmaFetchTrigger(void *param, void *tmrId) { SMA_VID(pSma), pItem->level, pRSmaInfo->suid); } break; default: { - smaWarn("vgId:%d, rsma fetch task not start for level:%" PRIi8 " suid:%" PRIi64 " since stat is unknown", - SMA_VID(pSma), pItem->level, pRSmaInfo->suid); + smaDebug("vgId:%d, rsma fetch task not start for level:%" PRIi8 " suid:%" PRIi64 " since stat is unknown", + SMA_VID(pSma), pItem->level, pRSmaInfo->suid); } break; } @@ -1578,183 +1590,62 @@ _end: tdReleaseSmaRef(smaMgmt.rsetId, pRSmaInfo->refId); } -/** - * @brief put rsma fetch msg to fetch queue - * - * @param pSma - * @param pInfo - * @param level - * @return int32_t - */ -static int32_t tdRSmaFetchSend(SSma *pSma, SRSmaInfo *pInfo, int8_t level) { - SRSmaFetchMsg fetchMsg = {.suid = pInfo->suid, .level = level}; - int32_t ret = 0; - int32_t contLen = 0; - SEncoder encoder = {0}; - tEncodeSize(tEncodeSRSmaFetchMsg, &fetchMsg, contLen, ret); - if (ret < 0) { - terrno = TSDB_CODE_OUT_OF_MEMORY; - tEncoderClear(&encoder); - goto _err; - } - - void *pBuf = rpcMallocCont(contLen + sizeof(SMsgHead)); - tEncoderInit(&encoder, POINTER_SHIFT(pBuf, sizeof(SMsgHead)), contLen); - if (tEncodeSRSmaFetchMsg(&encoder, &fetchMsg) < 0) { - terrno = TSDB_CODE_OUT_OF_MEMORY; - tEncoderClear(&encoder); - } - tEncoderClear(&encoder); - - ((SMsgHead *)pBuf)->vgId = SMA_VID(pSma); - ((SMsgHead *)pBuf)->contLen = contLen + sizeof(SMsgHead); - - SRpcMsg rpcMsg = { - .code = 0, - .msgType = TDMT_VND_FETCH_RSMA, - .pCont = pBuf, - .contLen = contLen + sizeof(SMsgHead), - }; - - if ((terrno = tmsgPutToQueue(&pSma->pVnode->msgCb, QUERY_QUEUE, &rpcMsg)) != 0) { - smaError("vgId:%d, failed to put rsma fetch msg into fetch-queue for suid:%" PRIi64 " level:%" PRIi8 " since %s", - SMA_VID(pSma), pInfo->suid, level, terrstr()); - goto _err; +static void tdFreeRSmaSubmitItems(SArray *pItems) { + for (int32_t i = 0; i < taosArrayGetSize(pItems); ++i) { + taosFreeQitem(*(void **)taosArrayGet(pItems, i)); } - - smaDebug("vgId:%d, success to put rsma fetch msg into fetch-queue for suid:%" PRIi64 " level:%" PRIi8, SMA_VID(pSma), - pInfo->suid, level); - - return TSDB_CODE_SUCCESS; -_err: - return TSDB_CODE_FAILED; } /** - * @brief fetch rsma data of level 2/3 and submit + * @brief fetch rsma result(consider the efficiency and functionality) * * @param pSma - * @param pMsg + * @param pInfo + * @param pSubmitArr * @return int32_t */ -int32_t smaProcessFetch(SSma *pSma, void *pMsg) { - SRpcMsg *pRpcMsg = (SRpcMsg *)pMsg; - SRSmaFetchMsg req = {0}; - SDecoder decoder = {0}; - void *pBuf = NULL; - SRSmaStat *pRSmaStat = NULL; - if (!pRpcMsg || pRpcMsg->contLen < sizeof(SMsgHead)) { - terrno = TSDB_CODE_RSMA_FETCH_MSG_MSSED_UP; - goto _err; - } - - pBuf = POINTER_SHIFT(pRpcMsg->pCont, sizeof(SMsgHead)); - - tDecoderInit(&decoder, pBuf, pRpcMsg->contLen); - if (tDecodeSRSmaFetchMsg(&decoder, &req) < 0) { - terrno = TSDB_CODE_INVALID_MSG; - goto _err; - } - - pRSmaStat = SMA_RSMA_STAT(pSma); - - if (atomic_val_compare_exchange_8(&pRSmaStat->execStat, 0, 1) == 0) { - SArray *pSubmitArr = NULL; - if (!(pSubmitArr = taosArrayInit(RSMA_SUBMIT_BATCH_SIZE, POINTER_BYTES))) { - terrno = TSDB_CODE_OUT_OF_MEMORY; - atomic_store_8(&pRSmaStat->execStat, 0); - goto _err; - } - tdRSmaConsumeAndFetch(pSma, req.suid, req.level, pSubmitArr); - atomic_store_8(&pRSmaStat->execStat, 0); - taosArrayDestroy(pSubmitArr); - } else { - int8_t level = req.level; - int8_t *val = taosHashGet(RSMA_FETCH_HASH(pRSmaStat), &req.suid, sizeof(req.suid)); - if (val) { - level |= (*val); - } - ASSERT(level >= 1 && level <= 3); - taosHashPut(RSMA_FETCH_HASH(pRSmaStat), &req.suid, sizeof(req.suid), &level, sizeof(level)); - } - - tDecoderClear(&decoder); - smaDebug("vgId:%d, success to process rsma fetch msg for suid:%" PRIi64 " level:%" PRIi8, SMA_VID(pSma), req.suid, - req.level); - return TSDB_CODE_SUCCESS; -_err: - tDecoderClear(&decoder); - smaError("vgId:%d, failed to process rsma fetch msg since %s", SMA_VID(pSma), terrstr()); - return TSDB_CODE_FAILED; -} - -static void tdFreeRSmaSubmitItems(SArray *pItems) { - for (int32_t i = 0; i < taosArrayGetSize(pItems); ++i) { - taosFreeQitem(*(void **)taosArrayGet(pItems, i)); - } -} - -static int32_t tdRSmaConsumeAndFetch(SSma *pSma, int64_t suid, int8_t level, SArray *pSubmitArr) { - SRSmaInfo *pInfo = tdAcquireRSmaInfoBySuid(pSma, suid); - if (!pInfo) { - return TSDB_CODE_SUCCESS; - } - - // step 1: consume submit req - int64_t qMemSize = 0; - if ((qMemSize = taosQueueMemorySize(pInfo->queue) > 0)) { - taosReadAllQitems(pInfo->queue, pInfo->qall); // queue has mutex lock - - SRSmaStat *pRSmaStat = SMA_RSMA_STAT(pSma); - atomic_fetch_sub_64(&pRSmaStat->qBufSize, qMemSize); - - taosArrayClear(pSubmitArr); - - while (1) { - void *msg = NULL; - taosGetQitem(pInfo->qall, (void **)&msg); - if (msg) { - if (taosArrayPush(pSubmitArr, &msg) < 0) { - tdFreeRSmaSubmitItems(pSubmitArr); - goto _err; - } - } else { - break; - } - } - - int32_t size = taosArrayGetSize(pSubmitArr); - if (size > 0) { - for (int32_t i = 1; i <= TSDB_RETENTION_L2; ++i) { - if (tdExecuteRSmaImpl(pSma, pSubmitArr->pData, size, STREAM_INPUT__MERGED_SUBMIT, pInfo, RSMA_EXEC_TIMEOUT, i) < - 0) { - tdFreeRSmaSubmitItems(pSubmitArr); - goto _err; - } - } - - tdFreeRSmaSubmitItems(pSubmitArr); - } - } - - // step 2: fetch rsma result +static int32_t tdRSmaFetchAllResult(SSma *pSma, SRSmaInfo *pInfo, SArray *pSubmitArr) { SSDataBlock dataBlock = {.info.type = STREAM_GET_ALL}; for (int8_t i = 1; i <= TSDB_RETENTION_L2; ++i) { - if (level & i) { + SRSmaInfoItem *pItem = RSMA_INFO_ITEM(pInfo, i - 1); + if (pItem->fetchLevel) { + pItem->fetchLevel = 0; qTaskInfo_t taskInfo = RSMA_INFO_QTASK(pInfo, i - 1); if (!taskInfo) { continue; } + + int64_t curMs = taosGetTimestampMs(); + if ((pItem->nSkipped * pItem->maxDelay) > RSMA_FETCH_DELAY_MAX) { + smaInfo("vgId:%d, suid:%" PRIi64 " level:%" PRIi8 " nSkipped:%" PRIi8 " maxDelay:%d, fetch executed", + SMA_VID(pSma), pInfo->suid, i, pItem->nSkipped, pItem->maxDelay); + } else if (((curMs - pInfo->lastRecv) < RSMA_FETCH_ACTIVE_MAX)) { + ++pItem->nSkipped; + smaDebug("vgId:%d, suid:%" PRIi64 " level:%" PRIi8 " curMs:%" PRIi64 " lastRecv:%" PRIi64 ", fetch skipped ", + SMA_VID(pSma), pInfo->suid, i, curMs, pInfo->lastRecv); + continue; + } else { + smaInfo("vgId:%d, suid:%" PRIi64 " level:%" PRIi8 " curMs:%" PRIi64 " lastRecv:%" PRIi64 ", fetch executed ", + SMA_VID(pSma), pInfo->suid, i, curMs, pInfo->lastRecv); + } + + pItem->nSkipped = 0; + if ((terrno = qSetMultiStreamInput(taskInfo, &dataBlock, 1, STREAM_INPUT__DATA_BLOCK)) < 0) { goto _err; } - SRSmaInfoItem *pItem = RSMA_INFO_ITEM(pInfo, i - 1); - if (tdRSmaFetchAndSubmitResult(pSma, taskInfo, pItem, pInfo->pTSchema, suid) < 0) { + if (tdRSmaExecAndSubmitResult(pSma, taskInfo, pItem, pInfo->pTSchema, pInfo->suid) < 0) { tdCleanupStreamInputDataBlock(taskInfo); goto _err; } tdCleanupStreamInputDataBlock(taskInfo); + smaInfo("vgId:%d, suid:%" PRIi64 " level:%" PRIi8 " nSkipped:%" PRIi8 " maxDelay:%d, fetch finished", + SMA_VID(pSma), pInfo->suid, i, pItem->nSkipped, pItem->maxDelay); + } else { + smaDebug("vgId:%d, suid:%" PRIi64 " level:%" PRIi8 " nSkipped:%" PRIi8 + " maxDelay:%d, fetch not executed as fetch level is %" PRIi8, + SMA_VID(pSma), pInfo->suid, i, pItem->nSkipped, pItem->maxDelay, pItem->fetchLevel); } } @@ -1766,6 +1657,45 @@ _err: return TSDB_CODE_FAILED; } +static int32_t tdRSmaBatchExec(SSma *pSma, SRSmaInfo *pInfo, STaosQall *qall, SArray *pSubmitArr, ERsmaExecType type) { + taosArrayClear(pSubmitArr); + while (1) { + void *msg = NULL; + taosGetQitem(qall, (void **)&msg); + if (msg) { + if (taosArrayPush(pSubmitArr, &msg) < 0) { + tdFreeRSmaSubmitItems(pSubmitArr); + goto _err; + } + } else { + break; + } + } + + int32_t size = taosArrayGetSize(pSubmitArr); + if (size > 0) { + for (int32_t i = 1; i <= TSDB_RETENTION_L2; ++i) { + if (tdExecuteRSmaImpl(pSma, pSubmitArr->pData, size, STREAM_INPUT__MERGED_SUBMIT, pInfo, type, i) < 0) { + tdFreeRSmaSubmitItems(pSubmitArr); + goto _err; + } + } + tdFreeRSmaSubmitItems(pSubmitArr); + } + return TSDB_CODE_SUCCESS; +_err: + while (1) { + void *msg = NULL; + taosGetQitem(qall, (void **)&msg); + if (msg) { + taosFreeQitem(msg); + } else { + break; + } + } + return TSDB_CODE_FAILED; +} + /** * @brief * @@ -1774,10 +1704,10 @@ _err: * @return int32_t */ int32_t tdRSmaProcessExecImpl(SSma *pSma, ERsmaExecType type) { + SVnode *pVnode = pSma->pVnode; SSmaEnv *pEnv = SMA_RSMA_ENV(pSma); SRSmaStat *pRSmaStat = (SRSmaStat *)SMA_ENV_STAT(pEnv); SHashObj *infoHash = NULL; - SArray *pSubmitQArr = NULL; SArray *pSubmitArr = NULL; bool isFetchAll = false; @@ -1786,106 +1716,97 @@ int32_t tdRSmaProcessExecImpl(SSma *pSma, ERsmaExecType type) { goto _err; } - if (type == RSMA_EXEC_OVERFLOW) { - taosRLockLatch(SMA_ENV_LOCK(pEnv)); - if (atomic_load_64(&pRSmaStat->qBufSize) < RSMA_QTASKEXEC_BUFSIZE) { - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); - return TSDB_CODE_SUCCESS; - } - taosRUnLockLatch(SMA_ENV_LOCK(pEnv)); - } - - if (!(pSubmitQArr = taosArrayInit(taosHashGetSize(infoHash), sizeof(SRSmaExecQItem)))) { - terrno = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - - if (!(pSubmitArr = taosArrayInit(RSMA_SUBMIT_BATCH_SIZE, POINTER_BYTES))) { + if (!(pSubmitArr = + taosArrayInit(TMIN(RSMA_SUBMIT_BATCH_SIZE, atomic_load_64(&pRSmaStat->nBufItems)), POINTER_BYTES))) { terrno = TSDB_CODE_OUT_OF_MEMORY; goto _err; } - // step 1: rsma exec - consume data in buffer queue for all suids - SRSmaExecQItem qItem = {0}; - void *pIter = taosHashIterate(infoHash, NULL); // infoHash has r/w lock - if (type == RSMA_EXEC_OVERFLOW) { - while (pIter) { - SRSmaInfo *pInfo = *(SRSmaInfo **)pIter; - if (taosQueueItemSize(pInfo->queue)) { - taosReadAllQitems(pInfo->queue, pInfo->qall); // queue has mutex lock - qItem.qall = &pInfo->qall; - qItem.pRSmaInfo = pIter; - taosArrayPush(pSubmitQArr, &qItem); - } - ASSERT(taosQueueItemSize(pInfo->queue) == 0); - pIter = taosHashIterate(infoHash, pIter); - } - } else if (type == RSMA_EXEC_COMMIT) { - while (pIter) { - SRSmaInfo *pInfo = *(SRSmaInfo **)pIter; - if (taosQueueItemSize(pInfo->iQueue)) { - taosReadAllQitems(pInfo->iQueue, pInfo->iQall); - qItem.qall = &pInfo->iQall; - qItem.pRSmaInfo = pIter; - taosArrayPush(pSubmitQArr, &qItem); - } - ASSERT(taosQueueItemSize(pInfo->iQueue) == 0); - pIter = taosHashIterate(infoHash, pIter); - } - } else { - ASSERT(0); - } - atomic_store_64(&pRSmaStat->qBufSize, 0); - - int32_t qSize = taosArrayGetSize(pSubmitQArr); - for (int32_t i = 0; i < qSize; ++i) { - SRSmaExecQItem *pItem = taosArrayGet(pSubmitQArr, i); - while (1) { - void *msg = NULL; - taosGetQitem(*(STaosQall **)pItem->qall, (void **)&msg); - if (msg) { - if (taosArrayPush(pSubmitArr, &msg) < 0) { - tdFreeRSmaSubmitItems(pSubmitArr); - goto _err; + bool isBusy = false; + while (true) { + isBusy = false; + // step 1: rsma exec - consume data in buffer queue for all suids + if (type == RSMA_EXEC_OVERFLOW || type == RSMA_EXEC_COMMIT) { + void *pIter = taosHashIterate(infoHash, NULL); // infoHash has r/w lock + while (pIter) { + SRSmaInfo *pInfo = *(SRSmaInfo **)pIter; + int64_t itemSize = 0; + if ((itemSize = taosQueueItemSize(pInfo->queue)) || RSMA_INFO_ITEM(pInfo, 0)->fetchLevel || + RSMA_INFO_ITEM(pInfo, 1)->fetchLevel) { + smaDebug("vgId:%d, queueItemSize is %" PRIi64 " execType:%" PRIi8, SMA_VID(pSma), itemSize, type); + if (atomic_val_compare_exchange_8(&pInfo->assigned, 0, 1) == 0) { + taosReadAllQitems(pInfo->queue, pInfo->qall); // queue has mutex lock + int32_t qallItemSize = taosQallItemSize(pInfo->qall); + if (qallItemSize > 0) { + tdRSmaBatchExec(pSma, pInfo, pInfo->qall, pSubmitArr, type); + } + + if (type == RSMA_EXEC_OVERFLOW) { + tdRSmaFetchAllResult(pSma, pInfo, pSubmitArr); + } + + if (qallItemSize > 0) { + // subtract the item size after the task finished, commit should wait for all items be consumed + atomic_fetch_sub_64(&pRSmaStat->nBufItems, qallItemSize); + isBusy = true; + } + ASSERT(1 == atomic_val_compare_exchange_8(&pInfo->assigned, 1, 0)); + } } - } else { + pIter = taosHashIterate(infoHash, pIter); + } + if (type == RSMA_EXEC_COMMIT) { break; } } - - int32_t size = taosArrayGetSize(pSubmitArr); - if (size > 0) { - SRSmaInfo *pInfo = *(SRSmaInfo **)pItem->pRSmaInfo; - for (int32_t i = 1; i <= TSDB_RETENTION_L2; ++i) { - if (tdExecuteRSmaImpl(pSma, pSubmitArr->pData, size, STREAM_INPUT__MERGED_SUBMIT, pInfo, type, i) < 0) { - tdFreeRSmaSubmitItems(pSubmitArr); - goto _err; +#if 0 + else if (type == RSMA_EXEC_COMMIT) { + while (pIter) { + SRSmaInfo *pInfo = *(SRSmaInfo **)pIter; + if (taosQueueItemSize(pInfo->iQueue)) { + if (atomic_val_compare_exchange_8(&pInfo->assigned, 0, 1) == 0) { + taosReadAllQitems(pInfo->iQueue, pInfo->iQall); // queue has mutex lock + int32_t qallItemSize = taosQallItemSize(pInfo->iQall); + if (qallItemSize > 0) { + atomic_fetch_sub_64(&pRSmaStat->nBufItems, qallItemSize); + nIdle = 0; + + // batch exec + tdRSmaBatchExec(pSma, pInfo, pInfo->qall, pSubmitArr, type); + } + + // tdRSmaFetchAllResult(pSma, pInfo, pSubmitArr); + ASSERT(1 == atomic_val_compare_exchange_8(&pInfo->assigned, 1, 0)); + } } + ASSERT(taosQueueItemSize(pInfo->iQueue) == 0); + pIter = taosHashIterate(infoHash, pIter); } - tdFreeRSmaSubmitItems(pSubmitArr); - taosArrayClear(pSubmitArr); + break; + } +#endif + else { + ASSERT(0); } - } - // step 2: rsma fetch - consume data in buffer queue for suids triggered by timer - if (taosHashGetSize(RSMA_FETCH_HASH(pRSmaStat)) <= 0) { - goto _end; - } - pIter = taosHashIterate(RSMA_FETCH_HASH(pRSmaStat), NULL); - if (pIter) { - tdRSmaConsumeAndFetch(pSma, *(int64_t *)taosHashGetKey(pIter, NULL), *(int8_t *)pIter, pSubmitArr); - while ((pIter = taosHashIterate(RSMA_FETCH_HASH(pRSmaStat), pIter))) { - tdRSmaConsumeAndFetch(pSma, *(int64_t *)taosHashGetKey(pIter, NULL), *(int8_t *)pIter, pSubmitArr); + if (atomic_load_64(&pRSmaStat->nBufItems) <= 0) { + if (pVnode->inClose) { + break; + } + tsem_wait(&pRSmaStat->notEmpty); + if (pVnode->inClose && (atomic_load_64(&pRSmaStat->nBufItems) <= 0)) { + smaInfo("vgId:%d, exec task end, inClose:%d, nBufItems:%" PRIi64, SMA_VID(pSma), pVnode->inClose, + atomic_load_64(&pRSmaStat->nBufItems)); + break; + } } - } + } // end of while(true) _end: taosArrayDestroy(pSubmitArr); - taosArrayDestroy(pSubmitQArr); return TSDB_CODE_SUCCESS; _err: taosArrayDestroy(pSubmitArr); - taosArrayDestroy(pSubmitQArr); return TSDB_CODE_FAILED; } @@ -1905,15 +1826,21 @@ int32_t smaProcessExec(SSma *pSma, void *pMsg) { goto _err; } smaDebug("vgId:%d, begin to process rsma exec msg by TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); - if (tdRSmaProcessExecImpl(pSma, RSMA_EXEC_OVERFLOW) < 0) { - goto _err; + + int8_t nOld = atomic_fetch_add_8(&pRSmaStat->nExecutor, 1); + + if (nOld < TMIN(RSMA_EXECUTOR_MAX, tsNumOfVnodeQueryThreads / 2)) { + if (tdRSmaProcessExecImpl(pSma, RSMA_EXEC_OVERFLOW) < 0) { + goto _err; + } + } else { + atomic_fetch_sub_8(&pRSmaStat->nExecutor, 1); } - atomic_store_8(&pRSmaStat->execStat, 0); smaDebug("vgId:%d, success to process rsma exec msg by TID:%p", SMA_VID(pSma), (void *)taosGetSelfPthreadId()); return TSDB_CODE_SUCCESS; _err: - atomic_store_8(&pRSmaStat->execStat, 0); + atomic_fetch_sub_8(&pRSmaStat->nExecutor, 1); smaError("vgId:%d, failed to process rsma exec msg by TID:%p since %s", SMA_VID(pSma), (void *)taosGetSelfPthreadId(), terrstr()); return TSDB_CODE_FAILED; diff --git a/source/dnode/vnode/src/sma/smaUtil.c b/source/dnode/vnode/src/sma/smaUtil.c index da7022248526d7371bf40a0e91c7a0f7d49b8a16..d771797963a5cd9d242fea1f4d65a5634f12b5e8 100644 --- a/source/dnode/vnode/src/sma/smaUtil.c +++ b/source/dnode/vnode/src/sma/smaUtil.c @@ -375,6 +375,9 @@ int32_t tdCloneRSmaInfo(SSma *pSma, SRSmaInfo *pInfo) { if (TABLE_IS_ROLLUP(mr.me.flags)) { param = &mr.me.stbEntry.rsmaParam; for (int32_t i = 0; i < TSDB_RETENTION_L2; ++i) { + if (!pInfo->iTaskInfo[i]) { + continue; + } if (tdCloneQTaskInfo(pSma, pInfo->taskInfo[i], pInfo->iTaskInfo[i], param, pInfo->suid, i) < 0) { goto _err; } diff --git a/source/dnode/vnode/src/tsdb/tsdbCache.c b/source/dnode/vnode/src/tsdb/tsdbCache.c index bb367ff8b1a91430b6864522955f4bb0b2f1580a..ed25783e9f416b7a0bada836c81ee801893d0e33 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCache.c +++ b/source/dnode/vnode/src/tsdb/tsdbCache.c @@ -266,14 +266,14 @@ int32_t tsdbCacheInsertLast(SLRUCache *pCache, tb_uid_t uid, STSRow *row, STsdb } for (++iCol; iCol < nCol; ++iCol) { - SLastCol *tTsVal = (SLastCol *)taosArrayGet(pLast, iCol); - if (keyTs >= tTsVal->ts) { - SColVal *tColVal = &tTsVal->colVal; + SLastCol *tTsVal1 = (SLastCol *)taosArrayGet(pLast, iCol); + if (keyTs >= tTsVal1->ts) { + SColVal *tColVal = &tTsVal1->colVal; SColVal colVal = {0}; tTSRowGetVal(row, pTSchema, iCol, &colVal); if (colVal.isNone || colVal.isNull) { - if (keyTs == tTsVal->ts && !tColVal->isNone && !tColVal->isNull) { + if (keyTs == tTsVal1->ts && !tColVal->isNone && !tColVal->isNull) { invalidate = true; break; @@ -284,6 +284,7 @@ int32_t tsdbCacheInsertLast(SLRUCache *pCache, tb_uid_t uid, STSRow *row, STsdb } } + _invalidate: taosMemoryFreeClear(pTSchema); taosLRUCacheRelease(pCache, h, invalidate); @@ -322,7 +323,7 @@ static int32_t getTableDelDataFromDelIdx(SDelFReader *pDelReader, SDelIdx *pDelI int32_t code = 0; if (pDelIdx) { - code = tsdbReadDelData(pDelReader, pDelIdx, aDelData, NULL); + code = tsdbReadDelData(pDelReader, pDelIdx, aDelData); } return code; @@ -393,8 +394,7 @@ static int32_t getTableDelIdx(SDelFReader *pDelFReader, tb_uid_t suid, tb_uid_t SDelIdx idx = {.suid = suid, .uid = uid}; // tMapDataReset(&delIdxMap); - // code = tsdbReadDelIdx(pDelFReader, &delIdxMap, NULL); - code = tsdbReadDelIdx(pDelFReader, pDelIdxArray, NULL); + code = tsdbReadDelIdx(pDelFReader, pDelIdxArray); if (code) goto _err; // code = tMapDataSearch(&delIdxMap, &idx, tGetDelIdx, tCmprDelIdx, pDelIdx); @@ -410,6 +410,178 @@ _err: return code; } +typedef enum { + SFSLASTNEXTROW_FS, + SFSLASTNEXTROW_FILESET, + SFSLASTNEXTROW_BLOCKDATA, + SFSLASTNEXTROW_BLOCKROW +} SFSLASTNEXTROWSTATES; + +typedef struct { + SFSLASTNEXTROWSTATES state; // [input] + STsdb *pTsdb; // [input] + SBlockIdx *pBlockIdxExp; // [input] + STSchema *pTSchema; // [input] + int32_t nFileSet; + int32_t iFileSet; + SArray *aDFileSet; + SDataFReader *pDataFReader; + SArray *aBlockL; + SBlockL *pBlockL; + SBlockData *pBlockDataL; + SBlockData blockDataL; + int32_t nRow; + int32_t iRow; + TSDBROW row; + /* + SArray *aBlockIdx; + SBlockIdx *pBlockIdx; + SMapData blockMap; + int32_t nBlock; + int32_t iBlock; + SBlock block; + */ +} SFSLastNextRowIter; + +static int32_t getNextRowFromFSLast(void *iter, TSDBROW **ppRow) { + SFSLastNextRowIter *state = (SFSLastNextRowIter *)iter; + int32_t code = 0; + + switch (state->state) { + case SFSLASTNEXTROW_FS: + // state->aDFileSet = state->pTsdb->pFS->cState->aDFileSet; + state->nFileSet = taosArrayGetSize(state->aDFileSet); + state->iFileSet = state->nFileSet; + + state->pBlockDataL = NULL; + + case SFSLASTNEXTROW_FILESET: { + SDFileSet *pFileSet = NULL; + _next_fileset: + if (--state->iFileSet >= 0) { + pFileSet = (SDFileSet *)taosArrayGet(state->aDFileSet, state->iFileSet); + } else { + if (state->pBlockDataL) { + tBlockDataDestroy(state->pBlockDataL, 1); + state->pBlockDataL = NULL; + } + + *ppRow = NULL; + return code; + } + + code = tsdbDataFReaderOpen(&state->pDataFReader, state->pTsdb, pFileSet); + if (code) goto _err; + + if (!state->aBlockL) { + state->aBlockL = taosArrayInit(0, sizeof(SBlockIdx)); + } else { + taosArrayClear(state->aBlockL); + } + + code = tsdbReadBlockL(state->pDataFReader, state->aBlockL); + if (code) goto _err; + + // SBlockL *pBlockL = (SBlockL *)taosArrayGet(state->aBlockL, state->iBlockL); + + state->pBlockL = taosArraySearch(state->aBlockL, state->pBlockIdxExp, tCmprBlockL, TD_EQ); + if (!state->pBlockL) { + goto _next_fileset; + } + + int64_t suid = state->pBlockL->suid; + int64_t uid = state->pBlockL->maxUid; + + if (!state->pBlockDataL) { + state->pBlockDataL = &state->blockDataL; + } + code = tBlockDataInit(state->pBlockDataL, suid, suid ? 0 : uid, state->pTSchema); + if (code) goto _err; + } + case SFSLASTNEXTROW_BLOCKDATA: + code = tsdbReadLastBlock(state->pDataFReader, state->pBlockL, state->pBlockDataL); + if (code) goto _err; + + state->nRow = state->blockDataL.nRow; + state->iRow = state->nRow - 1; + + if (!state->pBlockDataL->uid) { + while (state->pBlockIdxExp->uid != state->pBlockDataL->aUid[state->iRow]) { + --state->iRow; + } + } + + state->state = SFSLASTNEXTROW_BLOCKROW; + case SFSLASTNEXTROW_BLOCKROW: + if (state->pBlockDataL->uid) { + if (state->iRow >= 0) { + state->row = tsdbRowFromBlockData(state->pBlockDataL, state->iRow); + *ppRow = &state->row; + + if (--state->iRow < 0) { + state->state = SFSLASTNEXTROW_FILESET; + } + } + } else { + if (state->iRow >= 0 && state->pBlockIdxExp->uid == state->pBlockDataL->aUid[state->iRow]) { + state->row = tsdbRowFromBlockData(state->pBlockDataL, state->iRow); + *ppRow = &state->row; + + if (--state->iRow < 0 || state->pBlockIdxExp->uid != state->pBlockDataL->aUid[state->iRow]) { + state->state = SFSLASTNEXTROW_FILESET; + } + } + } + + return code; + default: + ASSERT(0); + break; + } + +_err: + if (state->pDataFReader) { + tsdbDataFReaderClose(&state->pDataFReader); + state->pDataFReader = NULL; + } + if (state->aBlockL) { + taosArrayDestroy(state->aBlockL); + state->aBlockL = NULL; + } + if (state->pBlockDataL) { + tBlockDataDestroy(state->pBlockDataL, 1); + state->pBlockDataL = NULL; + } + + *ppRow = NULL; + + return code; +} + +int32_t clearNextRowFromFSLast(void *iter) { + SFSLastNextRowIter *state = (SFSLastNextRowIter *)iter; + int32_t code = 0; + + if (!state) { + return code; + } + + if (state->pDataFReader) { + tsdbDataFReaderClose(&state->pDataFReader); + state->pDataFReader = NULL; + } + if (state->aBlockL) { + taosArrayDestroy(state->aBlockL); + state->aBlockL = NULL; + } + if (state->pBlockDataL) { + tBlockDataDestroy(state->pBlockDataL, 1); + state->pBlockDataL = NULL; + } + + return code; +} + typedef enum SFSNEXTROWSTATES { SFSNEXTROW_FS, SFSNEXTROW_FILESET, @@ -456,9 +628,9 @@ static int32_t getNextRowFromFS(void *iter, TSDBROW **ppRow) { if (--state->iFileSet >= 0) { pFileSet = (SDFileSet *)taosArrayGet(state->aDFileSet, state->iFileSet); } else { - // tBlockDataClear(&state->blockData, 1); + // tBlockDataDestroy(&state->blockData, 1); if (state->pBlockData) { - tBlockDataClear(state->pBlockData, 1); + tBlockDataDestroy(state->pBlockData, 1); state->pBlockData = NULL; } @@ -470,13 +642,12 @@ static int32_t getNextRowFromFS(void *iter, TSDBROW **ppRow) { if (code) goto _err; // tMapDataReset(&state->blockIdxMap); - // code = tsdbReadBlockIdx(state->pDataFReader, &state->blockIdxMap, NULL); if (!state->aBlockIdx) { state->aBlockIdx = taosArrayInit(0, sizeof(SBlockIdx)); } else { taosArrayClear(state->aBlockIdx); } - code = tsdbReadBlockIdx(state->pDataFReader, state->aBlockIdx, NULL); + code = tsdbReadBlockIdx(state->pDataFReader, state->aBlockIdx); if (code) goto _err; /* if (state->pBlockIdx) { */ @@ -492,8 +663,7 @@ static int32_t getNextRowFromFS(void *iter, TSDBROW **ppRow) { } tMapDataReset(&state->blockMap); - code = tsdbReadBlock(state->pDataFReader, state->pBlockIdx, &state->blockMap, NULL); - /* code = tsdbReadBlock(state->pDataFReader, &state->blockIdx, &state->blockMap, NULL); */ + code = tsdbReadBlock(state->pDataFReader, state->pBlockIdx, &state->blockMap); if (code) goto _err; state->nBlock = state->blockMap.nItem; @@ -502,7 +672,7 @@ static int32_t getNextRowFromFS(void *iter, TSDBROW **ppRow) { if (!state->pBlockData) { state->pBlockData = &state->blockData; - tBlockDataInit(&state->blockData); + tBlockDataCreate(&state->blockData); } } case SFSNEXTROW_BLOCKDATA: @@ -515,7 +685,7 @@ static int32_t getNextRowFromFS(void *iter, TSDBROW **ppRow) { tMapDataGetItemByIdx(&state->blockMap, state->iBlock, &block, tGetBlock); /* code = tsdbReadBlockData(state->pDataFReader, &state->blockIdx, &block, &state->blockData, NULL, NULL); */ - code = tsdbReadBlockData(state->pDataFReader, state->pBlockIdx, &block, state->pBlockData, NULL, NULL); + code = tsdbReadDataBlock(state->pDataFReader, &block, state->pBlockData); if (code) goto _err; state->nRow = state->blockData.nRow; @@ -560,8 +730,8 @@ _err: state->aBlockIdx = NULL; } if (state->pBlockData) { - // tBlockDataClear(&state->blockData, 1); - tBlockDataClear(state->pBlockData, 1); + // tBlockDataDestroy(&state->blockData, 1); + tBlockDataDestroy(state->pBlockData, 1); state->pBlockData = NULL; } @@ -587,8 +757,8 @@ int32_t clearNextRowFromFS(void *iter) { state->aBlockIdx = NULL; } if (state->pBlockData) { - // tBlockDataClear(&state->blockData, 1); - tBlockDataClear(state->pBlockData, 1); + // tBlockDataDestroy(&state->blockData, 1); + tBlockDataDestroy(state->pBlockData, 1); state->pBlockData = NULL; } @@ -730,18 +900,19 @@ typedef struct { SArray *pSkyline; int64_t iSkyline; - SBlockIdx idx; - SMemNextRowIter memState; - SMemNextRowIter imemState; - SFSNextRowIter fsState; - TSDBROW memRow, imemRow, fsRow; + SBlockIdx idx; + SMemNextRowIter memState; + SMemNextRowIter imemState; + SFSLastNextRowIter fsLastState; + SFSNextRowIter fsState; + TSDBROW memRow, imemRow, fsLastRow, fsRow; - TsdbNextRowState input[3]; + TsdbNextRowState input[4]; STsdbReadSnap *pReadSnap; STsdb *pTsdb; } CacheNextRowIter; -static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTsdb) { +static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTsdb, STSchema *pTSchema) { int code = 0; tb_uid_t suid = getTableSuidByUid(uid, pTsdb); @@ -750,12 +921,12 @@ static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTs STbData *pMem = NULL; if (pIter->pReadSnap->pMem) { - tsdbGetTbDataFromMemTable(pIter->pReadSnap->pMem, suid, uid, &pMem); + pMem = tsdbGetTbDataFromMemTable(pIter->pReadSnap->pMem, suid, uid); } STbData *pIMem = NULL; if (pIter->pReadSnap->pIMem) { - tsdbGetTbDataFromMemTable(pIter->pReadSnap->pIMem, suid, uid, &pIMem); + pIMem = tsdbGetTbDataFromMemTable(pIter->pReadSnap->pIMem, suid, uid); } pIter->pTsdb = pTsdb; @@ -768,7 +939,7 @@ static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTs if (pDelFile) { SDelFReader *pDelFReader; - code = tsdbDelFReaderOpen(&pDelFReader, pDelFile, pTsdb, NULL); + code = tsdbDelFReaderOpen(&pDelFReader, pDelFile, pTsdb); if (code) goto _err; code = getTableDelIdx(pDelFReader, suid, uid, &delIdx); @@ -787,6 +958,12 @@ static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTs pIter->idx = (SBlockIdx){.suid = suid, .uid = uid}; + pIter->fsLastState.state = (SFSLASTNEXTROWSTATES) SFSNEXTROW_FS; + pIter->fsLastState.pTsdb = pTsdb; + pIter->fsLastState.aDFileSet = pIter->pReadSnap->fs.aDFileSet; + pIter->fsLastState.pBlockIdxExp = &pIter->idx; + pIter->fsLastState.pTSchema = pTSchema; + pIter->fsState.state = SFSNEXTROW_FS; pIter->fsState.pTsdb = pTsdb; pIter->fsState.aDFileSet = pIter->pReadSnap->fs.aDFileSet; @@ -794,7 +971,9 @@ static int32_t nextRowIterOpen(CacheNextRowIter *pIter, tb_uid_t uid, STsdb *pTs pIter->input[0] = (TsdbNextRowState){&pIter->memRow, true, false, &pIter->memState, getNextRowFromMem, NULL}; pIter->input[1] = (TsdbNextRowState){&pIter->imemRow, true, false, &pIter->imemState, getNextRowFromMem, NULL}; - pIter->input[2] = + pIter->input[2] = (TsdbNextRowState){&pIter->fsLastRow, false, true, &pIter->fsLastState, getNextRowFromFSLast, + clearNextRowFromFSLast}; + pIter->input[3] = (TsdbNextRowState){&pIter->fsRow, false, true, &pIter->fsState, getNextRowFromFS, clearNextRowFromFS}; if (pMem) { @@ -819,7 +998,7 @@ _err: static int32_t nextRowIterClose(CacheNextRowIter *pIter) { int code = 0; - for (int i = 0; i < 3; ++i) { + for (int i = 0; i < 4; ++i) { if (pIter->input[i].nextRowClearFn) { pIter->input[i].nextRowClearFn(pIter->input[i].iter); } @@ -831,7 +1010,6 @@ static int32_t nextRowIterClose(CacheNextRowIter *pIter) { tsdbUntakeReadSnap(pIter->pTsdb, pIter->pReadSnap); - return code; _err: return code; } @@ -840,7 +1018,7 @@ _err: static int32_t nextRowIterGet(CacheNextRowIter *pIter, TSDBROW **ppRow) { int code = 0; - for (int i = 0; i < 3; ++i) { + for (int i = 0; i < 4; ++i) { if (pIter->input[i].next && !pIter->input[i].stop) { code = pIter->input[i].nextRowFn(pIter->input[i].iter, &pIter->input[i].pRow); if (code) goto _err; @@ -852,18 +1030,18 @@ static int32_t nextRowIterGet(CacheNextRowIter *pIter, TSDBROW **ppRow) { } } - if (pIter->input[0].stop && pIter->input[1].stop && pIter->input[2].stop) { + if (pIter->input[0].stop && pIter->input[1].stop && pIter->input[2].stop && pIter->input[3].stop) { *ppRow = NULL; return code; } - // select maxpoint(s) from mem, imem, fs - TSDBROW *max[3] = {0}; - int iMax[3] = {-1, -1, -1}; + // select maxpoint(s) from mem, imem, fs and last + TSDBROW *max[4] = {0}; + int iMax[4] = {-1, -1, -1, -1}; int nMax = 0; TSKEY maxKey = TSKEY_MIN; - for (int i = 0; i < 3; ++i) { + for (int i = 0; i < 4; ++i) { if (!pIter->input[i].stop && pIter->input[i].pRow != NULL) { TSDBKEY key = TSDBROW_KEY(pIter->input[i].pRow); @@ -881,13 +1059,13 @@ static int32_t nextRowIterGet(CacheNextRowIter *pIter, TSDBROW **ppRow) { } // delete detection - TSDBROW *merge[3] = {0}; - int iMerge[3] = {-1, -1, -1}; + TSDBROW *merge[4] = {0}; + int iMerge[4] = {-1, -1, -1, -1}; int nMerge = 0; for (int i = 0; i < nMax; ++i) { - TSDBKEY maxKey = TSDBROW_KEY(max[i]); + TSDBKEY maxKey1 = TSDBROW_KEY(max[i]); - bool deleted = tsdbKeyDeleted(&maxKey, pIter->pSkyline, &pIter->iSkyline); + bool deleted = tsdbKeyDeleted(&maxKey1, pIter->pSkyline, &pIter->iSkyline); if (!deleted) { iMerge[nMerge] = iMax[i]; merge[nMerge++] = max[i]; @@ -923,7 +1101,7 @@ static int32_t mergeLastRow(tb_uid_t uid, STsdb *pTsdb, bool *dup, STSRow **ppRo TSKEY lastRowTs = TSKEY_MAX; CacheNextRowIter iter = {0}; - nextRowIterOpen(&iter, uid, pTsdb); + nextRowIterOpen(&iter, uid, pTsdb, pTSchema); do { TSDBROW *pRow = NULL; @@ -1020,7 +1198,7 @@ static int32_t mergeLast(tb_uid_t uid, STsdb *pTsdb, SArray **ppLastArray) { TSKEY lastRowTs = TSKEY_MAX; CacheNextRowIter iter = {0}; - nextRowIterOpen(&iter, uid, pTsdb); + nextRowIterOpen(&iter, uid, pTsdb, pTSchema); do { TSDBROW *pRow = NULL; diff --git a/source/dnode/vnode/src/tsdb/tsdbCommit.c b/source/dnode/vnode/src/tsdb/tsdbCommit.c index 6e25166203948abe314e2405af77597ee9032a38..020f3b0bc6dedf70efa45a835056c8dd41d9a158 100644 --- a/source/dnode/vnode/src/tsdb/tsdbCommit.c +++ b/source/dnode/vnode/src/tsdb/tsdbCommit.c @@ -20,6 +20,12 @@ typedef struct { STSchema *pTSchema; } SSkmInfo; +typedef struct { + int64_t suid; + int64_t uid; + TSDBROW row; +} SRowInfo; + typedef struct { STsdb *pTsdb; /* commit data */ @@ -29,6 +35,7 @@ typedef struct { int32_t minRow; int32_t maxRow; int8_t cmprAlg; + SArray *aTbDataP; STsdbFS fs; // -------------- TSKEY nextKey; // reset by each table commit @@ -38,15 +45,27 @@ typedef struct { // commit file data struct { SDataFReader *pReader; - SArray *aBlockIdx; // SArray - SMapData mBlock; // SMapData, read from reader - SBlockData bData; + // data + SArray *aBlockIdx; // SArray + int32_t iBlockIdx; + SBlockIdx *pBlockIdx; + SMapData mBlock; // SMapData + SBlockData bData; + // last + SArray *aBlockL; // SArray + int32_t iBlockL; + SBlockData bDatal; + int32_t iRow; + SRowInfo *pRowInfo; + SRowInfo rowInfo; } dReader; struct { SDataFWriter *pWriter; SArray *aBlockIdx; // SArray + SArray *aBlockL; // SArray SMapData mBlock; // SMapData SBlockData bData; + SBlockData bDatal; } dWriter; SSkmInfo skmTable; SSkmInfo skmRow; @@ -162,10 +181,10 @@ static int32_t tsdbCommitDelStart(SCommitter *pCommitter) { SDelFile *pDelFileR = pCommitter->fs.pDelFile; if (pDelFileR) { - code = tsdbDelFReaderOpen(&pCommitter->pDelFReader, pDelFileR, pTsdb, NULL); + code = tsdbDelFReaderOpen(&pCommitter->pDelFReader, pDelFileR, pTsdb); if (code) goto _err; - code = tsdbReadDelIdx(pCommitter->pDelFReader, pCommitter->aDelIdx, NULL); + code = tsdbReadDelIdx(pCommitter->pDelFReader, pCommitter->aDelIdx); if (code) goto _err; } @@ -202,7 +221,7 @@ static int32_t tsdbCommitTableDel(SCommitter *pCommitter, STbData *pTbData, SDel suid = pDelIdx->suid; uid = pDelIdx->uid; - code = tsdbReadDelData(pCommitter->pDelFReader, pDelIdx, pCommitter->aDelData, NULL); + code = tsdbReadDelData(pCommitter->pDelFReader, pDelIdx, pCommitter->aDelData); if (code) goto _err; } else { taosArrayClear(pCommitter->aDelData); @@ -222,7 +241,7 @@ static int32_t tsdbCommitTableDel(SCommitter *pCommitter, STbData *pTbData, SDel } // write - code = tsdbWriteDelData(pCommitter->pDelFWriter, pCommitter->aDelData, NULL, &delIdx); + code = tsdbWriteDelData(pCommitter->pDelFWriter, pCommitter->aDelData, &delIdx); if (code) goto _err; // put delIdx @@ -243,7 +262,7 @@ static int32_t tsdbCommitDelEnd(SCommitter *pCommitter) { int32_t code = 0; STsdb *pTsdb = pCommitter->pTsdb; - code = tsdbWriteDelIdx(pCommitter->pDelFWriter, pCommitter->aDelIdxN, NULL); + code = tsdbWriteDelIdx(pCommitter->pDelFWriter, pCommitter->aDelIdxN); if (code) goto _err; code = tsdbUpdateDelFileHdr(pCommitter->pDelFWriter); @@ -271,44 +290,181 @@ _err: return code; } +static int32_t tsdbCommitterUpdateTableSchema(SCommitter *pCommitter, int64_t suid, int64_t uid) { + int32_t code = 0; + + if (suid) { + if (pCommitter->skmTable.suid == suid) goto _exit; + } else { + if (pCommitter->skmTable.uid == uid) goto _exit; + } + + pCommitter->skmTable.suid = suid; + pCommitter->skmTable.uid = uid; + tTSchemaDestroy(pCommitter->skmTable.pTSchema); + code = metaGetTbTSchemaEx(pCommitter->pTsdb->pVnode->pMeta, suid, uid, -1, &pCommitter->skmTable.pTSchema); + if (code) goto _exit; + +_exit: + return code; +} + +static int32_t tsdbCommitterUpdateRowSchema(SCommitter *pCommitter, int64_t suid, int64_t uid, int32_t sver) { + int32_t code = 0; + + if (pCommitter->skmRow.pTSchema) { + if (pCommitter->skmRow.suid == suid) { + if (suid == 0) { + if (pCommitter->skmRow.uid == uid && sver == pCommitter->skmRow.pTSchema->version) goto _exit; + } else { + if (sver == pCommitter->skmRow.pTSchema->version) goto _exit; + } + } + } + + pCommitter->skmRow.suid = suid; + pCommitter->skmRow.uid = uid; + tTSchemaDestroy(pCommitter->skmRow.pTSchema); + code = metaGetTbTSchemaEx(pCommitter->pTsdb->pVnode->pMeta, suid, uid, sver, &pCommitter->skmRow.pTSchema); + if (code) { + goto _exit; + } + +_exit: + return code; +} + +static int32_t tsdbCommitterNextLastRow(SCommitter *pCommitter) { + int32_t code = 0; + + ASSERT(pCommitter->dReader.pReader); + ASSERT(pCommitter->dReader.pRowInfo); + + SBlockData *pBlockDatal = &pCommitter->dReader.bDatal; + pCommitter->dReader.iRow++; + if (pCommitter->dReader.iRow < pBlockDatal->nRow) { + if (pBlockDatal->uid) { + pCommitter->dReader.pRowInfo->uid = pBlockDatal->uid; + } else { + pCommitter->dReader.pRowInfo->uid = pBlockDatal->aUid[pCommitter->dReader.iRow]; + } + pCommitter->dReader.pRowInfo->row = tsdbRowFromBlockData(pBlockDatal, pCommitter->dReader.iRow); + } else { + pCommitter->dReader.iBlockL++; + if (pCommitter->dReader.iBlockL < taosArrayGetSize(pCommitter->dReader.aBlockL)) { + SBlockL *pBlockL = (SBlockL *)taosArrayGet(pCommitter->dReader.aBlockL, pCommitter->dReader.iBlockL); + int64_t suid = pBlockL->suid; + int64_t uid = pBlockL->maxUid; + + code = tsdbCommitterUpdateTableSchema(pCommitter, suid, uid); + if (code) goto _exit; + + code = tBlockDataInit(pBlockDatal, suid, suid ? 0 : uid, pCommitter->skmTable.pTSchema); + if (code) goto _exit; + + code = tsdbReadLastBlock(pCommitter->dReader.pReader, pBlockL, pBlockDatal); + if (code) goto _exit; + + pCommitter->dReader.iRow = 0; + pCommitter->dReader.pRowInfo->suid = pBlockDatal->suid; + if (pBlockDatal->uid) { + pCommitter->dReader.pRowInfo->uid = pBlockDatal->uid; + } else { + pCommitter->dReader.pRowInfo->uid = pBlockDatal->aUid[0]; + } + pCommitter->dReader.pRowInfo->row = tsdbRowFromBlockData(pBlockDatal, pCommitter->dReader.iRow); + } else { + pCommitter->dReader.pRowInfo = NULL; + } + } + +_exit: + return code; +} + +static int32_t tsdbCommitterNextTableData(SCommitter *pCommitter) { + int32_t code = 0; + + ASSERT(pCommitter->dReader.pBlockIdx); + + pCommitter->dReader.iBlockIdx++; + if (pCommitter->dReader.iBlockIdx < taosArrayGetSize(pCommitter->dReader.aBlockIdx)) { + pCommitter->dReader.pBlockIdx = + (SBlockIdx *)taosArrayGet(pCommitter->dReader.aBlockIdx, pCommitter->dReader.iBlockIdx); + + code = tsdbReadBlock(pCommitter->dReader.pReader, pCommitter->dReader.pBlockIdx, &pCommitter->dReader.mBlock); + if (code) goto _exit; + + ASSERT(pCommitter->dReader.mBlock.nItem > 0); + } else { + pCommitter->dReader.pBlockIdx = NULL; + } + +_exit: + return code; +} + static int32_t tsdbCommitFileDataStart(SCommitter *pCommitter) { int32_t code = 0; STsdb *pTsdb = pCommitter->pTsdb; SDFileSet *pRSet = NULL; // memory + pCommitter->commitFid = tsdbKeyFid(pCommitter->nextKey, pCommitter->minutes, pCommitter->precision); + tsdbFidKeyRange(pCommitter->commitFid, pCommitter->minutes, pCommitter->precision, &pCommitter->minKey, + &pCommitter->maxKey); pCommitter->nextKey = TSKEY_MAX; - // old - taosArrayClear(pCommitter->dReader.aBlockIdx); - tMapDataReset(&pCommitter->dReader.mBlock); - tBlockDataReset(&pCommitter->dReader.bData); + // Reader pRSet = (SDFileSet *)taosArraySearch(pCommitter->fs.aDFileSet, &(SDFileSet){.fid = pCommitter->commitFid}, tDFileSetCmprFn, TD_EQ); if (pRSet) { code = tsdbDataFReaderOpen(&pCommitter->dReader.pReader, pTsdb, pRSet); if (code) goto _err; - code = tsdbReadBlockIdx(pCommitter->dReader.pReader, pCommitter->dReader.aBlockIdx, NULL); + // data + code = tsdbReadBlockIdx(pCommitter->dReader.pReader, pCommitter->dReader.aBlockIdx); if (code) goto _err; + + pCommitter->dReader.iBlockIdx = 0; + if (pCommitter->dReader.iBlockIdx < taosArrayGetSize(pCommitter->dReader.aBlockIdx)) { + pCommitter->dReader.pBlockIdx = + (SBlockIdx *)taosArrayGet(pCommitter->dReader.aBlockIdx, pCommitter->dReader.iBlockIdx); + + code = tsdbReadBlock(pCommitter->dReader.pReader, pCommitter->dReader.pBlockIdx, &pCommitter->dReader.mBlock); + if (code) goto _err; + } else { + pCommitter->dReader.pBlockIdx = NULL; + } + tBlockDataReset(&pCommitter->dReader.bData); + + // last + code = tsdbReadBlockL(pCommitter->dReader.pReader, pCommitter->dReader.aBlockL); + if (code) goto _err; + + pCommitter->dReader.iBlockL = -1; + pCommitter->dReader.iRow = -1; + pCommitter->dReader.pRowInfo = &pCommitter->dReader.rowInfo; + tBlockDataReset(&pCommitter->dReader.bDatal); + code = tsdbCommitterNextLastRow(pCommitter); + if (code) goto _err; + } else { + pCommitter->dReader.pBlockIdx = NULL; + pCommitter->dReader.pRowInfo = NULL; } - // new + // Writer SHeadFile fHead; SDataFile fData; SLastFile fLast; SSmaFile fSma; SDFileSet wSet = {.pHeadF = &fHead, .pDataF = &fData, .pLastF = &fLast, .pSmaF = &fSma}; - - taosArrayClear(pCommitter->dWriter.aBlockIdx); - tMapDataReset(&pCommitter->dWriter.mBlock); - tBlockDataReset(&pCommitter->dWriter.bData); if (pRSet) { wSet.diskId = pRSet->diskId; wSet.fid = pCommitter->commitFid; - fHead = (SHeadFile){.commitID = pCommitter->commitID, .offset = 0, .size = 0}; + fHead = (SHeadFile){.commitID = pCommitter->commitID, .size = 0, .offset = 0}; fData = *pRSet->pDataF; - fLast = (SLastFile){.commitID = pCommitter->commitID, .size = 0}; + fLast = (SLastFile){.commitID = pCommitter->commitID, .size = 0, .offset = 0}; fSma = *pRSet->pSmaF; } else { SDiskID did = {0}; @@ -319,14 +475,20 @@ static int32_t tsdbCommitFileDataStart(SCommitter *pCommitter) { wSet.diskId = did; wSet.fid = pCommitter->commitFid; - fHead = (SHeadFile){.commitID = pCommitter->commitID, .offset = 0, .size = 0}; + fHead = (SHeadFile){.commitID = pCommitter->commitID, .size = 0, .offset = 0}; fData = (SDataFile){.commitID = pCommitter->commitID, .size = 0}; - fLast = (SLastFile){.commitID = pCommitter->commitID, .size = 0}; + fLast = (SLastFile){.commitID = pCommitter->commitID, .size = 0, .offset = 0}; fSma = (SSmaFile){.commitID = pCommitter->commitID, .size = 0}; } code = tsdbDataFWriterOpen(&pCommitter->dWriter.pWriter, pTsdb, &wSet); if (code) goto _err; + taosArrayClear(pCommitter->dWriter.aBlockIdx); + taosArrayClear(pCommitter->dWriter.aBlockL); + tMapDataReset(&pCommitter->dWriter.mBlock); + tBlockDataReset(&pCommitter->dWriter.bData); + tBlockDataReset(&pCommitter->dWriter.bDatal); + _exit: return code; @@ -335,200 +497,192 @@ _err: return code; } -static int32_t tsdbCommitterUpdateTableSchema(SCommitter *pCommitter, int64_t suid, int64_t uid, int32_t sver) { - int32_t code = 0; - - if (pCommitter->skmTable.pTSchema) { - if (pCommitter->skmTable.suid == suid) { - if (suid == 0) { - if (pCommitter->skmTable.uid == uid && sver == pCommitter->skmTable.pTSchema->version) goto _exit; - } else { - if (sver == pCommitter->skmTable.pTSchema->version) goto _exit; - } - } - } +static int32_t tsdbCommitDataBlock(SCommitter *pCommitter, SBlock *pBlock) { + int32_t code = 0; + SBlockData *pBlockData = &pCommitter->dWriter.bData; + SBlock block; - pCommitter->skmTable.suid = suid; - pCommitter->skmTable.uid = uid; - tTSchemaDestroy(pCommitter->skmTable.pTSchema); - code = metaGetTbTSchemaEx(pCommitter->pTsdb->pVnode->pMeta, suid, uid, sver, &pCommitter->skmTable.pTSchema); - if (code) goto _exit; + ASSERT(pBlockData->nRow > 0); -_exit: - return code; -} + if (pBlock) { + block = *pBlock; // as a subblock + } else { + tBlockReset(&block); // as a new block + } -static int32_t tsdbCommitterUpdateRowSchema(SCommitter *pCommitter, int64_t suid, int64_t uid, int32_t sver) { - int32_t code = 0; + // info + block.nRow += pBlockData->nRow; + for (int32_t iRow = 0; iRow < pBlockData->nRow; iRow++) { + TSDBKEY key = {.ts = pBlockData->aTSKEY[iRow], .version = pBlockData->aVersion[iRow]}; - if (pCommitter->skmRow.pTSchema) { - if (pCommitter->skmRow.suid == suid) { - if (suid == 0) { - if (pCommitter->skmRow.uid == uid && sver == pCommitter->skmRow.pTSchema->version) goto _exit; - } else { - if (sver == pCommitter->skmRow.pTSchema->version) goto _exit; + if (iRow == 0) { + if (tsdbKeyCmprFn(&block.minKey, &key) > 0) { + block.minKey = key; + } + } else { + if (pBlockData->aTSKEY[iRow] == pBlockData->aTSKEY[iRow - 1]) { + block.hasDup = 1; } } - } - pCommitter->skmRow.suid = suid; - pCommitter->skmRow.uid = uid; - tTSchemaDestroy(pCommitter->skmRow.pTSchema); - code = metaGetTbTSchemaEx(pCommitter->pTsdb->pVnode->pMeta, suid, uid, sver, &pCommitter->skmRow.pTSchema); - if (code) { - goto _exit; + if (iRow == pBlockData->nRow - 1 && tsdbKeyCmprFn(&block.maxKey, &key) < 0) { + block.maxKey = key; + } + + block.minVer = TMIN(block.minVer, key.version); + block.maxVer = TMAX(block.maxVer, key.version); } -_exit: + // write + block.nSubBlock++; + code = tsdbWriteBlockData(pCommitter->dWriter.pWriter, pBlockData, &block.aSubBlock[block.nSubBlock - 1], + ((block.nSubBlock == 1) && !block.hasDup) ? &block.smaInfo : NULL, pCommitter->cmprAlg, 0); + if (code) goto _err; + + // put SBlock + code = tMapDataPutItem(&pCommitter->dWriter.mBlock, &block, tPutBlock); + if (code) goto _err; + + // clear + tBlockDataClear(pBlockData); + return code; -} -static int32_t tsdbCommitBlockData(SCommitter *pCommitter, SBlockData *pBlockData, SBlock *pBlock, SBlockIdx *pBlockIdx, - int8_t toDataOnly) { - int32_t code = 0; +_err: + tsdbError("vgId:%d tsdb commit data block failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); + return code; +} - if (pBlock->nSubBlock == 0) { - if (!toDataOnly && pBlockData->nRow < pCommitter->minRow) { - pBlock->last = 1; - } else { - pBlock->last = 0; - } +static int32_t tsdbCommitLastBlock(SCommitter *pCommitter) { + int32_t code = 0; + SBlockL blockL; + SBlockData *pBlockData = &pCommitter->dWriter.bDatal; + + ASSERT(pBlockData->nRow > 0); + + // info + blockL.suid = pBlockData->suid; + blockL.nRow = pBlockData->nRow; + blockL.minKey = TSKEY_MAX; + blockL.maxKey = TSKEY_MIN; + blockL.minVer = VERSION_MAX; + blockL.maxVer = VERSION_MIN; + for (int32_t iRow = 0; iRow < pBlockData->nRow; iRow++) { + blockL.minKey = TMIN(blockL.minKey, pBlockData->aTSKEY[iRow]); + blockL.maxKey = TMAX(blockL.maxKey, pBlockData->aTSKEY[iRow]); + blockL.minVer = TMIN(blockL.minVer, pBlockData->aVersion[iRow]); + blockL.maxVer = TMAX(blockL.maxVer, pBlockData->aVersion[iRow]); } + blockL.minUid = pBlockData->uid ? pBlockData->uid : pBlockData->aUid[0]; + blockL.maxUid = pBlockData->uid ? pBlockData->uid : pBlockData->aUid[pBlockData->nRow - 1]; - code = - tsdbWriteBlockData(pCommitter->dWriter.pWriter, pBlockData, NULL, NULL, pBlockIdx, pBlock, pCommitter->cmprAlg); + // write + code = tsdbWriteBlockData(pCommitter->dWriter.pWriter, pBlockData, &blockL.bInfo, NULL, pCommitter->cmprAlg, 1); if (code) goto _err; - code = tMapDataPutItem(&pCommitter->dWriter.mBlock, pBlock, tPutBlock); - if (code) goto _err; + // push SBlockL + if (taosArrayPush(pCommitter->dWriter.aBlockL, &blockL) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + + // clear + tBlockDataClear(pBlockData); return code; _err: + tsdbError("vgId:%d tsdb commit last block failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); return code; } -static int32_t tsdbMergeTableData(SCommitter *pCommitter, STbDataIter *pIter, SBlock *pBlockMerge, TSDBKEY toKey, - int8_t toDataOnly) { +static int32_t tsdbMergeCommitData(SCommitter *pCommitter, STbDataIter *pIter, SBlock *pBlock) { int32_t code = 0; - SBlockIdx *pBlockIdx = &(SBlockIdx){.suid = pIter->pTbData->suid, .uid = pIter->pTbData->uid}; - SBlockData *pBlockDataMerge = &pCommitter->dReader.bData; - SBlockData *pBlockData = &pCommitter->dWriter.bData; - SBlock block; - SBlock *pBlock = █ - TSDBROW *pRow1; - TSDBROW row2; - TSDBROW *pRow2 = &row2; + STbData *pTbData = pIter->pTbData; + SBlockData *pBlockDataR = &pCommitter->dReader.bData; + SBlockData *pBlockDataW = &pCommitter->dWriter.bData; - // read SBlockData - code = tsdbReadBlockData(pCommitter->dReader.pReader, pBlockIdx, pBlockMerge, pBlockDataMerge, NULL, NULL); + code = tsdbReadDataBlock(pCommitter->dReader.pReader, pBlock, pBlockDataR); if (code) goto _err; - code = tBlockDataSetSchema(pBlockData, pCommitter->skmTable.pTSchema); - if (code) goto _err; + tBlockDataClear(pBlockDataW); + int32_t iRow = 0; + TSDBROW row; + TSDBROW *pRow1 = tsdbTbDataIterGet(pIter); + TSDBROW *pRow2 = &row; + *pRow2 = tsdbRowFromBlockData(pBlockDataR, iRow); + while (pRow1 && pRow2) { + int32_t c = tsdbRowCmprFn(pRow1, pRow2); + + if (c < 0) { + code = tsdbCommitterUpdateRowSchema(pCommitter, pTbData->suid, pTbData->uid, TSDBROW_SVERSION(pRow1)); + if (code) goto _err; - // loop to merge - pRow1 = tsdbTbDataIterGet(pIter); - *pRow2 = tsdbRowFromBlockData(pBlockDataMerge, 0); - ASSERT(pRow1 && tsdbKeyCmprFn(&TSDBROW_KEY(pRow1), &toKey) < 0); - ASSERT(tsdbKeyCmprFn(&TSDBROW_KEY(pRow2), &toKey) < 0); - code = tsdbCommitterUpdateRowSchema(pCommitter, pBlockIdx->suid, pBlockIdx->uid, TSDBROW_SVERSION(pRow1)); - if (code) goto _err; + code = tBlockDataAppendRow(pBlockDataW, pRow1, pCommitter->skmRow.pTSchema, pTbData->uid); + if (code) goto _err; - tBlockReset(pBlock); - tBlockDataClearData(pBlockData); - while (true) { - if (pRow1 == NULL && pRow2 == NULL) { - if (pBlockData->nRow == 0) { - break; - } else { - goto _write_block; - } - } + // next + tsdbTbDataIterNext(pIter); + pRow1 = tsdbTbDataIterGet(pIter); + } else if (c > 0) { + code = tBlockDataAppendRow(pBlockDataW, pRow2, NULL, pTbData->uid); + if (code) goto _err; - if (pRow1 && pRow2) { - int32_t c = tsdbRowCmprFn(pRow1, pRow2); - if (c < 0) { - goto _append_mem_row; - } else if (c > 0) { - goto _append_block_row; + iRow++; + if (iRow < pBlockDataR->nRow) { + *pRow2 = tsdbRowFromBlockData(pBlockDataR, iRow); } else { - ASSERT(0); + pRow2 = NULL; } - } else if (pRow1) { - goto _append_mem_row; } else { - goto _append_block_row; - } - - _append_mem_row: - code = tBlockDataAppendRow(pBlockData, pRow1, pCommitter->skmRow.pTSchema); - if (code) goto _err; - - tsdbTbDataIterNext(pIter); - pRow1 = tsdbTbDataIterGet(pIter); - if (pRow1) { - if (tsdbKeyCmprFn(&TSDBROW_KEY(pRow1), &toKey) < 0) { - code = tsdbCommitterUpdateRowSchema(pCommitter, pBlockIdx->suid, pBlockIdx->uid, TSDBROW_SVERSION(pRow1)); - if (code) goto _err; - } else { - pRow1 = NULL; - } + ASSERT(0); } - if (pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { - goto _write_block; - } else { - continue; + // check + if (pBlockDataW->nRow >= pCommitter->maxRow * 4 / 5) { + code = tsdbCommitDataBlock(pCommitter, NULL); + if (code) goto _err; } + } - _append_block_row: - code = tBlockDataAppendRow(pBlockData, pRow2, NULL); + while (pRow2) { + code = tBlockDataAppendRow(pBlockDataW, pRow2, NULL, pTbData->uid); if (code) goto _err; - if (pRow2->iRow + 1 < pBlockDataMerge->nRow) { - *pRow2 = tsdbRowFromBlockData(pBlockDataMerge, pRow2->iRow + 1); + iRow++; + if (iRow < pBlockDataR->nRow) { + *pRow2 = tsdbRowFromBlockData(pBlockDataR, iRow); } else { pRow2 = NULL; } - if (pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { - goto _write_block; - } else { - continue; + // check + if (pBlockDataW->nRow >= pCommitter->maxRow * 4 / 5) { + code = tsdbCommitDataBlock(pCommitter, NULL); + if (code) goto _err; } + } - _write_block: - code = tsdbCommitBlockData(pCommitter, pBlockData, pBlock, pBlockIdx, toDataOnly); + // check + if (pBlockDataW->nRow > 0) { + code = tsdbCommitDataBlock(pCommitter, NULL); if (code) goto _err; - - tBlockReset(pBlock); - tBlockDataClearData(pBlockData); } return code; _err: - tsdbError("vgId:%d, tsdb merge block and mem failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); + tsdbError("vgId:%d, tsdb merge commit data failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); return code; } -static int32_t tsdbCommitTableMemData(SCommitter *pCommitter, STbDataIter *pIter, TSDBKEY toKey, int8_t toDataOnly) { +static int32_t tsdbCommitTableMemData(SCommitter *pCommitter, STbDataIter *pIter, TSDBKEY toKey) { int32_t code = 0; - TSDBROW *pRow; - SBlock block; - SBlock *pBlock = █ + STbData *pTbData = pIter->pTbData; SBlockData *pBlockData = &pCommitter->dWriter.bData; - int64_t suid = pIter->pTbData->suid; - int64_t uid = pIter->pTbData->uid; - - code = tBlockDataSetSchema(pBlockData, pCommitter->skmTable.pTSchema); - if (code) goto _err; - tBlockReset(pBlock); - tBlockDataClearData(pBlockData); - pRow = tsdbTbDataIterGet(pIter); - ASSERT(pRow && tsdbKeyCmprFn(&TSDBROW_KEY(pRow), &toKey) < 0); + tBlockDataClear(pBlockData); + TSDBROW *pRow = tsdbTbDataIterGet(pIter); while (true) { if (pRow == NULL) { if (pBlockData->nRow > 0) { @@ -539,33 +693,27 @@ static int32_t tsdbCommitTableMemData(SCommitter *pCommitter, STbDataIter *pIter } // update schema - code = tsdbCommitterUpdateRowSchema(pCommitter, suid, uid, TSDBROW_SVERSION(pRow)); + code = tsdbCommitterUpdateRowSchema(pCommitter, pTbData->suid, pTbData->uid, TSDBROW_SVERSION(pRow)); if (code) goto _err; // append - code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema); + code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema, pTbData->uid); if (code) goto _err; tsdbTbDataIterNext(pIter); pRow = tsdbTbDataIterGet(pIter); - // if (pRow && tsdbKeyCmprFn(&TSDBROW_KEY(pRow), &toKey) >= 0) pRow = NULL; - // crash on CI, use the block following if (pRow) { - TSDBKEY tmpKey = TSDBROW_KEY(pRow); - if (tsdbKeyCmprFn(&tmpKey, &toKey) >= 0) { + TSDBKEY rowKey = TSDBROW_KEY(pRow); + if (tsdbKeyCmprFn(&rowKey, &toKey) >= 0) { pRow = NULL; } } - if (pBlockData->nRow >= pCommitter->maxRow * 4 / 5) goto _write_block; - continue; - - _write_block: - code = tsdbCommitBlockData(pCommitter, pBlockData, pBlock, &(SBlockIdx){.suid = suid, .uid = uid}, toDataOnly); - if (code) goto _err; - - tBlockReset(pBlock); - tBlockDataClearData(pBlockData); + if (pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { + _write_block: + code = tsdbCommitDataBlock(pCommitter, NULL); + if (code) goto _err; + } } return code; @@ -575,65 +723,16 @@ _err: return code; } -static int32_t tsdbCommitTableDiskData(SCommitter *pCommitter, SBlock *pBlock, SBlockIdx *pBlockIdx) { - int32_t code = 0; - SBlock block; - - if (pBlock->last) { - code = tsdbReadBlockData(pCommitter->dReader.pReader, pBlockIdx, pBlock, &pCommitter->dReader.bData, NULL, NULL); - if (code) goto _err; - - tBlockReset(&block); - code = tsdbCommitBlockData(pCommitter, &pCommitter->dReader.bData, &block, pBlockIdx, 0); - if (code) goto _err; - } else { - code = tMapDataPutItem(&pCommitter->dWriter.mBlock, pBlock, tPutBlock); - if (code) goto _err; - } - - return code; - -_err: - tsdbError("vgId:%d, tsdb commit table disk data failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); - return code; -} - -static int32_t tsdbCommitTableDataEnd(SCommitter *pCommitter, int64_t suid, int64_t uid) { - int32_t code = 0; - SBlockIdx blockIdx = {.suid = suid, .uid = uid}; - SBlockIdx *pBlockIdx = &blockIdx; - - code = tsdbWriteBlock(pCommitter->dWriter.pWriter, &pCommitter->dWriter.mBlock, NULL, pBlockIdx); - if (code) goto _err; - - if (taosArrayPush(pCommitter->dWriter.aBlockIdx, pBlockIdx) == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - - return code; - -_err: - tsdbError("vgId:%d, commit table data end failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); - return code; -} +static int32_t tsdbGetNumOfRowsLessThan(STbDataIter *pIter, TSDBKEY key) { + int32_t nRow = 0; -static int32_t tsdbGetOvlpNRow(STbDataIter *pIter, SBlock *pBlock) { - int32_t nRow = 0; - TSDBROW *pRow; - TSDBKEY key; - int32_t c = 0; STbDataIter iter = *pIter; - - iter.pRow = NULL; while (true) { - pRow = tsdbTbDataIterGet(&iter); - + TSDBROW *pRow = tsdbTbDataIterGet(&iter); if (pRow == NULL) break; - key = TSDBROW_KEY(pRow); - c = tBlockCmprFn(&(SBlock){.maxKey = key, .minKey = key}, pBlock); - if (c == 0) { + int32_t c = tsdbKeyCmprFn(&TSDBROW_KEY(pRow), &key); + if (c < 0) { nRow++; tsdbTbDataIterNext(&iter); } else if (c > 0) { @@ -648,42 +747,33 @@ static int32_t tsdbGetOvlpNRow(STbDataIter *pIter, SBlock *pBlock) { static int32_t tsdbMergeAsSubBlock(SCommitter *pCommitter, STbDataIter *pIter, SBlock *pBlock) { int32_t code = 0; + STbData *pTbData = pIter->pTbData; SBlockData *pBlockData = &pCommitter->dWriter.bData; - SBlockIdx *pBlockIdx = &(SBlockIdx){.suid = pIter->pTbData->suid, .uid = pIter->pTbData->uid}; - SBlock block; - TSDBROW *pRow; - code = tBlockDataSetSchema(pBlockData, pCommitter->skmTable.pTSchema); - if (code) goto _err; - - pRow = tsdbTbDataIterGet(pIter); - code = tsdbCommitterUpdateRowSchema(pCommitter, pBlockIdx->suid, pBlockIdx->uid, TSDBROW_SVERSION(pRow)); - if (code) goto _err; + tBlockDataClear(pBlockData); + TSDBROW *pRow = tsdbTbDataIterGet(pIter); while (true) { if (pRow == NULL) break; - code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema); + + code = tsdbCommitterUpdateRowSchema(pCommitter, pTbData->suid, pTbData->uid, TSDBROW_SVERSION(pRow)); + if (code) goto _err; + + code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema, pTbData->uid); if (code) goto _err; tsdbTbDataIterNext(pIter); pRow = tsdbTbDataIterGet(pIter); if (pRow) { - TSDBKEY key = TSDBROW_KEY(pRow); - int32_t c = tBlockCmprFn(&(SBlock){.minKey = key, .maxKey = key}, pBlock); - - if (c == 0) { - code = - tsdbCommitterUpdateRowSchema(pCommitter, pIter->pTbData->suid, pIter->pTbData->uid, TSDBROW_SVERSION(pRow)); - if (code) goto _err; - } else if (c > 0) { + TSDBKEY rowKey = TSDBROW_KEY(pRow); + if (tsdbKeyCmprFn(&rowKey, &pBlock->maxKey) > 0) { pRow = NULL; - } else { - ASSERT(0); } } } - block = *pBlock; - code = tsdbCommitBlockData(pCommitter, pBlockData, &block, pBlockIdx, 0); + ASSERT(pBlockData->nRow > 0 && pBlock->nRow + pBlockData->nRow <= pCommitter->maxRow); + + code = tsdbCommitDataBlock(pCommitter, pBlock); if (code) goto _err; return code; @@ -693,176 +783,307 @@ _err: return code; } -static int32_t tsdbCommitTableData(SCommitter *pCommitter, STbData *pTbData, SBlockIdx *pBlockIdx) { - int32_t code = 0; - STbDataIter iter = {0}; - STbDataIter *pIter = &iter; - TSDBROW *pRow; - int32_t iBlock; - int32_t nBlock; - int64_t suid; - int64_t uid; +static int32_t tsdbMergeCommitLast(SCommitter *pCommitter, STbDataIter *pIter) { + int32_t code = 0; + STbData *pTbData = pIter->pTbData; + int32_t nRow = tsdbGetNumOfRowsLessThan(pIter, (TSDBKEY){.ts = pCommitter->maxKey + 1, .version = VERSION_MIN}); - if (pTbData) { - tsdbTbDataIterOpen(pTbData, &(TSDBKEY){.ts = pCommitter->minKey, .version = VERSION_MIN}, 0, pIter); - pRow = tsdbTbDataIterGet(pIter); - if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) pRow = NULL; + if (pCommitter->dReader.pRowInfo && tTABLEIDCmprFn(pTbData, pCommitter->dReader.pRowInfo) == 0) { + if (pCommitter->dReader.pRowInfo->suid) { // super table + for (int32_t iRow = pCommitter->dReader.iRow; iRow < pCommitter->dReader.bDatal.nRow; iRow++) { + if (pTbData->uid != pCommitter->dReader.bDatal.aUid[iRow]) break; + nRow++; + } + } else { // normal table + ASSERT(pCommitter->dReader.iRow == 0); + nRow += pCommitter->dReader.bDatal.nRow; + } + } - suid = pTbData->suid; - uid = pTbData->uid; - } else { - pIter = NULL; + if (nRow == 0) goto _exit; + + TSDBROW *pRow = tsdbTbDataIterGet(pIter); + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { pRow = NULL; } - if (pBlockIdx) { - code = tsdbReadBlock(pCommitter->dReader.pReader, pBlockIdx, &pCommitter->dReader.mBlock, NULL); - if (code) goto _err; + SRowInfo *pRowInfo = pCommitter->dReader.pRowInfo; + if (pRowInfo && pRowInfo->uid != pTbData->uid) { + pRowInfo = NULL; + } - nBlock = pCommitter->dReader.mBlock.nItem; - ASSERT(nBlock > 0); + while (nRow) { + SBlockData *pBlockData; + int8_t toData; - suid = pBlockIdx->suid; - uid = pBlockIdx->uid; - } else { - nBlock = 0; - } + if (nRow < pCommitter->minRow) { // to .last + toData = 0; + pBlockData = &pCommitter->dWriter.bDatal; - if (pRow == NULL && nBlock == 0) goto _exit; + // commit and reset block data schema if need + // QUESTION: Is there a case that pBlockData->nRow == 0 but need to change schema ? + if (pBlockData->suid || pBlockData->uid) { + if (pBlockData->suid != pTbData->suid || pBlockData->suid == 0) { + if (pBlockData->nRow > 0) { + code = tsdbCommitLastBlock(pCommitter); + if (code) goto _err; + } - // start =========== - tMapDataReset(&pCommitter->dWriter.mBlock); - SBlock block; - SBlock *pBlock = █ + tBlockDataReset(pBlockData); + } + } - iBlock = 0; - if (iBlock < nBlock) { - tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); - } else { - pBlock = NULL; - } + // set block data schema if need + if (pBlockData->suid == 0 && pBlockData->uid == 0) { + code = + tBlockDataInit(pBlockData, pTbData->suid, pTbData->suid ? 0 : pTbData->uid, pCommitter->skmTable.pTSchema); + if (code) goto _err; + } - if (pRow) { - code = tsdbCommitterUpdateTableSchema(pCommitter, pTbData->suid, pTbData->uid, pTbData->maxSkmVer); - if (code) goto _err; - } + if (pBlockData->nRow + nRow > pCommitter->maxRow) { + code = tsdbCommitLastBlock(pCommitter); + if (code) goto _err; + } + } else { // to .data + toData = 1; + pBlockData = &pCommitter->dWriter.bData; + ASSERT(pBlockData->nRow == 0); + } - // merge =========== - while (true) { - if (pRow == NULL && pBlock == NULL) break; + while (pRow && pRowInfo) { + int32_t c = tsdbRowCmprFn(pRow, &pRowInfo->row); + if (c < 0) { + code = tsdbCommitterUpdateRowSchema(pCommitter, pTbData->suid, pTbData->uid, TSDBROW_SVERSION(pRow)); + if (code) goto _err; - if (pRow && pBlock) { - if (pBlock->last) { - code = tsdbMergeTableData(pCommitter, pIter, pBlock, - (TSDBKEY){.ts = pCommitter->maxKey + 1, .version = VERSION_MIN}, 0); + code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema, pTbData->uid); if (code) goto _err; + tsdbTbDataIterNext(pIter); pRow = tsdbTbDataIterGet(pIter); - if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) pRow = NULL; - iBlock++; - if (iBlock < nBlock) { - tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); - } else { - pBlock = NULL; + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { + pRow = NULL; } + } else if (c > 0) { + code = tBlockDataAppendRow(pBlockData, &pRowInfo->row, NULL, pTbData->uid); + if (code) goto _err; + + code = tsdbCommitterNextLastRow(pCommitter); + if (code) goto _err; - ASSERT(pRow == NULL && pBlock == NULL); + pRowInfo = pCommitter->dReader.pRowInfo; + if (pRowInfo && pRowInfo->uid != pTbData->uid) { + pRowInfo = NULL; + } } else { - int32_t c = tBlockCmprFn(&(SBlock){.maxKey = TSDBROW_KEY(pRow), .minKey = TSDBROW_KEY(pRow)}, pBlock); - if (c > 0) { - // only disk data - code = tsdbCommitTableDiskData(pCommitter, pBlock, pBlockIdx); - if (code) goto _err; + ASSERT(0); + } - iBlock++; - if (iBlock < nBlock) { - tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); - } else { - pBlock = NULL; - } - } else if (c < 0) { - // only memory data - code = tsdbCommitTableMemData(pCommitter, pIter, pBlock->minKey, 1); + nRow--; + if (toData) { + if (nRow == 0 || pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { + code = tsdbCommitDataBlock(pCommitter, NULL); if (code) goto _err; + goto _outer_break; + } + } + } - pRow = tsdbTbDataIterGet(pIter); - if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) pRow = NULL; - } else { - // merge memory and disk - int32_t nOvlp = tsdbGetOvlpNRow(pIter, pBlock); - ASSERT(nOvlp); - if (pBlock->nRow + nOvlp <= pCommitter->maxRow && pBlock->nSubBlock < TSDB_MAX_SUBBLOCKS) { - code = tsdbMergeAsSubBlock(pCommitter, pIter, pBlock); - if (code) goto _err; - } else { - TSDBKEY toKey = {.ts = pCommitter->maxKey + 1, .version = VERSION_MIN}; - int8_t toDataOnly = 0; + while (pRow) { + code = tsdbCommitterUpdateRowSchema(pCommitter, pTbData->suid, pTbData->uid, TSDBROW_SVERSION(pRow)); + if (code) goto _err; - if (iBlock < nBlock - 1) { - toDataOnly = 1; + code = tBlockDataAppendRow(pBlockData, pRow, pCommitter->skmRow.pTSchema, pTbData->uid); + if (code) goto _err; - SBlock nextBlock = {0}; - tBlockReset(&nextBlock); - tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock + 1, &nextBlock, tGetBlock); - toKey = nextBlock.minKey; - } + tsdbTbDataIterNext(pIter); + pRow = tsdbTbDataIterGet(pIter); + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { + pRow = NULL; + } - code = tsdbMergeTableData(pCommitter, pIter, pBlock, toKey, toDataOnly); - if (code) goto _err; - } + nRow--; + if (toData) { + if (nRow == 0 || pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { + code = tsdbCommitDataBlock(pCommitter, NULL); + if (code) goto _err; + goto _outer_break; + } + } + } - pRow = tsdbTbDataIterGet(pIter); - if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) pRow = NULL; - iBlock++; - if (iBlock < nBlock) { - tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); - } else { - pBlock = NULL; - } + while (pRowInfo) { + code = tBlockDataAppendRow(pBlockData, &pRowInfo->row, NULL, pTbData->uid); + if (code) goto _err; + + code = tsdbCommitterNextLastRow(pCommitter); + if (code) goto _err; + + pRowInfo = pCommitter->dReader.pRowInfo; + if (pRowInfo && pRowInfo->uid != pTbData->uid) { + pRowInfo = NULL; + } + + nRow--; + if (toData) { + if (nRow == 0 || pBlockData->nRow >= pCommitter->maxRow * 4 / 5) { + code = tsdbCommitDataBlock(pCommitter, NULL); + if (code) goto _err; + goto _outer_break; } } - } else if (pBlock) { - code = tsdbCommitTableDiskData(pCommitter, pBlock, pBlockIdx); + } + + _outer_break: + ASSERT(nRow >= 0); + } + +_exit: + return code; + +_err: + tsdbError("vgId:%d tsdb merge commit last failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); + return code; +} + +static int32_t tsdbCommitTableData(SCommitter *pCommitter, STbData *pTbData) { + int32_t code = 0; + + ASSERT(pCommitter->dReader.pBlockIdx == NULL || tTABLEIDCmprFn(pCommitter->dReader.pBlockIdx, pTbData) >= 0); + ASSERT(pCommitter->dReader.pRowInfo == NULL || tTABLEIDCmprFn(pCommitter->dReader.pRowInfo, pTbData) >= 0); + + // merge commit table data + STbDataIter iter = {0}; + STbDataIter *pIter = &iter; + TSDBROW *pRow; + + tsdbTbDataIterOpen(pTbData, &(TSDBKEY){.ts = pCommitter->minKey, .version = VERSION_MIN}, 0, pIter); + pRow = tsdbTbDataIterGet(pIter); + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { + pRow = NULL; + } + + if (pRow == NULL) goto _exit; + + int32_t iBlock = 0; + SBlock block; + SBlock *pBlock = █ + if (pCommitter->dReader.pBlockIdx && tTABLEIDCmprFn(pTbData, pCommitter->dReader.pBlockIdx) == 0) { + tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); + } else { + pBlock = NULL; + } + + code = tsdbCommitterUpdateTableSchema(pCommitter, pTbData->suid, pTbData->uid); + if (code) goto _err; + + tMapDataReset(&pCommitter->dWriter.mBlock); + code = tBlockDataInit(&pCommitter->dReader.bData, pTbData->suid, pTbData->uid, pCommitter->skmTable.pTSchema); + if (code) goto _err; + code = tBlockDataInit(&pCommitter->dWriter.bData, pTbData->suid, pTbData->uid, pCommitter->skmTable.pTSchema); + if (code) goto _err; + + // .data merge + while (pBlock && pRow) { + int32_t c = tBlockCmprFn(pBlock, &(SBlock){.minKey = TSDBROW_KEY(pRow), .maxKey = TSDBROW_KEY(pRow)}); + if (c < 0) { // disk + code = tMapDataPutItem(&pCommitter->dWriter.mBlock, pBlock, tPutBlock); if (code) goto _err; + // next iBlock++; - if (iBlock < nBlock) { + if (iBlock < pCommitter->dReader.mBlock.nItem) { tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); } else { pBlock = NULL; } - } else { - code = - tsdbCommitTableMemData(pCommitter, pIter, (TSDBKEY){.ts = pCommitter->maxKey + 1, .version = VERSION_MIN}, 0); + } else if (c > 0) { // memory + code = tsdbCommitTableMemData(pCommitter, pIter, pBlock->minKey); if (code) goto _err; + // next + pRow = tsdbTbDataIterGet(pIter); + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { + pRow = NULL; + } + } else { // merge + int32_t nOvlp = tsdbGetNumOfRowsLessThan(pIter, pBlock->maxKey); + + ASSERT(nOvlp > 0); + + if (pBlock->nRow + nOvlp <= pCommitter->maxRow && pBlock->nSubBlock < TSDB_MAX_SUBBLOCKS) { + code = tsdbMergeAsSubBlock(pCommitter, pIter, pBlock); + if (code) goto _err; + } else { + code = tsdbMergeCommitData(pCommitter, pIter, pBlock); + if (code) goto _err; + } + + // next pRow = tsdbTbDataIterGet(pIter); - if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) pRow = NULL; - ASSERT(pRow == NULL); + if (pRow && TSDBROW_TS(pRow) > pCommitter->maxKey) { + pRow = NULL; + } + iBlock++; + if (iBlock < pCommitter->dReader.mBlock.nItem) { + tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); + } else { + pBlock = NULL; + } + } + } + + while (pBlock) { + code = tMapDataPutItem(&pCommitter->dWriter.mBlock, pBlock, tPutBlock); + if (code) goto _err; + + // next + iBlock++; + if (iBlock < pCommitter->dReader.mBlock.nItem) { + tMapDataGetItemByIdx(&pCommitter->dReader.mBlock, iBlock, pBlock, tGetBlock); + } else { + pBlock = NULL; } } - // end ===================== - code = tsdbCommitTableDataEnd(pCommitter, suid, uid); + // .data append and .last merge + code = tsdbMergeCommitLast(pCommitter, pIter); if (code) goto _err; + // end + if (pCommitter->dWriter.mBlock.nItem > 0) { + SBlockIdx blockIdx = {.suid = pTbData->suid, .uid = pTbData->uid}; + code = tsdbWriteBlock(pCommitter->dWriter.pWriter, &pCommitter->dWriter.mBlock, &blockIdx); + if (code) goto _err; + + if (taosArrayPush(pCommitter->dWriter.aBlockIdx, &blockIdx) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + } + _exit: - if (pIter) { - pRow = tsdbTbDataIterGet(pIter); - if (pRow) pCommitter->nextKey = TMIN(pCommitter->nextKey, TSDBROW_TS(pRow)); + pRow = tsdbTbDataIterGet(pIter); + if (pRow) { + pCommitter->nextKey = TMIN(pCommitter->nextKey, TSDBROW_TS(pRow)); } + return code; _err: - tsdbError("vgId:%d, tsdb commit table data failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); + tsdbError("vgId:%d tsdb commit table data failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); return code; } static int32_t tsdbCommitFileDataEnd(SCommitter *pCommitter) { int32_t code = 0; - // write blockIdx - code = tsdbWriteBlockIdx(pCommitter->dWriter.pWriter, pCommitter->dWriter.aBlockIdx, NULL); + // write aBlockIdx + code = tsdbWriteBlockIdx(pCommitter->dWriter.pWriter, pCommitter->dWriter.aBlockIdx); + if (code) goto _err; + + // write aBlockL + code = tsdbWriteBlockL(pCommitter->dWriter.pWriter, pCommitter->dWriter.aBlockL); if (code) goto _err; // update file header @@ -890,6 +1111,98 @@ _err: return code; } +static int32_t tsdbMoveCommitData(SCommitter *pCommitter, TABLEID toTable) { + int32_t code = 0; + + // .data + while (true) { + if (pCommitter->dReader.pBlockIdx == NULL || tTABLEIDCmprFn(pCommitter->dReader.pBlockIdx, &toTable) >= 0) break; + + SBlockIdx blockIdx = *pCommitter->dReader.pBlockIdx; + code = tsdbWriteBlock(pCommitter->dWriter.pWriter, &pCommitter->dReader.mBlock, &blockIdx); + if (code) goto _err; + + if (taosArrayPush(pCommitter->dWriter.aBlockIdx, &blockIdx) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + + code = tsdbCommitterNextTableData(pCommitter); + if (code) goto _err; + } + + // .last + while (true) { + if (pCommitter->dReader.pRowInfo == NULL || tTABLEIDCmprFn(pCommitter->dReader.pRowInfo, &toTable) >= 0) break; + + SBlockData *pBlockDataR = &pCommitter->dReader.bDatal; + SBlockData *pBlockDataW = &pCommitter->dWriter.bDatal; + tb_uid_t suid = pCommitter->dReader.pRowInfo->suid; + tb_uid_t uid = pCommitter->dReader.pRowInfo->uid; + + ASSERT((pBlockDataR->suid && !pBlockDataR->uid) || (!pBlockDataR->suid && pBlockDataR->uid)); + ASSERT(pBlockDataR->nRow > 0); + + // commit and reset block data schema if need + if (pBlockDataW->suid || pBlockDataW->uid) { + if (pBlockDataW->suid != suid || pBlockDataW->suid == 0) { + if (pBlockDataW->nRow > 0) { + code = tsdbCommitLastBlock(pCommitter); + if (code) goto _err; + } + tBlockDataReset(pBlockDataW); + } + } + + // set block data schema if need + if (pBlockDataW->suid == 0 && pBlockDataW->uid == 0) { + code = tsdbCommitterUpdateTableSchema(pCommitter, suid, uid); + if (code) goto _err; + + code = tBlockDataInit(pBlockDataW, suid, suid ? 0 : uid, pCommitter->skmTable.pTSchema); + if (code) goto _err; + } + + // check if it can make sure that one table data in one block + int32_t nRow = 0; + if (pBlockDataR->suid) { + int32_t iRow = pCommitter->dReader.iRow; + while ((iRow < pBlockDataR->nRow) && (pBlockDataR->aUid[iRow] == uid)) { + nRow++; + iRow++; + } + } else { + ASSERT(pCommitter->dReader.iRow == 0); + nRow = pBlockDataR->nRow; + } + + ASSERT(nRow > 0 && nRow < pCommitter->minRow); + + if (pBlockDataW->nRow + nRow > pCommitter->maxRow) { + ASSERT(pBlockDataW->nRow > 0); + + code = tsdbCommitLastBlock(pCommitter); + if (code) goto _err; + } + + while (nRow > 0) { + code = tBlockDataAppendRow(pBlockDataW, &pCommitter->dReader.pRowInfo->row, NULL, uid); + if (code) goto _err; + + code = tsdbCommitterNextLastRow(pCommitter); + if (code) goto _err; + + nRow--; + } + } + + return code; + +_err: + tsdbError("vgId:%d tsdb move commit data failed since %s", TD_VID(pCommitter->pTsdb->pVnode), tstrerror(code)); + return code; +} + static int32_t tsdbCommitFileData(SCommitter *pCommitter) { int32_t code = 0; STsdb *pTsdb = pCommitter->pTsdb; @@ -900,59 +1213,30 @@ static int32_t tsdbCommitFileData(SCommitter *pCommitter) { if (code) goto _err; // commit file data impl - int32_t iTbData = 0; - int32_t nTbData = taosArrayGetSize(pMemTable->aTbData); - int32_t iBlockIdx = 0; - int32_t nBlockIdx = taosArrayGetSize(pCommitter->dReader.aBlockIdx); - STbData *pTbData; - SBlockIdx *pBlockIdx; - - ASSERT(nTbData > 0); - - pTbData = (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData); - pBlockIdx = (iBlockIdx < nBlockIdx) ? (SBlockIdx *)taosArrayGet(pCommitter->dReader.aBlockIdx, iBlockIdx) : NULL; - while (pTbData || pBlockIdx) { - if (pTbData && pBlockIdx) { - int32_t c = tTABLEIDCmprFn(pTbData, pBlockIdx); - - if (c == 0) { - goto _commit_table_mem_and_disk; - } else if (c < 0) { - goto _commit_table_mem_data; - } else { - goto _commit_table_disk_data; - } - } else if (pBlockIdx) { - goto _commit_table_disk_data; - } else { - goto _commit_table_mem_data; - } + for (int32_t iTbData = 0; iTbData < taosArrayGetSize(pCommitter->aTbDataP); iTbData++) { + STbData *pTbData = (STbData *)taosArrayGetP(pCommitter->aTbDataP, iTbData); - _commit_table_mem_data: - code = tsdbCommitTableData(pCommitter, pTbData, NULL); + // move commit until current (suid, uid) + code = tsdbMoveCommitData(pCommitter, *(TABLEID *)pTbData); if (code) goto _err; - iTbData++; - pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData) : NULL; - continue; - - _commit_table_disk_data: - code = tsdbCommitTableData(pCommitter, NULL, pBlockIdx); + // commit current table data + code = tsdbCommitTableData(pCommitter, pTbData); if (code) goto _err; - iBlockIdx++; - pBlockIdx = (iBlockIdx < nBlockIdx) ? (SBlockIdx *)taosArrayGet(pCommitter->dReader.aBlockIdx, iBlockIdx) : NULL; - continue; + // move next reader table data if need + if (pCommitter->dReader.pBlockIdx && tTABLEIDCmprFn(pTbData, pCommitter->dReader.pBlockIdx) == 0) { + code = tsdbCommitterNextTableData(pCommitter); + if (code) goto _err; + } + } - _commit_table_mem_and_disk: - code = tsdbCommitTableData(pCommitter, pTbData, pBlockIdx); - if (code) goto _err; + code = tsdbMoveCommitData(pCommitter, (TABLEID){.suid = INT64_MAX, .uid = INT64_MAX}); + if (code) goto _err; - iBlockIdx++; - pBlockIdx = (iBlockIdx < nBlockIdx) ? (SBlockIdx *)taosArrayGet(pCommitter->dReader.aBlockIdx, iBlockIdx) : NULL; - iTbData++; - pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData) : NULL; - continue; + if (pCommitter->dWriter.bDatal.nRow > 0) { + code = tsdbCommitLastBlock(pCommitter); + if (code) goto _err; } // commit file data end @@ -987,6 +1271,11 @@ static int32_t tsdbStartCommit(STsdb *pTsdb, SCommitter *pCommitter) { pCommitter->minRow = pTsdb->pVnode->config.tsdbCfg.minRows; pCommitter->maxRow = pTsdb->pVnode->config.tsdbCfg.maxRows; pCommitter->cmprAlg = pTsdb->pVnode->config.tsdbCfg.compression; + pCommitter->aTbDataP = tsdbMemTableGetTbDataArray(pTsdb->imem); + if (pCommitter->aTbDataP == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } code = tsdbFSCopy(pTsdb, &pCommitter->fs); if (code) goto _err; @@ -1001,22 +1290,42 @@ _err: static int32_t tsdbCommitDataStart(SCommitter *pCommitter) { int32_t code = 0; + // Reader pCommitter->dReader.aBlockIdx = taosArrayInit(0, sizeof(SBlockIdx)); if (pCommitter->dReader.aBlockIdx == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } + code = tBlockDataCreate(&pCommitter->dReader.bData); + if (code) goto _exit; + + pCommitter->dReader.aBlockL = taosArrayInit(0, sizeof(SBlockL)); + if (pCommitter->dReader.aBlockL == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _exit; + } + + code = tBlockDataCreate(&pCommitter->dReader.bDatal); + if (code) goto _exit; + + // Writer pCommitter->dWriter.aBlockIdx = taosArrayInit(0, sizeof(SBlockIdx)); if (pCommitter->dWriter.aBlockIdx == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } - code = tBlockDataInit(&pCommitter->dReader.bData); + pCommitter->dWriter.aBlockL = taosArrayInit(0, sizeof(SBlockL)); + if (pCommitter->dWriter.aBlockL == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _exit; + } + + code = tBlockDataCreate(&pCommitter->dWriter.bData); if (code) goto _exit; - code = tBlockDataInit(&pCommitter->dWriter.bData); + code = tBlockDataCreate(&pCommitter->dWriter.bDatal); if (code) goto _exit; _exit: @@ -1024,12 +1333,19 @@ _exit: } static void tsdbCommitDataEnd(SCommitter *pCommitter) { + // Reader taosArrayDestroy(pCommitter->dReader.aBlockIdx); tMapDataClear(&pCommitter->dReader.mBlock); - tBlockDataClear(&pCommitter->dReader.bData, 1); + tBlockDataDestroy(&pCommitter->dReader.bData, 1); + taosArrayDestroy(pCommitter->dReader.aBlockL); + tBlockDataDestroy(&pCommitter->dReader.bDatal, 1); + + // Writer taosArrayDestroy(pCommitter->dWriter.aBlockIdx); + taosArrayDestroy(pCommitter->dWriter.aBlockL); tMapDataClear(&pCommitter->dWriter.mBlock); - tBlockDataClear(&pCommitter->dWriter.bData, 1); + tBlockDataDestroy(&pCommitter->dWriter.bData, 1); + tBlockDataDestroy(&pCommitter->dWriter.bDatal, 1); tTSchemaDestroy(pCommitter->skmTable.pTSchema); tTSchemaDestroy(pCommitter->skmRow.pTSchema); } @@ -1049,9 +1365,6 @@ static int32_t tsdbCommitData(SCommitter *pCommitter) { // impl ==================== pCommitter->nextKey = pMemTable->minKey; while (pCommitter->nextKey < TSKEY_MAX) { - pCommitter->commitFid = tsdbKeyFid(pCommitter->nextKey, pCommitter->minutes, pCommitter->precision); - tsdbFidKeyRange(pCommitter->commitFid, pCommitter->minutes, pCommitter->precision, &pCommitter->minKey, - &pCommitter->maxKey); code = tsdbCommitFileData(pCommitter); if (code) goto _err; } @@ -1088,13 +1401,13 @@ static int32_t tsdbCommitDel(SCommitter *pCommitter) { int32_t iDelIdx = 0; int32_t nDelIdx = taosArrayGetSize(pCommitter->aDelIdx); int32_t iTbData = 0; - int32_t nTbData = taosArrayGetSize(pMemTable->aTbData); + int32_t nTbData = taosArrayGetSize(pCommitter->aTbDataP); STbData *pTbData; SDelIdx *pDelIdx; ASSERT(nTbData > 0); - pTbData = (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData); + pTbData = (STbData *)taosArrayGetP(pCommitter->aTbDataP, iTbData); pDelIdx = (iDelIdx < nDelIdx) ? (SDelIdx *)taosArrayGet(pCommitter->aDelIdx, iDelIdx) : NULL; while (true) { if (pTbData == NULL && pDelIdx == NULL) break; @@ -1120,7 +1433,7 @@ static int32_t tsdbCommitDel(SCommitter *pCommitter) { if (code) goto _err; iTbData++; - pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData) : NULL; + pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pCommitter->aTbDataP, iTbData) : NULL; continue; _commit_disk_del: @@ -1136,7 +1449,7 @@ static int32_t tsdbCommitDel(SCommitter *pCommitter) { if (code) goto _err; iTbData++; - pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pMemTable->aTbData, iTbData) : NULL; + pTbData = (iTbData < nTbData) ? (STbData *)taosArrayGetP(pCommitter->aTbDataP, iTbData) : NULL; iDelIdx++; pDelIdx = (iDelIdx < nDelIdx) ? (SDelIdx *)taosArrayGet(pCommitter->aDelIdx, iDelIdx) : NULL; continue; @@ -1184,6 +1497,7 @@ static int32_t tsdbEndCommit(SCommitter *pCommitter, int32_t eno) { tsdbUnrefMemTable(pMemTable); tsdbFSDestroy(&pCommitter->fs); + taosArrayDestroy(pCommitter->aTbDataP); tsdbInfo("vgId:%d, tsdb end commit", TD_VID(pTsdb->pVnode)); return code; diff --git a/source/dnode/vnode/src/tsdb/tsdbFS.c b/source/dnode/vnode/src/tsdb/tsdbFS.c index 74f1aef1fc7acc699b8dbc23521d957a2865ba3a..247de993381d98713fa6a4ca1938c11b044c8cd6 100644 --- a/source/dnode/vnode/src/tsdb/tsdbFS.c +++ b/source/dnode/vnode/src/tsdb/tsdbFS.c @@ -576,10 +576,7 @@ int32_t tsdbFSCopy(STsdb *pTsdb, STsdbFS *pFS) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } - fSet.pHeadF->nRef = 0; - fSet.pHeadF->commitID = pSet->pHeadF->commitID; - fSet.pHeadF->size = pSet->pHeadF->size; - fSet.pHeadF->offset = pSet->pHeadF->offset; + *fSet.pHeadF = *pSet->pHeadF; // data fSet.pDataF = (SDataFile *)taosMemoryMalloc(sizeof(SDataFile)); @@ -587,9 +584,7 @@ int32_t tsdbFSCopy(STsdb *pTsdb, STsdbFS *pFS) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } - fSet.pDataF->nRef = 0; - fSet.pDataF->commitID = pSet->pDataF->commitID; - fSet.pDataF->size = pSet->pDataF->size; + *fSet.pDataF = *pSet->pDataF; // data fSet.pLastF = (SLastFile *)taosMemoryMalloc(sizeof(SLastFile)); @@ -597,9 +592,7 @@ int32_t tsdbFSCopy(STsdb *pTsdb, STsdbFS *pFS) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } - fSet.pLastF->nRef = 0; - fSet.pLastF->commitID = pSet->pLastF->commitID; - fSet.pLastF->size = pSet->pLastF->size; + *fSet.pLastF = *pSet->pLastF; // last fSet.pSmaF = (SSmaFile *)taosMemoryMalloc(sizeof(SSmaFile)); @@ -607,9 +600,7 @@ int32_t tsdbFSCopy(STsdb *pTsdb, STsdbFS *pFS) { code = TSDB_CODE_OUT_OF_MEMORY; goto _exit; } - fSet.pSmaF->nRef = 0; - fSet.pSmaF->commitID = pSet->pSmaF->commitID; - fSet.pSmaF->size = pSet->pSmaF->size; + *fSet.pSmaF = *pSet->pSmaF; if (taosArrayPush(pFS->aDFileSet, &fSet) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; diff --git a/source/dnode/vnode/src/tsdb/tsdbFile.c b/source/dnode/vnode/src/tsdb/tsdbFile.c index 52a102f911290dc7a40516d594fc378ff2942cf0..00d2ac848f6d599fef54d9957047521e27062c89 100644 --- a/source/dnode/vnode/src/tsdb/tsdbFile.c +++ b/source/dnode/vnode/src/tsdb/tsdbFile.c @@ -58,6 +58,7 @@ int32_t tPutLastFile(uint8_t *p, SLastFile *pLastFile) { n += tPutI64v(p ? p + n : p, pLastFile->commitID); n += tPutI64v(p ? p + n : p, pLastFile->size); + n += tPutI64v(p ? p + n : p, pLastFile->offset); return n; } @@ -67,6 +68,7 @@ static int32_t tGetLastFile(uint8_t *p, SLastFile *pLastFile) { n += tGetI64v(p + n, &pLastFile->commitID); n += tGetI64v(p + n, &pLastFile->size); + n += tGetI64v(p + n, &pLastFile->offset); return n; } @@ -186,11 +188,16 @@ int32_t tPutDFileSet(uint8_t *p, SDFileSet *pSet) { n += tPutI32v(p ? p + n : p, pSet->diskId.level); n += tPutI32v(p ? p + n : p, pSet->diskId.id); n += tPutI32v(p ? p + n : p, pSet->fid); + + // data n += tPutHeadFile(p ? p + n : p, pSet->pHeadF); n += tPutDataFile(p ? p + n : p, pSet->pDataF); - n += tPutLastFile(p ? p + n : p, pSet->pLastF); n += tPutSmaFile(p ? p + n : p, pSet->pSmaF); + // last + n += tPutU8(p ? p + n : p, 1); // for future compatibility + n += tPutLastFile(p ? p + n : p, pSet->pLastF); + return n; } @@ -200,11 +207,17 @@ int32_t tGetDFileSet(uint8_t *p, SDFileSet *pSet) { n += tGetI32v(p + n, &pSet->diskId.level); n += tGetI32v(p + n, &pSet->diskId.id); n += tGetI32v(p + n, &pSet->fid); + + // data n += tGetHeadFile(p + n, pSet->pHeadF); n += tGetDataFile(p + n, pSet->pDataF); - n += tGetLastFile(p + n, pSet->pLastF); n += tGetSmaFile(p + n, pSet->pSmaF); + // last + uint8_t nLast; + n += tGetU8(p + n, &nLast); + n += tGetLastFile(p + n, pSet->pLastF); + return n; } diff --git a/source/dnode/vnode/src/tsdb/tsdbMemTable.c b/source/dnode/vnode/src/tsdb/tsdbMemTable.c index 34b37ffe9b4ed8a15f85525c393b86c75133c696..a6628463f8ff231505052dc3a48d8c7a59a6eaa5 100644 --- a/source/dnode/vnode/src/tsdb/tsdbMemTable.c +++ b/source/dnode/vnode/src/tsdb/tsdbMemTable.c @@ -15,6 +15,7 @@ #include "tsdb.h" +#define MEM_MIN_HASH 1024 #define SL_MAX_LEVEL 5 #define SL_NODE_SIZE(l) (sizeof(SMemSkipListNode) + sizeof(SMemSkipListNode *) * (l)*2) @@ -45,12 +46,12 @@ int32_t tsdbMemTableCreate(STsdb *pTsdb, SMemTable **ppMemTable) { pMemTable->nRef = 1; pMemTable->minKey = TSKEY_MAX; pMemTable->maxKey = TSKEY_MIN; - pMemTable->minVersion = VERSION_MAX; - pMemTable->maxVersion = VERSION_MIN; pMemTable->nRow = 0; pMemTable->nDel = 0; - pMemTable->aTbData = taosArrayInit(128, sizeof(STbData *)); - if (pMemTable->aTbData == NULL) { + pMemTable->nTbData = 0; + pMemTable->nBucket = MEM_MIN_HASH; + pMemTable->aBucket = (STbData **)taosMemoryCalloc(pMemTable->nBucket, sizeof(STbData *)); + if (pMemTable->aBucket == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; taosMemoryFree(pMemTable); goto _err; @@ -68,37 +69,30 @@ _err: void tsdbMemTableDestroy(SMemTable *pMemTable) { if (pMemTable) { vnodeBufPoolUnRef(pMemTable->pPool); - taosArrayDestroy(pMemTable->aTbData); + taosMemoryFree(pMemTable->aBucket); taosMemoryFree(pMemTable); } } -static int32_t tbDataPCmprFn(const void *p1, const void *p2) { - STbData *pTbData1 = *(STbData **)p1; - STbData *pTbData2 = *(STbData **)p2; +static FORCE_INLINE STbData *tsdbGetTbDataFromMemTableImpl(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid) { + STbData *pTbData = pMemTable->aBucket[TABS(uid) % pMemTable->nBucket]; - if (pTbData1->suid < pTbData2->suid) { - return -1; - } else if (pTbData1->suid > pTbData2->suid) { - return 1; - } - - if (pTbData1->uid < pTbData2->uid) { - return -1; - } else if (pTbData1->uid > pTbData2->uid) { - return 1; + while (pTbData) { + if (pTbData->uid == uid) break; + pTbData = pTbData->next; } - return 0; + return pTbData; } -void tsdbGetTbDataFromMemTable(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid, STbData **ppTbData) { - STbData *pTbData = &(STbData){.suid = suid, .uid = uid}; + +STbData *tsdbGetTbDataFromMemTable(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid) { + STbData *pTbData; taosRLockLatch(&pMemTable->latch); - void *p = taosArraySearch(pMemTable->aTbData, &pTbData, tbDataPCmprFn, TD_EQ); + pTbData = tsdbGetTbDataFromMemTableImpl(pMemTable, suid, uid); taosRUnLockLatch(&pMemTable->latch); - *ppTbData = p ? *(STbData **)p : NULL; + return pTbData; } int32_t tsdbInsertTableData(STsdb *pTsdb, int64_t version, SSubmitMsgIter *pMsgIter, SSubmitBlk *pBlock, @@ -184,10 +178,6 @@ int32_t tsdbDeleteTableData(STsdb *pTsdb, int64_t version, tb_uid_t suid, tb_uid pTbData->pTail = pDelData; } - // update the state of pMemTable and other (todo) - - pMemTable->minVersion = TMIN(pMemTable->minVersion, version); - pMemTable->maxVersion = TMAX(pMemTable->maxVersion, version); pMemTable->nDel++; if (TSDB_CACHE_LAST_ROW(pMemTable->pTsdb->pVnode->config) && tsdbKeyCmprFn(&lastKey, &pTbData->maxKey) >= 0) { @@ -320,18 +310,44 @@ _exit: return pIter->pRow; } +static int32_t tsdbMemTableRehash(SMemTable *pMemTable) { + int32_t code = 0; + + int32_t nBucket = pMemTable->nBucket * 2; + STbData **aBucket = (STbData **)taosMemoryCalloc(nBucket, sizeof(STbData *)); + if (aBucket == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _exit; + } + + for (int32_t iBucket = 0; iBucket < pMemTable->nBucket; iBucket++) { + STbData *pTbData = pMemTable->aBucket[iBucket]; + + while (pTbData) { + STbData *pNext = pTbData->next; + + int32_t idx = TABS(pTbData->uid) % nBucket; + pTbData->next = aBucket[idx]; + aBucket[idx] = pTbData; + + pTbData = pNext; + } + } + + taosMemoryFree(pMemTable->aBucket); + pMemTable->nBucket = nBucket; + pMemTable->aBucket = aBucket; + +_exit: + return code; +} + static int32_t tsdbGetOrCreateTbData(SMemTable *pMemTable, tb_uid_t suid, tb_uid_t uid, STbData **ppTbData) { - int32_t code = 0; - int32_t idx = 0; - STbData *pTbData = NULL; - STbData *pTbDataT = &(STbData){.suid = suid, .uid = uid}; + int32_t code = 0; // get - idx = taosArraySearchIdx(pMemTable->aTbData, &pTbDataT, tbDataPCmprFn, TD_GE); - if (idx >= 0) { - pTbData = (STbData *)taosArrayGetP(pMemTable->aTbData, idx); - if (tbDataPCmprFn(&pTbDataT, &pTbData) == 0) goto _exit; - } + STbData *pTbData = tsdbGetTbDataFromMemTableImpl(pMemTable, suid, uid); + if (pTbData) goto _exit; // create SVBufPool *pPool = pMemTable->pTsdb->pVnode->inUse; @@ -346,9 +362,6 @@ static int32_t tsdbGetOrCreateTbData(SMemTable *pMemTable, tb_uid_t suid, tb_uid pTbData->uid = uid; pTbData->minKey = TSKEY_MAX; pTbData->maxKey = TSKEY_MIN; - pTbData->minVersion = VERSION_MAX; - pTbData->maxVersion = VERSION_MIN; - pTbData->maxSkmVer = -1; pTbData->pHead = NULL; pTbData->pTail = NULL; pTbData->sl.seed = taosRand(); @@ -367,21 +380,23 @@ static int32_t tsdbGetOrCreateTbData(SMemTable *pMemTable, tb_uid_t suid, tb_uid SL_NODE_FORWARD(pTbData->sl.pTail, iLevel) = NULL; } - void *p; - if (idx < 0) { - idx = taosArrayGetSize(pMemTable->aTbData); + taosWLockLatch(&pMemTable->latch); + + if (pMemTable->nTbData >= pMemTable->nBucket) { + code = tsdbMemTableRehash(pMemTable); + if (code) { + taosWUnLockLatch(&pMemTable->latch); + goto _err; + } } - taosWLockLatch(&pMemTable->latch); - p = taosArrayInsert(pMemTable->aTbData, idx, &pTbData); - taosWUnLockLatch(&pMemTable->latch); + int32_t idx = TABS(uid) % pMemTable->nBucket; + pTbData->next = pMemTable->aBucket[idx]; + pMemTable->aBucket[idx] = pTbData; + pMemTable->nTbData++; - tsdbDebug("vgId:%d, add table data %p at idx:%d", TD_VID(pMemTable->pTsdb->pVnode), pTbData, idx); + taosWUnLockLatch(&pMemTable->latch); - if (p == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } _exit: *ppTbData = pTbData; return code; @@ -591,15 +606,9 @@ static int32_t tsdbInsertTableDataImpl(SMemTable *pMemTable, STbData *pTbData, i tsdbCacheInsertLast(pMemTable->pTsdb->lruCache, pTbData->uid, pLastRow, pMemTable->pTsdb); } - pTbData->minVersion = TMIN(pTbData->minVersion, version); - pTbData->maxVersion = TMAX(pTbData->maxVersion, version); - pTbData->maxSkmVer = TMAX(pTbData->maxSkmVer, pMsgIter->sversion); - // SMemTable pMemTable->minKey = TMIN(pMemTable->minKey, pTbData->minKey); pMemTable->maxKey = TMAX(pMemTable->maxKey, pTbData->maxKey); - pMemTable->minVersion = TMIN(pMemTable->minVersion, pTbData->minVersion); - pMemTable->maxVersion = TMAX(pMemTable->maxVersion, pTbData->maxVersion); pMemTable->nRow += nRow; pRsp->numOfRows = nRow; @@ -624,3 +633,41 @@ void tsdbUnrefMemTable(SMemTable *pMemTable) { tsdbMemTableDestroy(pMemTable); } } + +static FORCE_INLINE int32_t tbDataPCmprFn(const void *p1, const void *p2) { + STbData *pTbData1 = *(STbData **)p1; + STbData *pTbData2 = *(STbData **)p2; + + if (pTbData1->suid < pTbData2->suid) { + return -1; + } else if (pTbData1->suid > pTbData2->suid) { + return 1; + } + + if (pTbData1->uid < pTbData2->uid) { + return -1; + } else if (pTbData1->uid > pTbData2->uid) { + return 1; + } + + return 0; +} + +SArray *tsdbMemTableGetTbDataArray(SMemTable *pMemTable) { + SArray *aTbDataP = taosArrayInit(pMemTable->nTbData, sizeof(STbData *)); + if (aTbDataP == NULL) goto _exit; + + for (int32_t iBucket = 0; iBucket < pMemTable->nBucket; iBucket++) { + STbData *pTbData = pMemTable->aBucket[iBucket]; + + while (pTbData) { + taosArrayPush(aTbDataP, &pTbData); + pTbData = pTbData->next; + } + } + + taosArraySort(aTbDataP, tbDataPCmprFn); + +_exit: + return aTbDataP; +} diff --git a/source/dnode/vnode/src/tsdb/tsdbRead.c b/source/dnode/vnode/src/tsdb/tsdbRead.c index 335b311d0044b3c035ae8db4f2d72d7ce0590ed8..947ab2e7ffc71299013b3cec8c3288d05017ec7d 100644 --- a/source/dnode/vnode/src/tsdb/tsdbRead.c +++ b/source/dnode/vnode/src/tsdb/tsdbRead.c @@ -15,7 +15,10 @@ #include "osDef.h" #include "tsdb.h" + #define ASCENDING_TRAVERSE(o) (o == TSDB_ORDER_ASC) +#define ALL_ROWS_CHECKED_INDEX (INT16_MIN) +#define DEFAULT_ROW_INDEX_VAL (-1) typedef enum { EXTERNAL_ROWS_PREV = 0x1, @@ -29,16 +32,23 @@ typedef struct { bool hasVal; } SIterInfo; +typedef struct { + int32_t numOfBlocks; + int32_t numOfLastBlocks; +} SBlockNumber; + typedef struct STableBlockScanInfo { uint64_t uid; TSKEY lastKey; - SMapData mapData; // block info (compressed) - SArray* pBlockList; // block data index list - SIterInfo iter; // mem buffer skip list iterator - SIterInfo iiter; // imem buffer skip list iterator - SArray* delSkyline; // delete info for this table - int32_t fileDelIndex; - bool iterInit; // whether to initialize the in-memory skip list iterator or not + SMapData mapData; // block info (compressed) + SArray* pBlockList; // block data index list + SIterInfo iter; // mem buffer skip list iterator + SIterInfo iiter; // imem buffer skip list iterator + SArray* delSkyline; // delete info for this table + int32_t fileDelIndex; // file block delete index + int32_t lastBlockDelIndex;// delete index for last block + bool iterInit; // whether to initialize the in-memory skip list iterator or not + int16_t indexInBlockL;// row position in last block } STableBlockScanInfo; typedef struct SBlockOrderWrapper { @@ -59,8 +69,10 @@ typedef struct SIOCostSummary { double buildmemBlock; int64_t headFileLoad; double headFileLoadTime; - int64_t smaData; + int64_t smaDataLoad; double smaLoadTime; + int64_t lastBlockLoad; + double lastBlockLoadTime; } SIOCostSummary; typedef struct SBlockLoadSuppInfo { @@ -71,11 +83,28 @@ typedef struct SBlockLoadSuppInfo { char** buildBuf; // build string tmp buffer, todo remove it later after all string format being updated. } SBlockLoadSuppInfo; +typedef struct SVersionRange { + uint64_t minVer; + uint64_t maxVer; +} SVersionRange; + +typedef struct SLastBlockReader { + SArray* pBlockL; + int32_t currentBlockIndex; + SBlockData lastBlockData; + STimeWindow window; + SVersionRange verRange; + int32_t order; + uint64_t uid; + int16_t* rowIndex; // row index ptr, usually from the STableBlockScanInfo->indexInBlockL +} SLastBlockReader; + typedef struct SFilesetIter { - int32_t numOfFiles; // number of total files - int32_t index; // current accessed index in the list - SArray* pFileList; // data file list - int32_t order; + int32_t numOfFiles; // number of total files + int32_t index; // current accessed index in the list + SArray* pFileList; // data file list + int32_t order; + SLastBlockReader* pLastBlockReader; // last file block reader } SFilesetIter; typedef struct SFileDataBlockInfo { @@ -87,9 +116,9 @@ typedef struct SFileDataBlockInfo { typedef struct SDataBlockIter { int32_t numOfBlocks; int32_t index; - SArray* blockList; // SArray + SArray* blockList; // SArray int32_t order; - SBlock block; // current SBlock data + SBlock block; // current SBlock data SHashObj* pTableMap; } SDataBlockIter; @@ -100,11 +129,6 @@ typedef struct SFileBlockDumpInfo { bool allDumped; } SFileBlockDumpInfo; -typedef struct SVersionRange { - uint64_t minVer; - uint64_t maxVer; -} SVersionRange; - typedef struct SReaderStatus { bool loadFromFile; // check file stage SHashObj* pTableMap; // SHash @@ -145,10 +169,11 @@ static int buildDataBlockFromBufImpl(STableBlockScanInfo* pBlockScanInfo, i static TSDBROW* getValidRow(SIterInfo* pIter, const SArray* pDelList, STsdbReader* pReader); static int32_t doMergeRowsInFileBlocks(SBlockData* pBlockData, STableBlockScanInfo* pScanInfo, STsdbReader* pReader, SRowMerger* pMerger); +static int32_t doMergeRowsInLastBlock(SLastBlockReader* pLastBlockReader, STableBlockScanInfo* pScanInfo, int64_t ts, SRowMerger* pMerger); static int32_t doMergeRowsInBuf(SIterInfo* pIter, uint64_t uid, int64_t ts, SArray* pDelList, SRowMerger* pMerger, STsdbReader* pReader); static int32_t doAppendRowFromTSRow(SSDataBlock* pBlock, STsdbReader* pReader, STSRow* pTSRow, uint64_t uid); -static int32_t doAppendRowFromBlock(SSDataBlock* pResBlock, STsdbReader* pReader, SBlockData* pBlockData, +static int32_t doAppendRowFromFileBlock(SSDataBlock* pResBlock, STsdbReader* pReader, SBlockData* pBlockData, int32_t rowIndex); static void setComposedBlockFlag(STsdbReader* pReader, bool composed); static bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey, int32_t order); @@ -162,6 +187,9 @@ static int32_t initDelSkylineIterator(STableBlockScanInfo* pBlockScanInfo, STsdb static STsdb* getTsdbByRetentions(SVnode* pVnode, TSKEY winSKey, SRetention* retentions, const char* idstr, int8_t* pLevel); static SVersionRange getQueryVerRange(SVnode* pVnode, SQueryTableDataCond* pCond, int8_t level); +static int64_t getCurrentKeyInLastBlock(SLastBlockReader* pLastBlockReader); +static bool hasDataInLastBlock(SLastBlockReader* pLastBlockReader); +static int32_t doBuildDataBlock(STsdbReader* pReader); static int32_t setColumnIdSlotList(STsdbReader* pReader, SSDataBlock* pBlock) { SBlockLoadSuppInfo* pSupInfo = &pReader->suppInfo; @@ -182,7 +210,6 @@ static int32_t setColumnIdSlotList(STsdbReader* pReader, SSDataBlock* pBlock) { if (IS_VAR_DATA_TYPE(pCol->info.type)) { pSupInfo->buildBuf[i] = taosMemoryMalloc(pCol->info.bytes); - // tsdbInfo("-------------------%d\n", pCol->info.bytes); } } @@ -199,7 +226,7 @@ static SHashObj* createDataBlockScanInfo(STsdbReader* pTsdbReader, const STableK } for (int32_t j = 0; j < numOfTables; ++j) { - STableBlockScanInfo info = {.lastKey = 0, .uid = idList[j].uid}; + STableBlockScanInfo info = {.lastKey = 0, .uid = idList[j].uid, .indexInBlockL = DEFAULT_ROW_INDEX_VAL}; if (ASCENDING_TRAVERSE(pTsdbReader->order)) { if (info.lastKey == INT64_MIN || info.lastKey < pTsdbReader->window.skey) { info.lastKey = pTsdbReader->window.skey; @@ -293,15 +320,36 @@ static void limitOutputBufferSize(const SQueryTableDataCond* pCond, int32_t* cap } // init file iterator -static int32_t initFilesetIterator(SFilesetIter* pIter, SArray* aDFileSet, int32_t order, const char* idstr) { +static int32_t initFilesetIterator(SFilesetIter* pIter, SArray* aDFileSet, STsdbReader* pReader/*int32_t order, const char* idstr*/) { size_t numOfFileset = taosArrayGetSize(aDFileSet); - pIter->index = ASCENDING_TRAVERSE(order) ? -1 : numOfFileset; - pIter->order = order; + pIter->index = ASCENDING_TRAVERSE(pReader->order) ? -1 : numOfFileset; + pIter->order = pReader->order; pIter->pFileList = aDFileSet; pIter->numOfFiles = numOfFileset; - tsdbDebug("init fileset iterator, total files:%d %s", pIter->numOfFiles, idstr); + if (pIter->pLastBlockReader == NULL) { + pIter->pLastBlockReader = taosMemoryCalloc(1, sizeof(struct SLastBlockReader)); + if (pIter->pLastBlockReader == NULL) { + int32_t code = TSDB_CODE_OUT_OF_MEMORY; + tsdbError("failed to prepare the last block iterator, code:%d %s", tstrerror(code), pReader->idStr); + return code; + } + + SLastBlockReader* pLReader = pIter->pLastBlockReader; + pLReader->pBlockL = taosArrayInit(4, sizeof(SBlockL)); + pLReader->order = pReader->order; + pLReader->window = pReader->window; + pLReader->verRange = pReader->verRange; + pLReader->currentBlockIndex = -1; + + int32_t code = tBlockDataCreate(&pLReader->lastBlockData); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + } + + tsdbDebug("init fileset iterator, total files:%d %s", pIter->numOfFiles, pReader->idStr); return TSDB_CODE_SUCCESS; } @@ -361,7 +409,7 @@ _err: static void resetDataBlockIterator(SDataBlockIter* pIter, int32_t order, SHashObj* pTableMap) { pIter->order = order; pIter->index = -1; - pIter->numOfBlocks = -1; + pIter->numOfBlocks = 0; if (pIter->blockList == NULL) { pIter->blockList = taosArrayInit(4, sizeof(SFileDataBlockInfo)); } else { @@ -419,7 +467,7 @@ static int32_t tsdbReaderCreate(SVnode* pVnode, SQueryTableDataCond* pCond, STsd pReader->pTsdb = getTsdbByRetentions(pVnode, pCond->twindows.skey, pVnode->config.tsdbCfg.retentions, idstr, &level); pReader->suid = pCond->suid; pReader->order = pCond->order; - pReader->capacity = capacity; + pReader->capacity = 4096; pReader->idStr = (idstr != NULL) ? strdup(idstr) : NULL; pReader->verRange = getQueryVerRange(pVnode, pCond, level); pReader->type = pCond->type; @@ -440,7 +488,7 @@ static int32_t tsdbReaderCreate(SVnode* pVnode, SQueryTableDataCond* pCond, STsd pSup->tsColAgg.colId = PRIMARYKEY_TIMESTAMP_COL_ID; - code = tBlockDataInit(&pReader->status.fileBlockData); + code = tBlockDataCreate(&pReader->status.fileBlockData); if (code != TSDB_CODE_SUCCESS) { terrno = code; goto _end; @@ -547,14 +595,14 @@ static int32_t doLoadBlockIndex(STsdbReader* pReader, SDataFReader* pFileReader, SArray* aBlockIdx = taosArrayInit(8, sizeof(SBlockIdx)); int64_t st = taosGetTimestampUs(); - int32_t code = tsdbReadBlockIdx(pFileReader, aBlockIdx, NULL); + int32_t code = tsdbReadBlockIdx(pFileReader, aBlockIdx); if (code != TSDB_CODE_SUCCESS) { goto _end; } size_t num = taosArrayGetSize(aBlockIdx); if (num == 0) { - taosArrayClear(aBlockIdx); + taosArrayDestroy(aBlockIdx); return TSDB_CODE_SUCCESS; } @@ -594,24 +642,29 @@ _end: return code; } -static int32_t doLoadFileBlock(STsdbReader* pReader, SArray* pIndexList, uint32_t* numOfValidTables, - int32_t* numOfBlocks) { - size_t numOfTables = taosArrayGetSize(pIndexList); - *numOfValidTables = 0; - - int64_t st = taosGetTimestampUs(); - size_t size = 0; - +static void cleanupTableScanInfo(SHashObj* pTableMap) { STableBlockScanInfo* px = NULL; while (1) { - px = taosHashIterate(pReader->status.pTableMap, px); + px = taosHashIterate(pTableMap, px); if (px == NULL) { break; } + // reset the index in last block when handing a new file + px->indexInBlockL = -1; tMapDataClear(&px->mapData); taosArrayClear(px->pBlockList); } +} + +static int32_t doLoadFileBlock(STsdbReader* pReader, SArray* pIndexList, SArray* pLastBlockIndex, + SBlockNumber * pBlockNum, SArray* pQualifiedLastBlock) { + int32_t numOfQTable = 0; + size_t sizeInDisk = 0; + size_t numOfTables = taosArrayGetSize(pIndexList); + + int64_t st = taosGetTimestampUs(); + cleanupTableScanInfo(pReader->status.pTableMap); for (int32_t i = 0; i < numOfTables; ++i) { SBlockIdx* pBlockIdx = taosArrayGet(pIndexList, i); @@ -619,9 +672,9 @@ static int32_t doLoadFileBlock(STsdbReader* pReader, SArray* pIndexList, uint32_ STableBlockScanInfo* pScanInfo = taosHashGet(pReader->status.pTableMap, &pBlockIdx->uid, sizeof(int64_t)); tMapDataReset(&pScanInfo->mapData); - tsdbReadBlock(pReader->pFileReader, pBlockIdx, &pScanInfo->mapData, NULL); + tsdbReadBlock(pReader->pFileReader, pBlockIdx, &pScanInfo->mapData); - size += pScanInfo->mapData.nData; + sizeInDisk += pScanInfo->mapData.nData; for (int32_t j = 0; j < pScanInfo->mapData.nItem; ++j) { SBlock block = {0}; tMapDataGetItemByIdx(&pScanInfo->mapData, j, &block, tGetBlock); @@ -632,7 +685,7 @@ static int32_t doLoadFileBlock(STsdbReader* pReader, SArray* pIndexList, uint32_ } // 2. version range check - if (block.minVersion > pReader->verRange.maxVer || block.maxVersion < pReader->verRange.minVer) { + if (block.minVer > pReader->verRange.maxVer || block.maxVer < pReader->verRange.minVer) { continue; } @@ -642,30 +695,54 @@ static int32_t doLoadFileBlock(STsdbReader* pReader, SArray* pIndexList, uint32_ return TSDB_CODE_OUT_OF_MEMORY; } - (*numOfBlocks) += 1; + pBlockNum->numOfBlocks += 1; } if (pScanInfo->pBlockList != NULL && taosArrayGetSize(pScanInfo->pBlockList) > 0) { - (*numOfValidTables) += 1; + numOfQTable += 1; } } + size_t numOfLast = taosArrayGetSize(pLastBlockIndex); + for(int32_t i = 0; i < numOfLast; ++i) { + SBlockL* pLastBlock = taosArrayGet(pLastBlockIndex, i); + if (pLastBlock->suid != pReader->suid) { + continue; + } + + { + // 1. time range check + if (pLastBlock->minKey > pReader->window.ekey || pLastBlock->maxKey < pReader->window.skey) { + continue; + } + + // 2. version range check + if (pLastBlock->minVer > pReader->verRange.maxVer || pLastBlock->maxVer < pReader->verRange.minVer) { + continue; + } + + pBlockNum->numOfLastBlocks += 1; + taosArrayPush(pQualifiedLastBlock, pLastBlock); + } + } + + int32_t total = pBlockNum->numOfLastBlocks + pBlockNum->numOfBlocks; + double el = (taosGetTimestampUs() - st) / 1000.0; - tsdbDebug("load block of %d tables completed, blocks:%d in %d tables, size:%.2f Kb, elapsed time:%.2f ms %s", - numOfTables, *numOfBlocks, *numOfValidTables, size / 1000.0, el, pReader->idStr); + tsdbDebug("load block of %d tables completed, blocks:%d in %d tables, lastBlock:%d, size:%.2f Kb, elapsed time:%.2f ms %s", + numOfTables, total, numOfQTable, pBlockNum->numOfLastBlocks, sizeInDisk + / 1000.0, el, pReader->idStr); - pReader->cost.numOfBlocks += (*numOfBlocks); + pReader->cost.numOfBlocks += total; pReader->cost.headFileLoadTime += el; return TSDB_CODE_SUCCESS; } -// todo remove pblock parameter -static void setBlockAllDumped(SFileBlockDumpInfo* pDumpInfo, SBlock* pBlock, int32_t order) { +static void setBlockAllDumped(SFileBlockDumpInfo* pDumpInfo, int64_t maxKey, int32_t order) { int32_t step = ASCENDING_TRAVERSE(order) ? 1 : -1; - pDumpInfo->allDumped = true; - pDumpInfo->lastKey = pBlock->maxKey.ts + step; + pDumpInfo->lastKey = maxKey + step; } static void doCopyColVal(SColumnInfoData* pColInfoData, int32_t rowIndex, int32_t colIndex, SColVal* pColVal, @@ -685,8 +762,13 @@ static void doCopyColVal(SColumnInfoData* pColInfoData, int32_t rowIndex, int32_ } static SFileDataBlockInfo* getCurrentBlockInfo(SDataBlockIter* pBlockIter) { - SFileDataBlockInfo* pFBlockInfo = taosArrayGet(pBlockIter->blockList, pBlockIter->index); - return pFBlockInfo; + if (taosArrayGetSize(pBlockIter->blockList) == 0) { + ASSERT(pBlockIter->numOfBlocks == taosArrayGetSize(pBlockIter->blockList)); + return NULL; + } + + SFileDataBlockInfo* pBlockInfo = taosArrayGet(pBlockIter->blockList, pBlockIter->index); + return pBlockInfo; } static SBlock* getCurrentBlock(SDataBlockIter* pBlockIter) { return &pBlockIter->block; } @@ -736,19 +818,20 @@ static int32_t copyBlockDataToSDataBlock(STsdbReader* pReader, STableBlockScanIn pColData = taosArrayGet(pResBlock->pDataBlock, i); SColData* pData = tBlockDataGetColDataByIdx(pBlockData, colIndex); - - if (pData->cid == pColData->info.colId) { + if (pData->cid < pColData->info.colId) { + colIndex += 1; + } else if (pData->cid == pColData->info.colId) { for (int32_t j = pDumpInfo->rowIndex; j < endIndex && j >= 0; j += step) { tColDataGetValue(pData, j, &cv); doCopyColVal(pColData, rowIndex++, i, &cv, pSupInfo); } colIndex += 1; + i += 1; ASSERT(rowIndex == remain); } else { // the specified column does not exist in file block, fill with null data colDataAppendNNULL(pColData, 0, remain); + i += 1; } - - i += 1; } while (i < numOfOutputCols) { @@ -760,7 +843,7 @@ static int32_t copyBlockDataToSDataBlock(STsdbReader* pReader, STableBlockScanIn pResBlock->info.rows = remain; pDumpInfo->rowIndex += step * remain; - setBlockAllDumped(pDumpInfo, pBlock, pReader->order); + setBlockAllDumped(pDumpInfo, pBlock->maxKey.ts, pReader->order); double elapsedTime = (taosGetTimestampUs() - st) / 1000.0; pReader->cost.blockLoadTime += elapsedTime; @@ -769,47 +852,77 @@ static int32_t copyBlockDataToSDataBlock(STsdbReader* pReader, STableBlockScanIn tsdbDebug("%p copy file block to sdatablock, global index:%d, table index:%d, brange:%" PRId64 "-%" PRId64 ", rows:%d, remain:%d, minVer:%" PRId64 ", maxVer:%" PRId64 ", elapsed time:%.2f ms, %s", pReader, pBlockIter->index, pFBlock->tbBlockIdx, pBlock->minKey.ts, pBlock->maxKey.ts, remain, unDumpedRows, - pBlock->minVersion, pBlock->maxVersion, elapsedTime, pReader->idStr); + pBlock->minVer, pBlock->maxVer, elapsedTime, pReader->idStr); return TSDB_CODE_SUCCESS; } -static int32_t doLoadFileBlockData(STsdbReader* pReader, SDataBlockIter* pBlockIter, - STableBlockScanInfo* pBlockScanInfo, SBlockData* pBlockData) { +static int32_t doLoadFileBlockData(STsdbReader* pReader, SDataBlockIter* pBlockIter, SBlockData* pBlockData) { int64_t st = taosGetTimestampUs(); + double elapsedTime = 0; + int32_t code = 0; - SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(pBlockIter); - SBlock* pBlock = getCurrentBlock(pBlockIter); + SFileDataBlockInfo* pBlockInfo = getCurrentBlockInfo(pBlockIter); + SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; - SSDataBlock* pResBlock = pReader->pResBlock; - int32_t numOfCols = blockDataGetNumOfCols(pResBlock); + if (pBlockInfo != NULL) { + SBlock* pBlock = getCurrentBlock(pBlockIter); + code = tsdbReadDataBlock(pReader->pFileReader, pBlock, pBlockData); + if (code != TSDB_CODE_SUCCESS) { + tsdbError("%p error occurs in loading file block, global index:%d, table index:%d, brange:%" PRId64 "-%" PRId64 + ", rows:%d, code:%s %s", + pReader, pBlockIter->index, pBlockInfo->tbBlockIdx, pBlock->minKey.ts, pBlock->maxKey.ts, pBlock->nRow, + tstrerror(code), pReader->idStr); + goto _error; + } - SBlockLoadSuppInfo* pSupInfo = &pReader->suppInfo; - SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; + elapsedTime = (taosGetTimestampUs() - st) / 1000.0; - SBlockIdx blockIdx = {.suid = pReader->suid, .uid = pBlockScanInfo->uid}; - int32_t code = - tsdbReadColData(pReader->pFileReader, &blockIdx, pBlock, pSupInfo->colIds, numOfCols, pBlockData, NULL, NULL); - if (code != TSDB_CODE_SUCCESS) { - goto _error; + tsdbDebug("%p load file block into buffer, global index:%d, table index:%d, brange:%" PRId64 "-%" PRId64 + ", rows:%d, minVer:%" PRId64 ", maxVer:%" PRId64 ", elapsed time:%.2f ms, %s", + pReader, pBlockIter->index, pBlockInfo->tbBlockIdx, pBlock->minKey.ts, pBlock->maxKey.ts, pBlock->nRow, + pBlock->minVer, pBlock->maxVer, elapsedTime, pReader->idStr); + } else { +#if 0 + SLastBlockReader* pLastBlockReader = pReader->status.fileIter.pLastBlockReader; + + uint64_t uid = pBlockInfo->uid; + SArray* pBlocks = pLastBlockReader->pBlockL; + + pLastBlockReader->currentBlockIndex = -1; + + // find the correct SBlockL + for(int32_t i = 0; i < taosArrayGetSize(pBlocks); ++i) { + SBlockL* pBlock = taosArrayGet(pBlocks, i); + if (pBlock->minUid >= uid && pBlock->maxUid <= uid) { + pLastBlockReader->currentBlockIndex = i; + break; + } + } + +// SBlockL* pBlockL = taosArrayGet(pLastBlockReader->pBlockL, *index); + code = tsdbReadLastBlock(pReader->pFileReader, pBlockL, pBlockData); + if (code != TSDB_CODE_SUCCESS) { + tsdbDebug("%p error occurs in loading last block into buffer, last block index:%d, total:%d brange:%" PRId64 "-%" PRId64 + ", rows:%d, minVer:%" PRId64 ", maxVer:%" PRId64 ", code:%s %s", + pReader, *index, pBlockIter->numOfBlocks.numOfLastBlocks, 0, 0, pBlockL->nRow, + pBlockL->minVer, pBlockL->maxVer, tstrerror(code), pReader->idStr); + goto _error; + } + + tsdbDebug("%p load last file block into buffer, last block index:%d, total:%d brange:%" PRId64 "-%" PRId64 + ", rows:%d, minVer:%" PRId64 ", maxVer:%" PRId64 ", elapsed time:%.2f ms, %s", + pReader, *index, pBlockIter->numOfBlocks.numOfLastBlocks, 0, 0, pBlockL->nRow, + pBlockL->minVer, pBlockL->maxVer, elapsedTime, pReader->idStr); +#endif } - double elapsedTime = (taosGetTimestampUs() - st) / 1000.0; pReader->cost.blockLoadTime += elapsedTime; - pDumpInfo->allDumped = false; - tsdbDebug("%p load file block into buffer, global index:%d, table index:%d, brange:%" PRId64 "-%" PRId64 - ", rows:%d, minVer:%" PRId64 ", maxVer:%" PRId64 ", elapsed time:%.2f ms, %s", - pReader, pBlockIter->index, pFBlock->tbBlockIdx, pBlock->minKey.ts, pBlock->maxKey.ts, pBlock->nRow, - pBlock->minVersion, pBlock->maxVersion, elapsedTime, pReader->idStr); return TSDB_CODE_SUCCESS; _error: - tsdbError("%p error occurs in loading file block, global index:%d, table index:%d, brange:%" PRId64 "-%" PRId64 - ", rows:%d, %s", - pReader, pBlockIter->index, pFBlock->tbBlockIdx, pBlock->minKey.ts, pBlock->maxKey.ts, pBlock->nRow, - pReader->idStr); return code; } @@ -865,10 +978,11 @@ static int32_t fileDataBlockOrderCompar(const void* pLeft, const void* pRight, v static int32_t doSetCurrentBlock(SDataBlockIter* pBlockIter) { SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(pBlockIter); - STableBlockScanInfo* pScanInfo = taosHashGet(pBlockIter->pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); - - int32_t* mapDataIndex = taosArrayGet(pScanInfo->pBlockList, pFBlock->tbBlockIdx); - tMapDataGetItemByIdx(&pScanInfo->mapData, *mapDataIndex, &pBlockIter->block, tGetBlock); + if (pFBlock != NULL) { + STableBlockScanInfo* pScanInfo = taosHashGet(pBlockIter->pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); + int32_t* mapDataIndex = taosArrayGet(pScanInfo->pBlockList, pFBlock->tbBlockIdx); + tMapDataGetItemByIdx(&pScanInfo->mapData, *mapDataIndex, &pBlockIter->block, tGetBlock); + } #if 0 qDebug("check file block, table uid:%"PRIu64" index:%d offset:%"PRId64", ", pScanInfo->uid, *mapDataIndex, pBlockIter->block.aSubBlock[0].offset); @@ -945,7 +1059,7 @@ static int32_t initBlockIterator(STsdbReader* pReader, SDataBlockIter* pBlockIte int64_t et = taosGetTimestampUs(); tsdbDebug("%p create blocks info struct completed for one table, %d blocks not sorted, elapsed time:%.2f ms %s", - pReader, cnt, (et - st) / 1000.0, pReader->idStr); + pReader, numOfBlocks, (et - st) / 1000.0, pReader->idStr); pBlockIter->index = asc ? 0 : (numOfBlocks - 1); cleanupBlockOrderSupporter(&sup); @@ -956,7 +1070,7 @@ static int32_t initBlockIterator(STsdbReader* pReader, SDataBlockIter* pBlockIte tsdbDebug("%p create data blocks info struct completed, %d blocks in %d tables %s", pReader, cnt, sup.numOfTables, pReader->idStr); - assert(cnt <= numOfBlocks && sup.numOfTables <= numOfTables); + ASSERT(cnt <= numOfBlocks && sup.numOfTables <= numOfTables); SMultiwayMergeTreeInfo* pTree = NULL; uint8_t ret = tMergeTreeCreate(&pTree, sup.numOfTables, &sup, fileDataBlockOrderCompar); @@ -983,7 +1097,7 @@ static int32_t initBlockIterator(STsdbReader* pReader, SDataBlockIter* pBlockIte } int64_t et = taosGetTimestampUs(); - tsdbDebug("%p %d data blocks access order completed, elapsed time:%.2f ms %s", pReader, cnt, (et - st) / 1000.0, + tsdbDebug("%p %d data blocks access order completed, elapsed time:%.2f ms %s", pReader, numOfBlocks, (et - st) / 1000.0, pReader->idStr); cleanupBlockOrderSupporter(&sup); taosMemoryFree(pTree); @@ -1014,8 +1128,8 @@ static bool blockIteratorNext(SDataBlockIter* pBlockIter) { static int32_t dataBlockPartiallyRequired(STimeWindow* pWindow, SVersionRange* pVerRange, SBlock* pBlock) { return (pWindow->ekey < pBlock->maxKey.ts && pWindow->ekey >= pBlock->minKey.ts) || (pWindow->skey > pBlock->minKey.ts && pWindow->skey <= pBlock->maxKey.ts) || - (pVerRange->minVer > pBlock->minVersion && pVerRange->minVer <= pBlock->maxVersion) || - (pVerRange->maxVer < pBlock->maxVersion && pVerRange->maxVer >= pBlock->minVersion); + (pVerRange->minVer > pBlock->minVer && pVerRange->minVer <= pBlock->maxVer) || + (pVerRange->maxVer < pBlock->maxVer && pVerRange->maxVer >= pBlock->minVer); } static SBlock* getNeighborBlockOfSameTable(SFileDataBlockInfo* pFBlockInfo, STableBlockScanInfo* pTableBlockScanInfo, @@ -1095,8 +1209,8 @@ static bool bufferDataInFileBlockGap(int32_t order, TSDBKEY key, SBlock* pBlock) } static bool keyOverlapFileBlock(TSDBKEY key, SBlock* pBlock, SVersionRange* pVerRange) { - return (key.ts >= pBlock->minKey.ts && key.ts <= pBlock->maxKey.ts) && (pBlock->maxVersion >= pVerRange->minVer) && - (pBlock->minVersion <= pVerRange->maxVer); + return (key.ts >= pBlock->minKey.ts && key.ts <= pBlock->maxKey.ts) && (pBlock->maxVer >= pVerRange->minVer) && + (pBlock->minVer <= pVerRange->maxVer); } static bool doCheckforDatablockOverlap(STableBlockScanInfo* pBlockScanInfo, const SBlock* pBlock) { @@ -1105,11 +1219,11 @@ static bool doCheckforDatablockOverlap(STableBlockScanInfo* pBlockScanInfo, cons for (int32_t i = pBlockScanInfo->fileDelIndex; i < num; i += 1) { TSDBKEY* p = taosArrayGet(pBlockScanInfo->delSkyline, i); if (p->ts >= pBlock->minKey.ts && p->ts <= pBlock->maxKey.ts) { - if (p->version >= pBlock->minVersion) { + if (p->version >= pBlock->minVer) { return true; } } else if (p->ts < pBlock->minKey.ts) { // p->ts < pBlock->minKey.ts - if (p->version >= pBlock->minVersion) { + if (p->version >= pBlock->minVer) { if (i < num - 1) { TSDBKEY* pnext = taosArrayGet(pBlockScanInfo->delSkyline, i + 1); if (i + 1 == num - 1) { // pnext is the last point @@ -1117,7 +1231,7 @@ static bool doCheckforDatablockOverlap(STableBlockScanInfo* pBlockScanInfo, cons return true; } } else { - if (pnext->ts >= pBlock->minKey.ts && pnext->version >= pBlock->minVersion) { + if (pnext->ts >= pBlock->minKey.ts && pnext->version >= pBlock->minVer) { return true; } } @@ -1169,7 +1283,7 @@ static bool overlapWithDelSkyline(STableBlockScanInfo* pBlockScanInfo, const SBl // 4. output buffer should be large enough to hold all rows in current block // 5. delete info should not overlap with current block data static bool fileBlockShouldLoad(STsdbReader* pReader, SFileDataBlockInfo* pFBlock, SBlock* pBlock, - STableBlockScanInfo* pScanInfo, TSDBKEY key) { + STableBlockScanInfo* pScanInfo, TSDBKEY key, SLastBlockReader* pLastBlockReader) { int32_t neighborIndex = 0; SBlock* pNeighbor = getNeighborBlockOfSameTable(pFBlock, pScanInfo, &neighborIndex, pReader->order); @@ -1184,8 +1298,30 @@ static bool fileBlockShouldLoad(STsdbReader* pReader, SFileDataBlockInfo* pFBloc bool hasDup = (pBlock->nSubBlock == 1) ? pBlock->hasDup : true; bool overlapWithDel = overlapWithDelSkyline(pScanInfo, pBlock, pReader->order); - return (overlapWithNeighbor || hasDup || dataBlockPartiallyRequired(&pReader->window, &pReader->verRange, pBlock) || - keyOverlapFileBlock(key, pBlock, &pReader->verRange) || (pBlock->nRow > pReader->capacity) || overlapWithDel); + // todo here we need to each key in the last files to identify if it is really overlapped with last block + bool overlapWithlastBlock = false; + if (taosArrayGetSize(pLastBlockReader->pBlockL) > 0 && (pLastBlockReader->currentBlockIndex != -1)) { + SBlockL *pBlockL = taosArrayGet(pLastBlockReader->pBlockL, pLastBlockReader->currentBlockIndex); + overlapWithlastBlock = !(pBlock->maxKey.ts < pBlockL->minKey || pBlock->minKey.ts > pBlockL->maxKey); + } + + bool moreThanOutputCapacity = pBlock->nRow > pReader->capacity; + bool partiallyRequired = dataBlockPartiallyRequired(&pReader->window, &pReader->verRange, pBlock); + bool overlapWithKey = keyOverlapFileBlock(key, pBlock, &pReader->verRange); + + bool loadDataBlock = (overlapWithNeighbor || hasDup || partiallyRequired || overlapWithKey || + moreThanOutputCapacity || overlapWithDel || overlapWithlastBlock); + + // log the reason why load the datablock for profile + if (loadDataBlock) { + tsdbDebug("%p uid:%" PRIu64 + " need to load the datablock, reason overlapwithneighborblock:%d, hasDup:%d, partiallyRequired:%d, " + "overlapWithKey:%d, greaterThanBuf:%d, overlapWithDel:%d, overlapWithlastBlock:%d, %s", + pReader, pFBlock->uid, overlapWithNeighbor, hasDup, partiallyRequired, overlapWithKey, + moreThanOutputCapacity, overlapWithDel, overlapWithlastBlock, pReader->idStr); + } + + return loadDataBlock; } static int32_t buildDataBlockFromBuf(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, int64_t endKey) { @@ -1224,7 +1360,7 @@ static bool tryCopyDistinctRowFromFileBlock(STsdbReader* pReader, SBlockData* pB int64_t nextKey = pBlockData->aTSKEY[pDumpInfo->rowIndex + step]; if (nextKey != key) { // merge is not needed - doAppendRowFromBlock(pReader->pResBlock, pReader, pBlockData, pDumpInfo->rowIndex); + doAppendRowFromFileBlock(pReader->pResBlock, pReader, pBlockData, pDumpInfo->rowIndex); pDumpInfo->rowIndex += step; return true; } @@ -1258,8 +1394,124 @@ static FORCE_INLINE STSchema* doGetSchemaForTSRow(int32_t sversion, STsdbReader* return pReader->pMemSchema; } +static int32_t doMergeBufAndFileRows_Rv(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, TSDBROW* pRow, + SIterInfo* pIter, int64_t key, SLastBlockReader* pLastBlockReader) { + SRowMerger merge = {0}; + STSRow* pTSRow = NULL; + SBlockData* pBlockData = &pReader->status.fileBlockData; + SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; + + int64_t tsLast = INT64_MIN; + if (pLastBlockReader->lastBlockData.nRow > 0) { + tsLast = getCurrentKeyInLastBlock(pLastBlockReader); + } + + TSDBKEY k = TSDBROW_KEY(pRow); + TSDBROW fRow = tsdbRowFromBlockData(pBlockData, pDumpInfo->rowIndex); + + SBlockData* pLastBlockData = &pLastBlockReader->lastBlockData; + + int64_t minKey = 0; + if (pReader->order == TSDB_ORDER_ASC) { + minKey = INT64_MAX; // chosen the minimum value + if (minKey > tsLast && pLastBlockReader->lastBlockData.nRow > 0) { + minKey = tsLast; + } + + if (minKey > k.ts) { + minKey = k.ts; + } + + if (minKey > key && pBlockData->nRow > 0) { + minKey = key; + } + } else { + minKey = INT64_MIN; + if (minKey < tsLast && pLastBlockReader->lastBlockData.nRow > 0) { + minKey = tsLast; + } + + if (minKey < k.ts) { + minKey = k.ts; + } + + if (minKey < key && pBlockData->nRow > 0) { + minKey = key; + } + } + + bool init = false; + + // ASC: file block ---> last block -----> imem -----> mem + //DESC: mem -----> imem -----> last block -----> file block + if (pReader->order == TSDB_ORDER_ASC) { + if (minKey == key) { + init = true; + tRowMergerInit(&merge, &fRow, pReader->pSchema); + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + } + + if (minKey == tsLast) { + TSDBROW fRow1 = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + if (init) { + tRowMerge(&merge, &fRow1); + } else { + init = true; + tRowMergerInit(&merge, &fRow1, pReader->pSchema); + } + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, tsLast, &merge); + } + + if (minKey == k.ts) { + if (init) { + tRowMerge(&merge, pRow); + } else { + init = true; + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(pRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, pRow, pSchema); + } + doMergeRowsInBuf(pIter, pBlockScanInfo->uid, k.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + } else { + if (minKey == k.ts) { + init = true; + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(pRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, pRow, pSchema); + doMergeRowsInBuf(pIter, pBlockScanInfo->uid, k.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + + if (minKey == tsLast) { + TSDBROW fRow1 = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + if (init) { + tRowMerge(&merge, &fRow1); + } else { + init = true; + tRowMergerInit(&merge, &fRow1, pReader->pSchema); + } + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, tsLast, &merge); + } + + if (minKey == key) { + if (init) { + tRowMerge(&merge, &fRow); + } else { + init = true; + tRowMergerInit(&merge, &fRow, pReader->pSchema); + } + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + } + } + + tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; +} + static int32_t doMergeBufAndFileRows(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, TSDBROW* pRow, - SIterInfo* pIter, int64_t key) { + SIterInfo* pIter, int64_t key, SLastBlockReader* pLastBlockReader) { SRowMerger merge = {0}; STSRow* pTSRow = NULL; SBlockData* pBlockData = &pReader->status.fileBlockData; @@ -1331,12 +1583,159 @@ static int32_t doMergeBufAndFileRows(STsdbReader* pReader, STableBlockScanInfo* return TSDB_CODE_SUCCESS; } -static int32_t doMergeThreeLevelRows(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo) { +static int32_t doMergeMultiLevelRowsRv(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, SBlockData* pBlockData, SLastBlockReader* pLastBlockReader) { + SRowMerger merge = {0}; + STSRow* pTSRow = NULL; + + SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; + SArray* pDelList = pBlockScanInfo->delSkyline; + + TSDBROW* pRow = getValidRow(&pBlockScanInfo->iter, pDelList, pReader); + TSDBROW* piRow = getValidRow(&pBlockScanInfo->iiter, pDelList, pReader); + ASSERT(pRow != NULL && piRow != NULL); + + SBlockData* pLastBlockData = &pLastBlockReader->lastBlockData; + int64_t tsLast = getCurrentKeyInLastBlock(pLastBlockReader); + + int64_t key = pBlockData->aTSKEY[pDumpInfo->rowIndex]; + + TSDBKEY k = TSDBROW_KEY(pRow); + TSDBKEY ik = TSDBROW_KEY(piRow); + + int64_t minKey = 0;//INT64_MAX; + if (ASCENDING_TRAVERSE(pReader->order)) { + minKey = INT64_MAX; // let's find the minimum + if (minKey > k.ts) { + minKey = k.ts; + } + + if (minKey > ik.ts) { + minKey = ik.ts; + } + + if (minKey > key && pBlockData->nRow > 0) { + minKey = key; + } + + if (minKey > tsLast && pLastBlockData->nRow > 0) { + minKey = tsLast; + } + } else { + minKey = INT64_MIN; // let find the maximum ts value + if (minKey < k.ts) { + minKey = k.ts; + } + + if (minKey < ik.ts) { + minKey = ik.ts; + } + + if (minKey < key && pBlockData->nRow > 0) { + minKey = key; + } + + if (minKey < tsLast && pLastBlockData->nRow > 0) { + minKey = tsLast; + } + } + + bool init = false; + + // ASC: file block -----> last block -----> imem -----> mem + // DESC: mem -----> imem -----> last block -----> file block + if (ASCENDING_TRAVERSE(pReader->order)) { + if (minKey == key) { + init = true; + TSDBROW fRow = tsdbRowFromBlockData(pBlockData, pDumpInfo->rowIndex); + tRowMergerInit(&merge, &fRow, pReader->pSchema); + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + } + + if (minKey == tsLast) { + TSDBROW fRow1 = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + if (init) { + tRowMerge(&merge, &fRow1); + } else { + init = true; + tRowMergerInit(&merge, &fRow1, pReader->pSchema); + } + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, tsLast, &merge); + } + + if (minKey == ik.ts) { + if (init) { + tRowMerge(&merge, piRow); + } else { + init = true; + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(piRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, piRow, pSchema); + } + doMergeRowsInBuf(&pBlockScanInfo->iiter, pBlockScanInfo->uid, ik.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + + if (minKey == k.ts) { + if (init) { + tRowMerge(&merge, pRow); + } else { + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(pRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, pRow, pSchema); + } + doMergeRowsInBuf(&pBlockScanInfo->iter, pBlockScanInfo->uid, k.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + } else { + if (minKey == k.ts) { + init = true; + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(pRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, pRow, pSchema); + doMergeRowsInBuf(&pBlockScanInfo->iter, pBlockScanInfo->uid, k.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + + if (minKey == ik.ts) { + if (init) { + tRowMerge(&merge, piRow); + } else { + init = true; + STSchema* pSchema = doGetSchemaForTSRow(TSDBROW_SVERSION(piRow), pReader, pBlockScanInfo->uid); + tRowMergerInit(&merge, piRow, pSchema); + } + doMergeRowsInBuf(&pBlockScanInfo->iiter, pBlockScanInfo->uid, ik.ts, pBlockScanInfo->delSkyline, &merge, pReader); + } + + if (minKey == tsLast) { + TSDBROW fRow1 = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + if (init) { + tRowMerge(&merge, &fRow1); + } else { + init = true; + tRowMergerInit(&merge, &fRow1, pReader->pSchema); + } + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, tsLast, &merge); + } + + if (minKey == key) { + TSDBROW fRow = tsdbRowFromBlockData(pBlockData, pDumpInfo->rowIndex); + if (!init) { + tRowMergerInit(&merge, &fRow, pReader->pSchema); + } else { + tRowMerge(&merge, &fRow); + } + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + } + } + + tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; +} + +static int32_t doMergeThreeLevelRows(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, SBlockData* pBlockData) { SRowMerger merge = {0}; STSRow* pTSRow = NULL; SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; - SBlockData* pBlockData = &pReader->status.fileBlockData; SArray* pDelList = pBlockScanInfo->delSkyline; TSDBROW* pRow = getValidRow(&pBlockScanInfo->iter, pDelList, pReader); @@ -1477,6 +1876,14 @@ static int32_t doMergeThreeLevelRows(STsdbReader* pReader, STableBlockScanInfo* static bool isValidFileBlockRow(SBlockData* pBlockData, SFileBlockDumpInfo* pDumpInfo, STableBlockScanInfo* pBlockScanInfo, STsdbReader* pReader) { + // it is an multi-table data block + if (pBlockData->aUid != NULL) { + uint64_t uid = pBlockData->aUid[pDumpInfo->rowIndex]; + if (uid != pBlockScanInfo->uid) { // move to next row + return false; + } + } + // check for version and time range int64_t ver = pBlockData->aVersion[pDumpInfo->rowIndex]; if (ver > pReader->verRange.maxVer || ver < pReader->verRange.minVer) { @@ -1498,39 +1905,191 @@ static bool isValidFileBlockRow(SBlockData* pBlockData, SFileBlockDumpInfo* pDum static bool outOfTimeWindow(int64_t ts, STimeWindow* pWindow) { return (ts > pWindow->ekey) || (ts < pWindow->skey); } -static int32_t buildComposedDataBlockImpl(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo) { +static void initLastBlockReader(SLastBlockReader* pLastBlockReader, uint64_t uid, int16_t* startPos) { + pLastBlockReader->uid = uid; + pLastBlockReader->rowIndex = startPos; + + if (*startPos == -1) { + if (ASCENDING_TRAVERSE(pLastBlockReader->order)) { + // do nothing + } else { + *startPos = pLastBlockReader->lastBlockData.nRow; + } + } +} + +static void setAllRowsChecked(SLastBlockReader *pLastBlockReader) { + *pLastBlockReader->rowIndex = ALL_ROWS_CHECKED_INDEX; +} + +static bool nextRowInLastBlock(SLastBlockReader *pLastBlockReader, STableBlockScanInfo* pBlockScanInfo) { + int32_t step = (pLastBlockReader->order == TSDB_ORDER_ASC) ? 1 : -1; + if (*pLastBlockReader->rowIndex == ALL_ROWS_CHECKED_INDEX) { + return false; + } + + *(pLastBlockReader->rowIndex) += step; + + SBlockData* pBlockData = &pLastBlockReader->lastBlockData; + for(int32_t i = *(pLastBlockReader->rowIndex); i < pBlockData->nRow && i >= 0; i += step) { + if (pBlockData->aUid != NULL && pBlockData->aUid[i] != pLastBlockReader->uid) { + continue; + } + + int64_t ts = pBlockData->aTSKEY[i]; + if (ts < pLastBlockReader->window.skey) { + continue; + } + + int64_t ver = pBlockData->aVersion[i]; + if (ver < pLastBlockReader->verRange.minVer) { + continue; + } + + // no data any more, todo opt handle desc case + if (ts > pLastBlockReader->window.ekey) { + continue; + } + + // todo opt handle desc case + if (ver > pLastBlockReader->verRange.maxVer) { + continue; + } + + TSDBKEY k = {.ts = ts, .version = ver}; + if (hasBeenDropped(pBlockScanInfo->delSkyline, &pBlockScanInfo->lastBlockDelIndex, &k, pLastBlockReader->order)) { + continue; + } + + *(pLastBlockReader->rowIndex) = i; + return true; + } + + // set all data is consumed in last block + setAllRowsChecked(pLastBlockReader); + return false; +} + +static int64_t getCurrentKeyInLastBlock(SLastBlockReader* pLastBlockReader) { + SBlockData* pBlockData = &pLastBlockReader->lastBlockData; + return pBlockData->aTSKEY[*pLastBlockReader->rowIndex]; +} + +static bool hasDataInLastBlock(SLastBlockReader* pLastBlockReader) { + if (*pLastBlockReader->rowIndex == ALL_ROWS_CHECKED_INDEX) { + return false; + } + return true; +} + +// todo refactor +static int32_t buildComposedDataBlockImpl(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo, + SBlockData* pBlockData, SLastBlockReader* pLastBlockReader) { SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; - SBlockData* pBlockData = &pReader->status.fileBlockData; - int64_t key = pBlockData->aTSKEY[pDumpInfo->rowIndex]; + int64_t key = (pBlockData->nRow > 0)? pBlockData->aTSKEY[pDumpInfo->rowIndex]:INT64_MIN; TSDBROW* pRow = getValidRow(&pBlockScanInfo->iter, pBlockScanInfo->delSkyline, pReader); TSDBROW* piRow = getValidRow(&pBlockScanInfo->iiter, pBlockScanInfo->delSkyline, pReader); if (pBlockScanInfo->iter.hasVal && pBlockScanInfo->iiter.hasVal) { - return doMergeThreeLevelRows(pReader, pBlockScanInfo); + return doMergeMultiLevelRowsRv(pReader, pBlockScanInfo, pBlockData, pLastBlockReader); } else { - // imem + file + // imem + file + last block if (pBlockScanInfo->iiter.hasVal) { - return doMergeBufAndFileRows(pReader, pBlockScanInfo, piRow, &pBlockScanInfo->iiter, key); + return doMergeBufAndFileRows_Rv(pReader, pBlockScanInfo, piRow, &pBlockScanInfo->iiter, key, pLastBlockReader); } // mem + file if (pBlockScanInfo->iter.hasVal) { - return doMergeBufAndFileRows(pReader, pBlockScanInfo, pRow, &pBlockScanInfo->iter, key); + return doMergeBufAndFileRows_Rv(pReader, pBlockScanInfo, pRow, &pBlockScanInfo->iter, key, pLastBlockReader); } - // imem & mem are all empty, only file exist - if (tryCopyDistinctRowFromFileBlock(pReader, pBlockData, key, pDumpInfo)) { - return TSDB_CODE_SUCCESS; - } else { + if (pBlockData->nRow > 0) { TSDBROW fRow = tsdbRowFromBlockData(pBlockData, pDumpInfo->rowIndex); + // no last block + if (pLastBlockReader->lastBlockData.nRow == 0 || (!hasDataInLastBlock(pLastBlockReader))) { + if (tryCopyDistinctRowFromFileBlock(pReader, pBlockData, key, pDumpInfo)) { + return TSDB_CODE_SUCCESS; + } else { + STSRow* pTSRow = NULL; + SRowMerger merge = {0}; + + tRowMergerInit(&merge, &fRow, pReader->pSchema); + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; + } + } + + // row in last file block + int64_t ts = getCurrentKeyInLastBlock(pLastBlockReader); + if (ts < key) { // save rows in last block + SBlockData* pLastBlockData = &pLastBlockReader->lastBlockData; + + STSRow* pTSRow = NULL; + SRowMerger merge = {0}; + + TSDBROW fRow1 = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + + tRowMergerInit(&merge, &fRow1, pReader->pSchema); + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, ts, &merge); + tRowMergerGetRow(&merge, &pTSRow); + + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; + } else if (ts == key) { + STSRow* pTSRow = NULL; + SRowMerger merge = {0}; + + tRowMergerInit(&merge, &fRow, pReader->pSchema); + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, ts, &merge); + + tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; + } else { // ts > key, asc; todo handle desc + // imem & mem are all empty, only file exist + if (tryCopyDistinctRowFromFileBlock(pReader, pBlockData, key, pDumpInfo)) { + return TSDB_CODE_SUCCESS; + } else { + STSRow* pTSRow = NULL; + SRowMerger merge = {0}; + + tRowMergerInit(&merge, &fRow, pReader->pSchema); + doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); + + taosMemoryFree(pTSRow); + tRowMergerClear(&merge); + return TSDB_CODE_SUCCESS; + } + } + } else { // only last block exists + SBlockData* pLastBlockData = &pLastBlockReader->lastBlockData; + int64_t tsLastBlock = getCurrentKeyInLastBlock(pLastBlockReader); + STSRow* pTSRow = NULL; SRowMerger merge = {0}; + TSDBROW fRow = tsdbRowFromBlockData(pLastBlockData, *pLastBlockReader->rowIndex); + tRowMergerInit(&merge, &fRow, pReader->pSchema); - doMergeRowsInFileBlocks(pBlockData, pBlockScanInfo, pReader, &merge); + doMergeRowsInLastBlock(pLastBlockReader, pBlockScanInfo, tsLastBlock, &merge); tRowMergerGetRow(&merge, &pTSRow); + doAppendRowFromTSRow(pReader->pResBlock, pReader, pTSRow, pBlockScanInfo->uid); taosMemoryFree(pTSRow); @@ -1540,41 +2099,59 @@ static int32_t buildComposedDataBlockImpl(STsdbReader* pReader, STableBlockScanI } } -static int32_t buildComposedDataBlock(STsdbReader* pReader, STableBlockScanInfo* pBlockScanInfo) { +static int32_t buildComposedDataBlock(STsdbReader* pReader) { SSDataBlock* pResBlock = pReader->pResBlock; + SFileDataBlockInfo* pBlockInfo = getCurrentBlockInfo(&pReader->status.blockIter); + + STableBlockScanInfo* pBlockScanInfo = NULL; + if (pBlockInfo != NULL) { + pBlockScanInfo = taosHashGet(pReader->status.pTableMap, &pBlockInfo->uid, sizeof(pBlockInfo->uid)); + } else { + pBlockScanInfo = pReader->status.pTableIter; + } + + SLastBlockReader* pLastBlockReader = pReader->status.fileIter.pLastBlockReader; SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; SBlockData* pBlockData = &pReader->status.fileBlockData; int32_t step = ASCENDING_TRAVERSE(pReader->order) ? 1 : -1; - int32_t numOfSub = 1; - int64_t st = taosGetTimestampUs(); while (1) { // todo check the validate of row in file block { - if (!isValidFileBlockRow(pBlockData, pDumpInfo, pBlockScanInfo, pReader)) { + bool hasBlockData = false; + + while (pBlockData->nRow > 0) { // find the first qualified row in data block + if (isValidFileBlockRow(pBlockData, pDumpInfo, pBlockScanInfo, pReader)) { + hasBlockData = true; + break; + } + pDumpInfo->rowIndex += step; SBlock* pBlock = getCurrentBlock(&pReader->status.blockIter); - numOfSub = pBlock->nSubBlock; - if (pDumpInfo->rowIndex >= pBlock->nRow || pDumpInfo->rowIndex < 0) { - setBlockAllDumped(pDumpInfo, pBlock, pReader->order); + setBlockAllDumped(pDumpInfo, pBlock->maxKey.ts, pReader->order); break; } + } - continue; + bool hasBlockLData = hasDataInLastBlock(pLastBlockReader); + + // no data in last block and block, no need to proceed. + if ((hasBlockData == false) && (hasBlockLData == false)) { + break; } } - buildComposedDataBlockImpl(pReader, pBlockScanInfo); - SBlock* pBlock = getCurrentBlock(&pReader->status.blockIter); + buildComposedDataBlockImpl(pReader, pBlockScanInfo, pBlockData, pLastBlockReader); // currently loaded file data block is consumed - if (pDumpInfo->rowIndex >= pBlock->nRow || pDumpInfo->rowIndex < 0) { - setBlockAllDumped(pDumpInfo, pBlock, pReader->order); + if ((pBlockData->nRow > 0) && (pDumpInfo->rowIndex >= pBlockData->nRow || pDumpInfo->rowIndex < 0)) { + SBlock* pBlock = getCurrentBlock(&pReader->status.blockIter); + setBlockAllDumped(pDumpInfo, pBlock->maxKey.ts, pReader->order); break; } @@ -1589,9 +2166,8 @@ static int32_t buildComposedDataBlock(STsdbReader* pReader, STableBlockScanInfo* setComposedBlockFlag(pReader, true); int64_t et = taosGetTimestampUs(); - tsdbDebug("%p uid:%" PRIu64 ", composed data block created, subBlock:%d, brange:%" PRIu64 "-%" PRIu64 - " rows:%d, elapsed time:%.2f ms %s", - pReader, pBlockScanInfo->uid, numOfSub, pResBlock->info.window.skey, pResBlock->info.window.ekey, + tsdbDebug("%p uid:%" PRIu64 ", composed data block created, brange:%" PRIu64 "-%" PRIu64 " rows:%d, elapsed time:%.2f ms %s", + pReader, pBlockScanInfo->uid, pResBlock->info.window.skey, pResBlock->info.window.ekey, pResBlock->info.rows, (et - st) / 1000.0, pReader->idStr); return TSDB_CODE_SUCCESS; @@ -1617,7 +2193,7 @@ static int32_t initMemDataIterator(STableBlockScanInfo* pBlockScanInfo, STsdbRea STbData* d = NULL; if (pReader->pReadSnap->pMem != NULL) { - tsdbGetTbDataFromMemTable(pReader->pReadSnap->pMem, pReader->suid, pBlockScanInfo->uid, &d); + d = tsdbGetTbDataFromMemTable(pReader->pReadSnap->pMem, pReader->suid, pBlockScanInfo->uid); if (d != NULL) { code = tsdbTbDataIterCreate(d, &startKey, backward, &pBlockScanInfo->iter.iter); if (code == TSDB_CODE_SUCCESS) { @@ -1638,7 +2214,7 @@ static int32_t initMemDataIterator(STableBlockScanInfo* pBlockScanInfo, STsdbRea STbData* di = NULL; if (pReader->pReadSnap->pIMem != NULL) { - tsdbGetTbDataFromMemTable(pReader->pReadSnap->pIMem, pReader->suid, pBlockScanInfo->uid, &di); + di = tsdbGetTbDataFromMemTable(pReader->pReadSnap->pIMem, pReader->suid, pBlockScanInfo->uid); if (di != NULL) { code = tsdbTbDataIterCreate(di, &startKey, backward, &pBlockScanInfo->iiter.iter); if (code == TSDB_CODE_SUCCESS) { @@ -1677,7 +2253,7 @@ int32_t initDelSkylineIterator(STableBlockScanInfo* pBlockScanInfo, STsdbReader* SDelFile* pDelFile = pReader->pReadSnap->fs.pDelFile; if (pDelFile) { SDelFReader* pDelFReader = NULL; - code = tsdbDelFReaderOpen(&pDelFReader, pDelFile, pTsdb, NULL); + code = tsdbDelFReaderOpen(&pDelFReader, pDelFile, pTsdb); if (code != TSDB_CODE_SUCCESS) { goto _err; } @@ -1688,7 +2264,7 @@ int32_t initDelSkylineIterator(STableBlockScanInfo* pBlockScanInfo, STsdbReader* goto _err; } - code = tsdbReadDelIdx(pDelFReader, aDelIdx, NULL); + code = tsdbReadDelIdx(pDelFReader, aDelIdx); if (code != TSDB_CODE_SUCCESS) { taosArrayDestroy(aDelIdx); tsdbDelFReaderClose(&pDelFReader); @@ -1699,7 +2275,7 @@ int32_t initDelSkylineIterator(STableBlockScanInfo* pBlockScanInfo, STsdbReader* SDelIdx* pIdx = taosArraySearch(aDelIdx, &idx, tCmprDelIdx, TD_EQ); if (pIdx != NULL) { - code = tsdbReadDelData(pDelFReader, pIdx, pDelData, NULL); + code = tsdbReadDelData(pDelFReader, pIdx, pDelData); } taosArrayDestroy(aDelIdx); @@ -1737,6 +2313,7 @@ int32_t initDelSkylineIterator(STableBlockScanInfo* pBlockScanInfo, STsdbReader* ASCENDING_TRAVERSE(pReader->order) ? 0 : taosArrayGetSize(pBlockScanInfo->delSkyline) - 1; pBlockScanInfo->iiter.index = pBlockScanInfo->iter.index; pBlockScanInfo->fileDelIndex = pBlockScanInfo->iter.index; + pBlockScanInfo->lastBlockDelIndex = pBlockScanInfo->iter.index; return code; _err: @@ -1744,12 +2321,9 @@ _err: return code; } -static TSDBKEY getCurrentKeyInBuf(SDataBlockIter* pBlockIter, STsdbReader* pReader) { +static TSDBKEY getCurrentKeyInBuf(STableBlockScanInfo* pScanInfo, STsdbReader* pReader) { TSDBKEY key = {.ts = TSKEY_INITIAL_VAL}; - SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(pBlockIter); - STableBlockScanInfo* pScanInfo = taosHashGet(pReader->status.pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); - initMemDataIterator(pScanInfo, pReader); TSDBROW* pRow = getValidRow(&pScanInfo->iter, pScanInfo->delSkyline, pReader); if (pRow != NULL) { @@ -1767,11 +2341,15 @@ static TSDBKEY getCurrentKeyInBuf(SDataBlockIter* pBlockIter, STsdbReader* pRead return key; } -static int32_t moveToNextFile(STsdbReader* pReader, int32_t* numOfBlocks) { +static int32_t moveToNextFile(STsdbReader* pReader, SBlockNumber* pBlockNum) { SReaderStatus* pStatus = &pReader->status; + pBlockNum->numOfBlocks = 0; + pBlockNum->numOfLastBlocks = 0; size_t numOfTables = taosHashGetSize(pReader->status.pTableMap); SArray* pIndexList = taosArrayInit(numOfTables, sizeof(SBlockIdx)); + SArray* pLastBlocks = pStatus->fileIter.pLastBlockReader->pBlockL; + taosArrayClear(pLastBlocks); while (1) { bool hasNext = filesetIteratorNext(&pStatus->fileIter, pReader); @@ -1786,18 +2364,34 @@ static int32_t moveToNextFile(STsdbReader* pReader, int32_t* numOfBlocks) { return code; } - if (taosArrayGetSize(pIndexList) > 0) { - uint32_t numOfValidTable = 0; - code = doLoadFileBlock(pReader, pIndexList, &numOfValidTable, numOfBlocks); + code = tsdbReadBlockL(pReader->pFileReader, pLastBlocks); + if (code != TSDB_CODE_SUCCESS) { + taosArrayDestroy(pIndexList); + return code; + } + + if (taosArrayGetSize(pIndexList) > 0 || taosArrayGetSize(pLastBlocks) > 0) { + SArray* pQLastBlock = taosArrayInit(4, sizeof(SBlockL)); + + code = doLoadFileBlock(pReader, pIndexList, pLastBlocks, pBlockNum, pQLastBlock); if (code != TSDB_CODE_SUCCESS) { taosArrayDestroy(pIndexList); + taosArrayDestroy(pQLastBlock); return code; } - if (numOfValidTable > 0) { + if (pBlockNum->numOfBlocks + pBlockNum->numOfLastBlocks > 0) { + ASSERT(taosArrayGetSize(pQLastBlock) == pBlockNum->numOfLastBlocks); + taosArrayClear(pLastBlocks); + taosArrayAddAll(pLastBlocks, pQLastBlock); + + taosArrayDestroy(pQLastBlock); break; } + + taosArrayDestroy(pQLastBlock); } + // no blocks in current file, try next files } @@ -1805,28 +2399,177 @@ static int32_t moveToNextFile(STsdbReader* pReader, int32_t* numOfBlocks) { return TSDB_CODE_SUCCESS; } +static int32_t doLoadRelatedLastBlock(SLastBlockReader* pLastBlockReader, STableBlockScanInfo *pBlockScanInfo, STsdbReader* pReader) { + SArray* pBlocks = pLastBlockReader->pBlockL; + SBlockL* pBlock = NULL; + + uint64_t uid = pBlockScanInfo->uid; + int32_t totalLastBlocks = (int32_t)taosArrayGetSize(pBlocks); + + initMemDataIterator(pBlockScanInfo, pReader); + + // find the correct SBlockL. todo binary search + int32_t index = -1; + for (int32_t i = 0; i < totalLastBlocks; ++i) { + SBlockL* p = taosArrayGet(pBlocks, i); + if (p->minUid <= uid && p->maxUid >= uid) { + index = i; + pBlock = p; + break; + } + } + + if (index == -1) { + pLastBlockReader->currentBlockIndex = index; + tBlockDataReset(&pLastBlockReader->lastBlockData); + return TSDB_CODE_SUCCESS; + } + + // the required last datablock has already loaded + if (index == pLastBlockReader->currentBlockIndex) { + return TSDB_CODE_SUCCESS; + } + + int64_t st = taosGetTimestampUs(); + int32_t code = tBlockDataInit(&pLastBlockReader->lastBlockData, pReader->suid, pReader->suid ? 0 : uid, pReader->pSchema); + if (code != TSDB_CODE_SUCCESS) { + tsdbError("%p init block data failed, code:%s %s", pReader, tstrerror(code), pReader->idStr); + return code; + } + + code = tsdbReadLastBlock(pReader->pFileReader, pBlock, &pLastBlockReader->lastBlockData); + + double el = (taosGetTimestampUs() - st) / 1000.0; + if (code != TSDB_CODE_SUCCESS) { + tsdbError("%p error occurs in loading last block into buffer, last block index:%d, total:%d code:%s %s", pReader, + pLastBlockReader->currentBlockIndex, totalLastBlocks, tstrerror(code), pReader->idStr); + } else { + tsdbDebug("%p load last block completed, uid:%" PRIu64 + " last block index:%d, total:%d rows:%d, minVer:%d, maxVer:%d, brange:%" PRId64 " - %" PRId64 + " elapsed time:%.2f ms, %s", + pReader, uid, pLastBlockReader->currentBlockIndex, totalLastBlocks, pBlock->nRow, pBlock->minVer, + pBlock->maxVer, pBlock->minKey, pBlock->maxKey, el, pReader->idStr); + } + + pLastBlockReader->currentBlockIndex = index; + pReader->cost.lastBlockLoad += 1; + pReader->cost.lastBlockLoadTime += el; + + return TSDB_CODE_SUCCESS; +} + +static int32_t doLoadLastBlockSequentially(STsdbReader* pReader) { + SReaderStatus* pStatus = &pReader->status; + SLastBlockReader* pLastBlockReader = pStatus->fileIter.pLastBlockReader; + + while(1) { + if (pStatus->pTableIter == NULL) { + pStatus->pTableIter = taosHashIterate(pStatus->pTableMap, NULL); + if (pStatus->pTableIter == NULL) { + return TSDB_CODE_SUCCESS; + } + } + + // load the last data block of current table + // todo opt perf by avoiding load last block repeatly + STableBlockScanInfo* pScanInfo = pStatus->pTableIter; + int32_t code = doLoadRelatedLastBlock(pLastBlockReader, pScanInfo, pReader); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + if (pLastBlockReader->currentBlockIndex != -1) { + initLastBlockReader(pLastBlockReader, pScanInfo->uid, &pScanInfo->indexInBlockL); + int32_t index = pScanInfo->indexInBlockL; + if (index == DEFAULT_ROW_INDEX_VAL || index == pLastBlockReader->lastBlockData.nRow) { + bool hasData = nextRowInLastBlock(pLastBlockReader, pScanInfo); + if (!hasData) { // current table does not have rows in last block, try next table + pStatus->pTableIter = taosHashIterate(pStatus->pTableMap, pStatus->pTableIter); + if (pStatus->pTableIter == NULL) { + return TSDB_CODE_SUCCESS; + } + continue; + } + } + } else { // no data in last block, try next table + pStatus->pTableIter = taosHashIterate(pStatus->pTableMap, pStatus->pTableIter); + if (pStatus->pTableIter == NULL) { + return TSDB_CODE_SUCCESS; + } + continue; + } + + code = doBuildDataBlock(pReader); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + if (pReader->pResBlock->info.rows > 0) { + return TSDB_CODE_SUCCESS; + } + + // current table is exhausted, let's try next table + pStatus->pTableIter = taosHashIterate(pStatus->pTableMap, pStatus->pTableIter); + if (pStatus->pTableIter == NULL) { + return TSDB_CODE_SUCCESS; + } + } +} + static int32_t doBuildDataBlock(STsdbReader* pReader) { + TSDBKEY key = {0}; int32_t code = TSDB_CODE_SUCCESS; + SBlock* pBlock = NULL; - SReaderStatus* pStatus = &pReader->status; - SDataBlockIter* pBlockIter = &pStatus->blockIter; + SReaderStatus* pStatus = &pReader->status; + SDataBlockIter* pBlockIter = &pStatus->blockIter; + STableBlockScanInfo* pScanInfo = NULL; + SFileDataBlockInfo* pBlockInfo = getCurrentBlockInfo(pBlockIter); + SLastBlockReader* pLastBlockReader = pReader->status.fileIter.pLastBlockReader; - SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(pBlockIter); - STableBlockScanInfo* pScanInfo = taosHashGet(pStatus->pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); + if (pBlockInfo != NULL) { + pScanInfo = taosHashGet(pReader->status.pTableMap, &pBlockInfo->uid, sizeof(pBlockInfo->uid)); + } else { + pScanInfo = pReader->status.pTableIter; + } - SBlock* pBlock = getCurrentBlock(pBlockIter); + if (pBlockInfo != NULL) { + pBlock = getCurrentBlock(pBlockIter); + } + + { + key = getCurrentKeyInBuf(pScanInfo, pReader); - TSDBKEY key = getCurrentKeyInBuf(pBlockIter, pReader); - if (fileBlockShouldLoad(pReader, pFBlock, pBlock, pScanInfo, key)) { + // load the last data block of current table + code = doLoadRelatedLastBlock(pLastBlockReader, pScanInfo, pReader); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + // note: the lastblock may be null here + initLastBlockReader(pLastBlockReader, pScanInfo->uid, &pScanInfo->indexInBlockL); + if (pScanInfo->indexInBlockL == DEFAULT_ROW_INDEX_VAL || pScanInfo->indexInBlockL == pLastBlockReader->lastBlockData.nRow) { + bool hasData = nextRowInLastBlock(pLastBlockReader, pScanInfo); + } + } + + if (pBlockInfo == NULL) { // build data block from last data file + ASSERT(pBlockIter->numOfBlocks == 0); + code = buildComposedDataBlock(pReader); + } else if (fileBlockShouldLoad(pReader, pBlockInfo, pBlock, pScanInfo, key, pLastBlockReader)) { tBlockDataReset(&pStatus->fileBlockData); - tBlockDataClearData(&pStatus->fileBlockData); - code = doLoadFileBlockData(pReader, pBlockIter, pScanInfo, &pStatus->fileBlockData); + code = tBlockDataInit(&pStatus->fileBlockData, pReader->suid, pScanInfo->uid, pReader->pSchema); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + code = doLoadFileBlockData(pReader, pBlockIter, &pStatus->fileBlockData); if (code != TSDB_CODE_SUCCESS) { return code; } // build composed data block - code = buildComposedDataBlock(pReader, pScanInfo); + code = buildComposedDataBlock(pReader); } else if (bufferDataInFileBlockGap(pReader->order, key, pBlock)) { // data in memory that are earlier than current file block // todo rows in buffer should be less than the file block in asc, greater than file block in desc @@ -1838,7 +2581,7 @@ static int32_t doBuildDataBlock(STsdbReader* pReader) { pInfo->uid = pScanInfo->uid; pInfo->window = (STimeWindow){.skey = pBlock->minKey.ts, .ekey = pBlock->maxKey.ts}; setComposedBlockFlag(pReader, false); - setBlockAllDumped(&pStatus->fBlockDumpInfo, pBlock, pReader->order); + setBlockAllDumped(&pStatus->fBlockDumpInfo, pBlock->maxKey.ts, pReader->order); } return code; @@ -1890,20 +2633,29 @@ static void initBlockDumpInfo(STsdbReader* pReader, SDataBlockIter* pBlockIter) } static int32_t initForFirstBlockInFile(STsdbReader* pReader, SDataBlockIter* pBlockIter) { - int32_t numOfBlocks = 0; - int32_t code = moveToNextFile(pReader, &numOfBlocks); + SBlockNumber num = {0}; + + int32_t code = moveToNextFile(pReader, &num); if (code != TSDB_CODE_SUCCESS) { return code; } // all data files are consumed, try data in buffer - if (numOfBlocks == 0) { + if (num.numOfBlocks + num.numOfLastBlocks == 0) { pReader->status.loadFromFile = false; return code; } // initialize the block iterator for a new fileset - code = initBlockIterator(pReader, pBlockIter, numOfBlocks); + if (num.numOfBlocks > 0) { + code = initBlockIterator(pReader, pBlockIter, num.numOfBlocks); + } else { // no block data, only last block exists + tBlockDataReset(&pReader->status.fileBlockData); + resetDataBlockIterator(pBlockIter, pReader->order, pReader->status.pTableMap); + } + + SLastBlockReader* pLReader = pReader->status.fileIter.pLastBlockReader; + pLReader->currentBlockIndex = -1; // set the correct start position according to the query time window initBlockDumpInfo(pReader, pBlockIter); @@ -1921,14 +2673,47 @@ static int32_t buildBlockFromFiles(STsdbReader* pReader) { SDataBlockIter* pBlockIter = &pReader->status.blockIter; - while (1) { - SFileDataBlockInfo* pFBlock = getCurrentBlockInfo(&pReader->status.blockIter); - STableBlockScanInfo* pScanInfo = taosHashGet(pReader->status.pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); + if (pBlockIter->numOfBlocks == 0) { + _begin: + code = doLoadLastBlockSequentially(pReader); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + if (pReader->pResBlock->info.rows > 0) { + return TSDB_CODE_SUCCESS; + } + + // all data blocks are checked in this last block file, now let's try the next file + if (pReader->status.pTableIter == NULL) { + code = initForFirstBlockInFile(pReader, pBlockIter); + + // error happens or all the data files are completely checked + if ((code != TSDB_CODE_SUCCESS) || (pReader->status.loadFromFile == false)) { + return code; + } + + // this file does not have data files, let's start check the last block file if exists + if (pBlockIter->numOfBlocks == 0) { + goto _begin; + } + } + + code = doBuildDataBlock(pReader); + if (code != TSDB_CODE_SUCCESS) { + return code; + } + + if (pReader->pResBlock->info.rows > 0) { + return TSDB_CODE_SUCCESS; + } + } + while (1) { SFileBlockDumpInfo* pDumpInfo = &pReader->status.fBlockDumpInfo; if (fileBlockPartiallyRead(pDumpInfo, asc)) { // file data block is partially loaded - code = buildComposedDataBlock(pReader, pScanInfo); + code = buildComposedDataBlock(pReader); } else { // current block are exhausted, try the next file block if (pDumpInfo->allDumped) { @@ -1936,17 +2721,25 @@ static int32_t buildBlockFromFiles(STsdbReader* pReader) { bool hasNext = blockIteratorNext(&pReader->status.blockIter); if (hasNext) { // check for the next block in the block accessed order list initBlockDumpInfo(pReader, pBlockIter); - } else { // data blocks in current file are exhausted, let's try the next file now + } else if (taosArrayGetSize(pReader->status.fileIter.pLastBlockReader->pBlockL) > 0) { // data blocks in current file are exhausted, let's try the next file now + tBlockDataReset(&pReader->status.fileBlockData); + resetDataBlockIterator(pBlockIter, pReader->order, pReader->status.pTableMap); + goto _begin; + } else { code = initForFirstBlockInFile(pReader, pBlockIter); // error happens or all the data files are completely checked if ((code != TSDB_CODE_SUCCESS) || (pReader->status.loadFromFile == false)) { return code; } + + // this file does not have blocks, let's start check the last block file + if (pBlockIter->numOfBlocks == 0) { + goto _begin; + } } } - // current block is not loaded yet, or data in buffer may overlap with the file block. code = doBuildDataBlock(pReader); } @@ -2014,39 +2807,6 @@ SVersionRange getQueryVerRange(SVnode* pVnode, SQueryTableDataCond* pCond, int8_ return (SVersionRange){.minVer = startVer, .maxVer = endVer}; } -// // todo not unref yet, since it is not support multi-group interpolation query -// static UNUSED_FUNC void changeQueryHandleForInterpQuery(STsdbReader* pHandle) { -// // filter the queried time stamp in the first place -// STsdbReader* pTsdbReadHandle = (STsdbReader*)pHandle; - -// // starts from the buffer in case of descending timestamp order check data blocks -// size_t numOfTables = taosArrayGetSize(pTsdbReadHandle->pTableCheckInfo); - -// int32_t i = 0; -// while (i < numOfTables) { -// STableBlockScanInfo* pCheckInfo = taosArrayGet(pTsdbReadHandle->pTableCheckInfo, i); - -// // the first qualified table for interpolation query -// // if ((pTsdbReadHandle->window.skey <= pCheckInfo->pTableObj->lastKey) && -// // (pCheckInfo->pTableObj->lastKey != TSKEY_INITIAL_VAL)) { -// // break; -// // } - -// i++; -// } - -// // there are no data in all the tables -// if (i == numOfTables) { -// return; -// } - -// STableBlockScanInfo info = *(STableBlockScanInfo*)taosArrayGet(pTsdbReadHandle->pTableCheckInfo, i); -// taosArrayClear(pTsdbReadHandle->pTableCheckInfo); - -// info.lastKey = pTsdbReadHandle->window.skey; -// taosArrayPush(pTsdbReadHandle->pTableCheckInfo, &info); -// } - bool hasBeenDropped(const SArray* pDelList, int32_t* index, TSDBKEY* pKey, int32_t order) { ASSERT(pKey != NULL); if (pDelList == NULL) { @@ -2265,8 +3025,7 @@ static int32_t checkForNeighborFileBlock(STsdbReader* pReader, STableBlockScanIn // 3. load the neighbor block, and set it to be the currently accessed file data block tBlockDataReset(&pStatus->fileBlockData); - tBlockDataClearData(&pStatus->fileBlockData); - int32_t code = doLoadFileBlockData(pReader, pBlockIter, pScanInfo, &pStatus->fileBlockData); + int32_t code = doLoadFileBlockData(pReader, pBlockIter, &pStatus->fileBlockData); if (code != TSDB_CODE_SUCCESS) { return code; } @@ -2315,6 +3074,21 @@ int32_t doMergeRowsInFileBlocks(SBlockData* pBlockData, STableBlockScanInfo* pSc return TSDB_CODE_SUCCESS; } +// todo check if the rows are dropped or not +int32_t doMergeRowsInLastBlock(SLastBlockReader* pLastBlockReader, STableBlockScanInfo* pScanInfo, int64_t ts, SRowMerger* pMerger) { + while(nextRowInLastBlock(pLastBlockReader, pScanInfo)) { + int64_t next1 = getCurrentKeyInLastBlock(pLastBlockReader); + if (next1 == ts) { + TSDBROW fRow1 = tsdbRowFromBlockData(&pLastBlockReader->lastBlockData, *pLastBlockReader->rowIndex); + tRowMerge(pMerger, &fRow1); + } else { + break; + } + } + + return TSDB_CODE_SUCCESS; +} + void doMergeMultiRows(TSDBROW* pRow, uint64_t uid, SIterInfo* pIter, SArray* pDelList, STSRow** pTSRow, STsdbReader* pReader, bool* freeTSRow) { TSDBROW* pNextRow = NULL; @@ -2487,7 +3261,7 @@ int32_t doAppendRowFromTSRow(SSDataBlock* pBlock, STsdbReader* pReader, STSRow* return TSDB_CODE_SUCCESS; } -int32_t doAppendRowFromBlock(SSDataBlock* pResBlock, STsdbReader* pReader, SBlockData* pBlockData, int32_t rowIndex) { +int32_t doAppendRowFromFileBlock(SSDataBlock* pResBlock, STsdbReader* pReader, SBlockData* pBlockData, int32_t rowIndex) { int32_t i = 0, j = 0; int32_t outputRowIndex = pResBlock->info.rows; @@ -2564,7 +3338,7 @@ int32_t tsdbSetTableId(STsdbReader* pReader, int64_t uid) { ASSERT(pReader != NULL); taosHashClear(pReader->status.pTableMap); - STableBlockScanInfo info = {.lastKey = 0, .uid = uid}; + STableBlockScanInfo info = {.lastKey = 0, .uid = uid, .indexInBlockL = DEFAULT_ROW_INDEX_VAL}; taosHashPut(pReader->status.pTableMap, &info.uid, sizeof(uint64_t), &info, sizeof(info)); return TDB_CODE_SUCCESS; } @@ -2615,6 +3389,7 @@ int32_t tsdbReaderOpen(SVnode* pVnode, SQueryTableDataCond* pCond, SArray* pTabl pCond->order = TSDB_ORDER_ASC; } + // here we only need one more row, so the capacity is set to be ONE. code = tsdbReaderCreate(pVnode, pCond, &pReader->innerReader[0], 1, idstr); if (code != TSDB_CODE_SUCCESS) { goto _err; @@ -2658,7 +3433,7 @@ int32_t tsdbReaderOpen(SVnode* pVnode, SQueryTableDataCond* pCond, SArray* pTabl if (pReader->type == TIMEWINDOW_RANGE_CONTAINED) { SDataBlockIter* pBlockIter = &pReader->status.blockIter; - initFilesetIterator(&pReader->status.fileIter, pReader->pReadSnap->fs.aDFileSet, pReader->order, pReader->idStr); + initFilesetIterator(&pReader->status.fileIter, pReader->pReadSnap->fs.aDFileSet, pReader); resetDataBlockIterator(&pReader->status.blockIter, pReader->order, pReader->status.pTableMap); // no data in files, let's try buffer in memory @@ -2679,8 +3454,7 @@ int32_t tsdbReaderOpen(SVnode* pVnode, SQueryTableDataCond* pCond, SArray* pTabl goto _err; } - initFilesetIterator(&pPrevReader->status.fileIter, pPrevReader->pReadSnap->fs.aDFileSet, pPrevReader->order, - pPrevReader->idStr); + initFilesetIterator(&pPrevReader->status.fileIter, pPrevReader->pReadSnap->fs.aDFileSet, pPrevReader); resetDataBlockIterator(&pPrevReader->status.blockIter, pPrevReader->order, pReader->status.pTableMap); // no data in files, let's try buffer in memory @@ -2720,7 +3494,7 @@ void tsdbReaderClose(STsdbReader* pReader) { } } taosMemoryFree(pSupInfo->buildBuf); - tBlockDataClear(&pReader->status.fileBlockData, true); + tBlockDataDestroy(&pReader->status.fileBlockData, true); cleanupDataBlockIterator(&pReader->status.blockIter); @@ -2732,15 +3506,23 @@ void tsdbReaderClose(STsdbReader* pReader) { tsdbDataFReaderClose(&pReader->pFileReader); } + SFilesetIter* pFilesetIter = &pReader->status.fileIter; + if (pFilesetIter->pLastBlockReader != NULL) { + tBlockDataDestroy(&pFilesetIter->pLastBlockReader->lastBlockData, true); + taosArrayDestroy(pFilesetIter->pLastBlockReader->pBlockL); + taosMemoryFree(pFilesetIter->pLastBlockReader); + } + SIOCostSummary* pCost = &pReader->cost; tsdbDebug("%p :io-cost summary: head-file:%" PRIu64 ", head-file time:%.2f ms, SMA:%" PRId64 " SMA-time:%.2f ms, fileBlocks:%" PRId64 ", fileBlocks-time:%.2f ms, " - "build in-memory-block-time:%.2f ms, STableBlockScanInfo size:%.2f Kb %s", - pReader, pCost->headFileLoad, pCost->headFileLoadTime, pCost->smaData, pCost->smaLoadTime, - pCost->numOfBlocks, pCost->blockLoadTime, pCost->buildmemBlock, - numOfTables * sizeof(STableBlockScanInfo) / 1000.0, pReader->idStr); + "build in-memory-block-time:%.2f ms, lastBlocks:%" PRId64 + ", lastBlocks-time:%.2f ms, STableBlockScanInfo size:%.2f Kb %s", + pReader, pCost->headFileLoad, pCost->headFileLoadTime, pCost->smaDataLoad, pCost->smaLoadTime, + pCost->numOfBlocks, pCost->blockLoadTime, pCost->buildmemBlock, pCost->lastBlockLoad, + pCost->lastBlockLoadTime, numOfTables * sizeof(STableBlockScanInfo) / 1000.0, pReader->idStr); taosMemoryFree(pReader->idStr); taosMemoryFree(pReader->pSchema); @@ -2857,7 +3639,7 @@ int32_t tsdbRetrieveDatablockSMA(STsdbReader* pReader, SColumnDataAgg*** pBlockS SBlockLoadSuppInfo* pSup = &pReader->suppInfo; if (tBlockHasSma(pBlock)) { - code = tsdbReadBlockSma(pReader->pFileReader, pBlock, pSup->pColAgg, NULL); + code = tsdbReadBlockSma(pReader->pFileReader, pBlock, pSup->pColAgg); if (code != TSDB_CODE_SUCCESS) { tsdbDebug("vgId:%d, failed to load block SMA for uid %" PRIu64 ", code:%s, %s", 0, pFBlock->uid, tstrerror(code), pReader->idStr); @@ -2902,7 +3684,7 @@ int32_t tsdbRetrieveDatablockSMA(STsdbReader* pReader, SColumnDataAgg*** pBlockS double elapsed = (taosGetTimestampUs() - stime) / 1000.0; pReader->cost.smaLoadTime += elapsed; - pReader->cost.smaData += 1; + pReader->cost.smaDataLoad += 1; *pBlockStatis = pSup->plist; @@ -2923,11 +3705,15 @@ static SArray* doRetrieveDataBlock(STsdbReader* pReader) { STableBlockScanInfo* pBlockScanInfo = taosHashGet(pStatus->pTableMap, &pFBlock->uid, sizeof(pFBlock->uid)); tBlockDataReset(&pStatus->fileBlockData); - tBlockDataClearData(&pStatus->fileBlockData); - int32_t code = doLoadFileBlockData(pReader, &pStatus->blockIter, pBlockScanInfo, &pStatus->fileBlockData); + int32_t code = tBlockDataInit(&pStatus->fileBlockData, pReader->suid, pBlockScanInfo->uid, pReader->pSchema); if (code != TSDB_CODE_SUCCESS) { - tBlockDataClear(&pStatus->fileBlockData, 1); + terrno = code; + return NULL; + } + code = doLoadFileBlockData(pReader, &pStatus->blockIter, &pStatus->fileBlockData); + if (code != TSDB_CODE_SUCCESS) { + tBlockDataDestroy(&pStatus->fileBlockData, 1); terrno = code; return NULL; } @@ -2969,7 +3755,7 @@ int32_t tsdbReaderReset(STsdbReader* pReader, SQueryTableDataCond* pCond) { int32_t numOfTables = taosHashGetSize(pReader->status.pTableMap); tsdbDataFReaderClose(&pReader->pFileReader); - initFilesetIterator(&pReader->status.fileIter, pReader->pReadSnap->fs.aDFileSet, pReader->order, pReader->idStr); + initFilesetIterator(&pReader->status.fileIter, pReader->pReadSnap->fs.aDFileSet, pReader); resetDataBlockIterator(&pReader->status.blockIter, pReader->order, pReader->status.pTableMap); resetDataBlockScanInfo(pReader->status.pTableMap); @@ -3078,7 +3864,7 @@ int64_t tsdbGetNumOfRowsInMemTable(STsdbReader* pReader) { STbData* d = NULL; if (pReader->pTsdb->mem != NULL) { - tsdbGetTbDataFromMemTable(pReader->pReadSnap->pMem, pReader->suid, pBlockScanInfo->uid, &d); + d = tsdbGetTbDataFromMemTable(pReader->pReadSnap->pMem, pReader->suid, pBlockScanInfo->uid); if (d != NULL) { rows += tsdbGetNRowsInTbData(d); } @@ -3086,7 +3872,7 @@ int64_t tsdbGetNumOfRowsInMemTable(STsdbReader* pReader) { STbData* di = NULL; if (pReader->pTsdb->imem != NULL) { - tsdbGetTbDataFromMemTable(pReader->pReadSnap->pIMem, pReader->suid, pBlockScanInfo->uid, &di); + di = tsdbGetTbDataFromMemTable(pReader->pReadSnap->pIMem, pReader->suid, pBlockScanInfo->uid); if (di != NULL) { rows += tsdbGetNRowsInTbData(di); } diff --git a/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c b/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c index ea9c3e5313d509cd3f8476a2a33c3cc6344ea564..c8f3862071b3628fdefd26df58ea3cb01e80d302 100644 --- a/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c +++ b/source/dnode/vnode/src/tsdb/tsdbReaderWriter.c @@ -15,8 +15,6 @@ #include "tsdb.h" -#define TSDB_FILE_DLMT ((uint32_t)0xF00AFA0F) - // SDelFWriter ==================================================== int32_t tsdbDelFWriterOpen(SDelFWriter **ppWriter, SDelFile *pFile, STsdb *pTsdb) { int32_t code = 0; @@ -63,6 +61,7 @@ _err: int32_t tsdbDelFWriterClose(SDelFWriter **ppWriter, int8_t sync) { int32_t code = 0; SDelFWriter *pWriter = *ppWriter; + STsdb *pTsdb = pWriter->pTsdb; // sync if (sync && taosFsyncFile(pWriter->pWriteH) < 0) { @@ -76,47 +75,47 @@ int32_t tsdbDelFWriterClose(SDelFWriter **ppWriter, int8_t sync) { goto _err; } + for (int32_t iBuf = 0; iBuf < sizeof(pWriter->aBuf) / sizeof(uint8_t *); iBuf++) { + tFree(pWriter->aBuf[iBuf]); + } + taosMemoryFree(pWriter); + *ppWriter = NULL; return code; _err: - tsdbError("vgId:%d, failed to close del file writer since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); + tsdbError("vgId:%d, failed to close del file writer since %s", TD_VID(pTsdb->pVnode), tstrerror(code)); return code; } -int32_t tsdbWriteDelData(SDelFWriter *pWriter, SArray *aDelData, uint8_t **ppBuf, SDelIdx *pDelIdx) { - int32_t code = 0; - uint8_t *pBuf = NULL; - int64_t size; - int64_t n; - SBlockDataHdr hdr = {.delimiter = TSDB_FILE_DLMT, .suid = pDelIdx->suid, .uid = pDelIdx->uid}; - - if (!ppBuf) ppBuf = &pBuf; +int32_t tsdbWriteDelData(SDelFWriter *pWriter, SArray *aDelData, SDelIdx *pDelIdx) { + int32_t code = 0; + int64_t size; + int64_t n; // prepare - size = sizeof(hdr); + size = sizeof(uint32_t); for (int32_t iDelData = 0; iDelData < taosArrayGetSize(aDelData); iDelData++) { size += tPutDelData(NULL, taosArrayGet(aDelData, iDelData)); } size += sizeof(TSCKSUM); // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pWriter->aBuf[0], size); if (code) goto _err; // build n = 0; - *(SBlockDataHdr *)(*ppBuf) = hdr; - n += sizeof(hdr); + n += tPutU32(pWriter->aBuf[0] + n, TSDB_FILE_DLMT); for (int32_t iDelData = 0; iDelData < taosArrayGetSize(aDelData); iDelData++) { - n += tPutDelData(*ppBuf + n, taosArrayGet(aDelData, iDelData)); + n += tPutDelData(pWriter->aBuf[0] + n, taosArrayGet(aDelData, iDelData)); } - taosCalcChecksumAppend(0, *ppBuf, size); + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); ASSERT(n + sizeof(TSCKSUM) == size); // write - n = taosWriteFile(pWriter->pWriteH, *ppBuf, size); + n = taosWriteFile(pWriter->pWriteH, pWriter->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -129,48 +128,42 @@ int32_t tsdbWriteDelData(SDelFWriter *pWriter, SArray *aDelData, uint8_t **ppBuf pDelIdx->size = size; pWriter->fDel.size += size; - tFree(pBuf); return code; _err: tsdbError("vgId:%d, failed to write del data since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } -int32_t tsdbWriteDelIdx(SDelFWriter *pWriter, SArray *aDelIdx, uint8_t **ppBuf) { +int32_t tsdbWriteDelIdx(SDelFWriter *pWriter, SArray *aDelIdx) { int32_t code = 0; int64_t size; int64_t n; - uint8_t *pBuf = NULL; SDelIdx *pDelIdx; - if (!ppBuf) ppBuf = &pBuf; - // prepare - size = 0; - size += tPutU32(NULL, TSDB_FILE_DLMT); + size = sizeof(uint32_t); for (int32_t iDelIdx = 0; iDelIdx < taosArrayGetSize(aDelIdx); iDelIdx++) { size += tPutDelIdx(NULL, taosArrayGet(aDelIdx, iDelIdx)); } size += sizeof(TSCKSUM); // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pWriter->aBuf[0], size); if (code) goto _err; // build n = 0; - n += tPutU32(*ppBuf + n, TSDB_FILE_DLMT); + n += tPutU32(pWriter->aBuf[0] + n, TSDB_FILE_DLMT); for (int32_t iDelIdx = 0; iDelIdx < taosArrayGetSize(aDelIdx); iDelIdx++) { - n += tPutDelIdx(*ppBuf + n, taosArrayGet(aDelIdx, iDelIdx)); + n += tPutDelIdx(pWriter->aBuf[0] + n, taosArrayGet(aDelIdx, iDelIdx)); } - taosCalcChecksumAppend(0, *ppBuf, size); + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); ASSERT(n + sizeof(TSCKSUM) == size); // write - n = taosWriteFile(pWriter->pWriteH, *ppBuf, size); + n = taosWriteFile(pWriter->pWriteH, pWriter->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -180,12 +173,10 @@ int32_t tsdbWriteDelIdx(SDelFWriter *pWriter, SArray *aDelIdx, uint8_t **ppBuf) pWriter->fDel.offset = pWriter->fDel.size; pWriter->fDel.size += size; - tFree(pBuf); return code; _err: tsdbError("vgId:%d, write del idx failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } @@ -225,9 +216,11 @@ struct SDelFReader { STsdb *pTsdb; SDelFile fDel; TdFilePtr pReadH; + + uint8_t *aBuf[1]; }; -int32_t tsdbDelFReaderOpen(SDelFReader **ppReader, SDelFile *pFile, STsdb *pTsdb, uint8_t **ppBuf) { +int32_t tsdbDelFReaderOpen(SDelFReader **ppReader, SDelFile *pFile, STsdb *pTsdb) { int32_t code = 0; char fname[TSDB_FILENAME_LEN]; SDelFReader *pDelFReader; @@ -252,32 +245,6 @@ int32_t tsdbDelFReaderOpen(SDelFReader **ppReader, SDelFile *pFile, STsdb *pTsdb goto _err; } -#if 0 - // load and check hdr if buffer is given - if (ppBuf) { - code = tRealloc(ppBuf, TSDB_FHDR_SIZE); - if (code) { - goto _err; - } - - n = taosReadFile(pDelFReader->pReadH, *ppBuf, TSDB_FHDR_SIZE); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } else if (n < TSDB_FHDR_SIZE) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } - - if (!taosCheckChecksumWhole(*ppBuf, TSDB_FHDR_SIZE)) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } - - // TODO: check the content - } -#endif - _exit: *ppReader = pDelFReader; return code; @@ -297,6 +264,9 @@ int32_t tsdbDelFReaderClose(SDelFReader **ppReader) { code = TAOS_SYSTEM_ERROR(errno); goto _exit; } + for (int32_t iBuf = 0; iBuf < sizeof(pReader->aBuf) / sizeof(uint8_t *); iBuf++) { + tFree(pReader->aBuf[iBuf]); + } taosMemoryFree(pReader); } *ppReader = NULL; @@ -305,16 +275,13 @@ _exit: return code; } -int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData, uint8_t **ppBuf) { - int32_t code = 0; - int64_t offset = pDelIdx->offset; - int64_t size = pDelIdx->size; - int64_t n; - uint8_t *pBuf = NULL; - SBlockDataHdr *pHdr; - SDelData *pDelData = &(SDelData){0}; +int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData) { + int32_t code = 0; + int64_t offset = pDelIdx->offset; + int64_t size = pDelIdx->size; + int64_t n; - if (!ppBuf) ppBuf = &pBuf; + taosArrayClear(aDelData); // seek if (taosLSeekFile(pReader->pReadH, offset, SEEK_SET) < 0) { @@ -323,11 +290,11 @@ int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData } // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; // read - n = taosReadFile(pReader->pReadH, *ppBuf, size); + n = taosReadFile(pReader->pReadH, pReader->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -337,23 +304,21 @@ int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData } // check - if (!taosCheckChecksumWhole(*ppBuf, size)) { + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } // // decode n = 0; - pHdr = (SBlockDataHdr *)(*ppBuf + n); - ASSERT(pHdr->delimiter == TSDB_FILE_DLMT); - ASSERT(pHdr->suid == pDelIdx->suid); - ASSERT(pHdr->uid == pDelIdx->uid); - n += sizeof(*pHdr); - taosArrayClear(aDelData); + + uint32_t delimiter; + n += tGetU32(pReader->aBuf[0] + n, &delimiter); while (n < size - sizeof(TSCKSUM)) { - n += tGetDelData(*ppBuf + n, pDelData); + SDelData delData; + n += tGetDelData(pReader->aBuf[0] + n, &delData); - if (taosArrayPush(aDelData, pDelData) == NULL) { + if (taosArrayPush(aDelData, &delData) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } @@ -361,25 +326,20 @@ int32_t tsdbReadDelData(SDelFReader *pReader, SDelIdx *pDelIdx, SArray *aDelData ASSERT(n == size - sizeof(TSCKSUM)); - tFree(pBuf); return code; _err: tsdbError("vgId:%d, read del data failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } -int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx, uint8_t **ppBuf) { - int32_t code = 0; - int32_t n; - int64_t offset = pReader->fDel.offset; - int64_t size = pReader->fDel.size - offset; - uint32_t delimiter; - uint8_t *pBuf = NULL; - SDelIdx *pDelIdx = &(SDelIdx){0}; +int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx) { + int32_t code = 0; + int32_t n; + int64_t offset = pReader->fDel.offset; + int64_t size = pReader->fDel.size - offset; - if (!ppBuf) ppBuf = &pBuf; + taosArrayClear(aDelIdx); // seek if (taosLSeekFile(pReader->pReadH, offset, SEEK_SET) < 0) { @@ -388,11 +348,11 @@ int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx, uint8_t **ppBuf) { } // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; // read - n = taosReadFile(pReader->pReadH, *ppBuf, size); + n = taosReadFile(pReader->pReadH, pReader->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -402,21 +362,23 @@ int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx, uint8_t **ppBuf) { } // check - if (!taosCheckChecksumWhole(*ppBuf, size)) { + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } // decode n = 0; - n += tGetU32(*ppBuf + n, &delimiter); + uint32_t delimiter; + n += tGetU32(pReader->aBuf[0] + n, &delimiter); ASSERT(delimiter == TSDB_FILE_DLMT); - taosArrayClear(aDelIdx); while (n < size - sizeof(TSCKSUM)) { - n += tGetDelIdx(*ppBuf + n, pDelIdx); + SDelIdx delIdx; - if (taosArrayPush(aDelIdx, pDelIdx) == NULL) { + n += tGetDelIdx(pReader->aBuf[0] + n, &delIdx); + + if (taosArrayPush(aDelIdx, &delIdx) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } @@ -424,12 +386,10 @@ int32_t tsdbReadDelIdx(SDelFReader *pReader, SArray *aDelIdx, uint8_t **ppBuf) { ASSERT(n == size - sizeof(TSCKSUM)); - tFree(pBuf); return code; _err: tsdbError("vgId:%d, read del idx failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } @@ -441,6 +401,8 @@ struct SDataFReader { TdFilePtr pDataFD; TdFilePtr pLastFD; TdFilePtr pSmaFD; + + uint8_t *aBuf[3]; }; int32_t tsdbDataFReaderOpen(SDataFReader **ppReader, STsdb *pTsdb, SDFileSet *pSet) { @@ -523,6 +485,10 @@ int32_t tsdbDataFReaderClose(SDataFReader **ppReader) { goto _err; } + for (int32_t iBuf = 0; iBuf < sizeof((*ppReader)->aBuf) / sizeof(uint8_t *); iBuf++) { + tFree((*ppReader)->aBuf[iBuf]); + } + taosMemoryFree(*ppReader); _exit: @@ -534,19 +500,20 @@ _err: return code; } -int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx, uint8_t **ppBuf) { - int32_t code = 0; - int64_t offset = pReader->pSet->pHeadF->offset; - int64_t size = pReader->pSet->pHeadF->size - offset; - uint8_t *pBuf = NULL; - int64_t n; - uint32_t delimiter; - SBlockIdx blockIdx; +int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx) { + int32_t code = 0; + int64_t offset = pReader->pSet->pHeadF->offset; + int64_t size = pReader->pSet->pHeadF->size - offset; + int64_t n; + uint32_t delimiter; - if (!ppBuf) ppBuf = &pBuf; + taosArrayClear(aBlockIdx); + if (size == 0) { + goto _exit; + } // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; // seek @@ -556,7 +523,7 @@ int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx, uint8_t **ppB } // read - n = taosReadFile(pReader->pHeadFD, *ppBuf, size); + n = taosReadFile(pReader->pHeadFD, pReader->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -566,19 +533,19 @@ int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx, uint8_t **ppB } // check - if (!taosCheckChecksumWhole(*ppBuf, size)) { + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } // decode n = 0; - n = tGetU32(*ppBuf + n, &delimiter); + n = tGetU32(pReader->aBuf[0] + n, &delimiter); ASSERT(delimiter == TSDB_FILE_DLMT); - taosArrayClear(aBlockIdx); while (n < size - sizeof(TSCKSUM)) { - n += tGetBlockIdx(*ppBuf + n, &blockIdx); + SBlockIdx blockIdx; + n += tGetBlockIdx(pReader->aBuf[0] + n, &blockIdx); if (taosArrayPush(aBlockIdx, &blockIdx) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; @@ -588,38 +555,38 @@ int32_t tsdbReadBlockIdx(SDataFReader *pReader, SArray *aBlockIdx, uint8_t **ppB ASSERT(n + sizeof(TSCKSUM) == size); - tFree(pBuf); +_exit: return code; _err: tsdbError("vgId:%d, read block idx failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } -int32_t tsdbReadBlock(SDataFReader *pReader, SBlockIdx *pBlockIdx, SMapData *mBlock, uint8_t **ppBuf) { - int32_t code = 0; - int64_t offset = pBlockIdx->offset; - int64_t size = pBlockIdx->size; - uint8_t *pBuf = NULL; - int64_t n; - int64_t tn; - SBlockDataHdr hdr; +int32_t tsdbReadBlockL(SDataFReader *pReader, SArray *aBlockL) { + int32_t code = 0; + int64_t offset = pReader->pSet->pLastF->offset; + int64_t size = pReader->pSet->pLastF->size - offset; + int64_t n; + uint32_t delimiter; - if (!ppBuf) ppBuf = &pBuf; + taosArrayClear(aBlockL); + if (size == 0) { + goto _exit; + } // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; // seek - if (taosLSeekFile(pReader->pHeadFD, offset, SEEK_SET) < 0) { + if (taosLSeekFile(pReader->pLastFD, offset, SEEK_SET) < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; } // read - n = taosReadFile(pReader->pHeadFD, *ppBuf, size); + n = taosReadFile(pReader->pLastFD, pReader->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -629,245 +596,116 @@ int32_t tsdbReadBlock(SDataFReader *pReader, SBlockIdx *pBlockIdx, SMapData *mBl } // check - if (!taosCheckChecksumWhole(*ppBuf, size)) { + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } // decode - hdr = *(SBlockDataHdr *)(*ppBuf); - ASSERT(hdr.delimiter == TSDB_FILE_DLMT); - ASSERT(hdr.suid == pBlockIdx->suid); - ASSERT(hdr.uid == pBlockIdx->uid); + n = 0; + n = tGetU32(pReader->aBuf[0] + n, &delimiter); + ASSERT(delimiter == TSDB_FILE_DLMT); - n = sizeof(hdr); - tn = tGetMapData(*ppBuf + n, mBlock); - if (tn < 0) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; + while (n < size - sizeof(TSCKSUM)) { + SBlockL blockl; + n += tGetBlockL(pReader->aBuf[0] + n, &blockl); + + if (taosArrayPush(aBlockL, &blockl) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } } - n += tn; + ASSERT(n + sizeof(TSCKSUM) == size); - tFree(pBuf); +_exit: return code; _err: - tsdbError("vgId:%d, read block failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); + tsdbError("vgId:%d read blockl failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } -static int32_t tsdbReadBlockDataKey(SBlockData *pBlockData, SSubBlock *pSubBlock, uint8_t *pBuf, uint8_t **ppBuf) { +int32_t tsdbReadBlock(SDataFReader *pReader, SBlockIdx *pBlockIdx, SMapData *mBlock) { int32_t code = 0; - int64_t size = pSubBlock->szVersion + pSubBlock->szTSKEY + sizeof(TSCKSUM); + int64_t offset = pBlockIdx->offset; + int64_t size = pBlockIdx->size; int64_t n; + int64_t tn; - if (!taosCheckChecksumWhole(pBuf, size)) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } - - code = tRealloc((uint8_t **)&pBlockData->aVersion, sizeof(int64_t) * pSubBlock->nRow); - if (code) goto _err; - code = tRealloc((uint8_t **)&pBlockData->aTSKEY, sizeof(TSKEY) * pSubBlock->nRow); + // alloc + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; - if (pSubBlock->cmprAlg == NO_COMPRESSION) { - ASSERT(pSubBlock->szVersion == sizeof(int64_t) * pSubBlock->nRow); - ASSERT(pSubBlock->szTSKEY == sizeof(TSKEY) * pSubBlock->nRow); - - // VERSION - memcpy(pBlockData->aVersion, pBuf, pSubBlock->szVersion); - - // TSKEY - memcpy(pBlockData->aTSKEY, pBuf + pSubBlock->szVersion, pSubBlock->szTSKEY); - } else { - size = sizeof(int64_t) * pSubBlock->nRow + COMP_OVERFLOW_BYTES; - if (pSubBlock->cmprAlg == TWO_STAGE_COMP) { - code = tRealloc(ppBuf, size); - if (code) goto _err; - } - - // VERSION - n = tsDecompressBigint(pBuf, pSubBlock->szVersion, pSubBlock->nRow, (char *)pBlockData->aVersion, - sizeof(int64_t) * pSubBlock->nRow, pSubBlock->cmprAlg, *ppBuf, size); - if (n < 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - - // TSKEY - n = tsDecompressTimestamp(pBuf + pSubBlock->szVersion, pSubBlock->szTSKEY, pSubBlock->nRow, - (char *)pBlockData->aTSKEY, sizeof(TSKEY) * pSubBlock->nRow, pSubBlock->cmprAlg, *ppBuf, - size); - if (n < 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } + // seek + if (taosLSeekFile(pReader->pHeadFD, offset, SEEK_SET) < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _err; } - return code; - -_err: - return code; -} - -static int32_t tsdbReadColDataImpl(SSubBlock *pSubBlock, SBlockCol *pBlockCol, SColData *pColData, uint8_t *pBuf, - uint8_t **ppBuf) { - int32_t code = 0; - int64_t size; - int64_t n; - - if (!taosCheckChecksumWhole(pBuf, pBlockCol->szBitmap + pBlockCol->szOffset + pBlockCol->szValue + sizeof(TSCKSUM))) { + // read + n = taosReadFile(pReader->pHeadFD, pReader->aBuf[0], size); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _err; + } else if (n < size) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } - pColData->nVal = pSubBlock->nRow; - pColData->flag = pBlockCol->flag; - - // BITMAP - if (pBlockCol->flag != HAS_VALUE) { - ASSERT(pBlockCol->szBitmap); - - size = BIT2_SIZE(pColData->nVal); - code = tRealloc(&pColData->pBitMap, size); - if (code) goto _err; - - code = tRealloc(ppBuf, size + COMP_OVERFLOW_BYTES); - if (code) goto _err; - - n = tsDecompressTinyint(pBuf, pBlockCol->szBitmap, size, pColData->pBitMap, size, TWO_STAGE_COMP, *ppBuf, - size + COMP_OVERFLOW_BYTES); - if (n <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - - ASSERT(n == size); - } else { - ASSERT(pBlockCol->szBitmap == 0); - } - pBuf = pBuf + pBlockCol->szBitmap; - - // OFFSET - if (IS_VAR_DATA_TYPE(pColData->type)) { - ASSERT(pBlockCol->szOffset); - - size = sizeof(int32_t) * pColData->nVal; - code = tRealloc((uint8_t **)&pColData->aOffset, size); - if (code) goto _err; - - code = tRealloc(ppBuf, size + COMP_OVERFLOW_BYTES); - if (code) goto _err; - - n = tsDecompressInt(pBuf, pBlockCol->szOffset, pColData->nVal, (char *)pColData->aOffset, size, TWO_STAGE_COMP, - *ppBuf, size + COMP_OVERFLOW_BYTES); - if (n <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - - ASSERT(n == size); - } else { - ASSERT(pBlockCol->szOffset == 0); - } - pBuf = pBuf + pBlockCol->szOffset; - - // VALUE - pColData->nData = pBlockCol->szOrigin; - - code = tRealloc(&pColData->pData, pColData->nData); - if (code) goto _err; - - if (pSubBlock->cmprAlg == NO_COMPRESSION) { - memcpy(pColData->pData, pBuf, pColData->nData); - } else { - if (pSubBlock->cmprAlg == TWO_STAGE_COMP) { - code = tRealloc(ppBuf, pColData->nData + COMP_OVERFLOW_BYTES); - if (code) goto _err; - } - - n = tDataTypes[pBlockCol->type].decompFunc(pBuf, pBlockCol->szValue, pSubBlock->nRow, pColData->pData, - pColData->nData, pSubBlock->cmprAlg, *ppBuf, - pColData->nData + COMP_OVERFLOW_BYTES); - if (n < 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - - ASSERT(n == pColData->nData); - } - - return code; - -_err: - return code; -} - -static int32_t tsdbReadBlockCol(SSubBlock *pSubBlock, uint8_t *p, SArray *aBlockCol) { - int32_t code = 0; - int32_t n = 0; - SBlockCol blockCol; - SBlockCol *pBlockCol = &blockCol; - - if (!taosCheckChecksumWhole(p, pSubBlock->szBlockCol + sizeof(TSCKSUM))) { + // check + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { code = TSDB_CODE_FILE_CORRUPTED; goto _err; } - n += sizeof(SBlockDataHdr); - while (n < pSubBlock->szBlockCol) { - n += tGetBlockCol(p + n, pBlockCol); + // decode + n = 0; - if (taosArrayPush(aBlockCol, pBlockCol) == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - } + uint32_t delimiter; + n += tGetU32(pReader->aBuf[0] + n, &delimiter); + ASSERT(delimiter == TSDB_FILE_DLMT); - ASSERT(n == pSubBlock->szBlockCol); + tn = tGetMapData(pReader->aBuf[0] + n, mBlock); + if (tn < 0) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + n += tn; + ASSERT(n + sizeof(TSCKSUM) == size); return code; _err: + tsdbError("vgId:%d, read block failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } -static int32_t tsdbReadSubColData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, int32_t iSubBlock, - int16_t *aColId, int32_t nCol, SBlockData *pBlockData, uint8_t **ppBuf1, - uint8_t **ppBuf2) { - TdFilePtr pFD = pBlock->last ? pReader->pLastFD : pReader->pDataFD; - SSubBlock *pSubBlock = &pBlock->aSubBlock[iSubBlock]; - SArray *aBlockCol = NULL; - int32_t code = 0; - int64_t offset; - int64_t size; - int64_t n; +int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnDataAgg) { + int32_t code = 0; + SSmaInfo *pSmaInfo = &pBlock->smaInfo; - tBlockDataReset(pBlockData); - pBlockData->nRow = pSubBlock->nRow; + ASSERT(pSmaInfo->size > 0); - // TSDBKEY and SBlockCol - if (nCol == 1) { - offset = pSubBlock->offset + pSubBlock->szBlockCol + sizeof(TSCKSUM); - size = pSubBlock->szVersion + pSubBlock->szTSKEY + sizeof(TSCKSUM); - } else { - offset = pSubBlock->offset; - size = pSubBlock->szBlockCol + sizeof(TSCKSUM) + pSubBlock->szVersion + pSubBlock->szTSKEY + sizeof(TSCKSUM); - } + taosArrayClear(aColumnDataAgg); - code = tRealloc(ppBuf1, size); + // alloc + int32_t size = pSmaInfo->size + sizeof(TSCKSUM); + code = tRealloc(&pReader->aBuf[0], size); if (code) goto _err; - n = taosLSeekFile(pFD, offset, SEEK_SET); + // seek + int64_t n = taosLSeekFile(pReader->pSmaFD, pSmaInfo->offset, SEEK_SET); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; + } else if (n < pSmaInfo->offset) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _err; } - n = taosReadFile(pFD, *ppBuf1, size); + // read + n = taosReadFile(pReader->pSmaFD, pReader->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -876,337 +714,203 @@ static int32_t tsdbReadSubColData(SDataFReader *pReader, SBlockIdx *pBlockIdx, S goto _err; } - if (nCol == 1) { - code = tsdbReadBlockDataKey(pBlockData, pSubBlock, *ppBuf1, ppBuf2); - if (code) goto _err; - - goto _exit; - } else { - aBlockCol = taosArrayInit(0, sizeof(SBlockCol)); - if (aBlockCol == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - - code = tsdbReadBlockCol(pSubBlock, *ppBuf1, aBlockCol); - if (code) goto _err; - - code = tsdbReadBlockDataKey(pBlockData, pSubBlock, *ppBuf1 + pSubBlock->szBlockCol + sizeof(TSCKSUM), ppBuf2); - if (code) goto _err; + // check + if (!taosCheckChecksumWhole(pReader->aBuf[0], size)) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _err; } - for (int32_t iCol = 1; iCol < nCol; iCol++) { - void *p = taosArraySearch(aBlockCol, &(SBlockCol){.cid = aColId[iCol]}, tBlockColCmprFn, TD_EQ); - - if (p) { - SBlockCol *pBlockCol = (SBlockCol *)p; - SColData *pColData; - - ASSERT(pBlockCol->flag && pBlockCol->flag != HAS_NONE); - - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _err; - - tColDataInit(pColData, pBlockCol->cid, pBlockCol->type, pBlockCol->smaOn); - if (pBlockCol->flag == HAS_NULL) { - for (int32_t iRow = 0; iRow < pSubBlock->nRow; iRow++) { - code = tColDataAppendValue(pColData, &COL_VAL_NULL(pBlockCol->cid, pBlockCol->type)); - if (code) goto _err; - } - } else { - offset = pSubBlock->offset + pSubBlock->szBlockCol + sizeof(TSCKSUM) + pSubBlock->szVersion + - pSubBlock->szTSKEY + sizeof(TSCKSUM) + pBlockCol->offset; - size = pBlockCol->szBitmap + pBlockCol->szOffset + pBlockCol->szValue + sizeof(TSCKSUM); - - code = tRealloc(ppBuf1, size); - if (code) goto _err; - - // seek - n = taosLSeekFile(pFD, offset, SEEK_SET); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } - - // read - n = taosReadFile(pFD, *ppBuf1, size); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } else if (n < size) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } + // decode + n = 0; + while (n < pSmaInfo->size) { + SColumnDataAgg sma; - code = tsdbReadColDataImpl(pSubBlock, pBlockCol, pColData, *ppBuf1, ppBuf2); - if (code) goto _err; - } + n += tGetColumnDataAgg(pReader->aBuf[0] + n, &sma); + if (taosArrayPush(aColumnDataAgg, &sma) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; } } -_exit: - taosArrayDestroy(aBlockCol); return code; _err: - taosArrayDestroy(aBlockCol); + tsdbError("vgId:%d tsdb read block sma failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } -int32_t tsdbReadColData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, int16_t *aColId, int32_t nCol, - SBlockData *pBlockData, uint8_t **ppBuf1, uint8_t **ppBuf2) { - int32_t code = 0; - uint8_t *pBuf1 = NULL; - uint8_t *pBuf2 = NULL; +static int32_t tsdbReadBlockDataImpl(SDataFReader *pReader, SBlockInfo *pBlkInfo, int8_t fromLast, + SBlockData *pBlockData) { + int32_t code = 0; - ASSERT(aColId[0] == PRIMARYKEY_TIMESTAMP_COL_ID); + tBlockDataClear(pBlockData); - if (!ppBuf1) ppBuf1 = &pBuf1; - if (!ppBuf2) ppBuf2 = &pBuf2; + TdFilePtr pFD = fromLast ? pReader->pLastFD : pReader->pDataFD; - code = tsdbReadSubColData(pReader, pBlockIdx, pBlock, 0, aColId, nCol, pBlockData, ppBuf1, ppBuf2); + // uid + version + tskey + code = tsdbReadAndCheck(pFD, pBlkInfo->offset, &pReader->aBuf[0], pBlkInfo->szKey, 1); if (code) goto _err; + SDiskDataHdr hdr; + uint8_t *p = pReader->aBuf[0] + tGetDiskDataHdr(pReader->aBuf[0], &hdr); - if (pBlock->nSubBlock > 1) { - SBlockData *pBlockData1 = &(SBlockData){0}; - SBlockData *pBlockData2 = &(SBlockData){0}; - - tBlockDataInit(pBlockData1); - tBlockDataInit(pBlockData2); - for (int32_t iSubBlock = 1; iSubBlock < pBlock->nSubBlock; iSubBlock++) { - code = tsdbReadSubColData(pReader, pBlockIdx, pBlock, iSubBlock, aColId, nCol, pBlockData1, ppBuf1, ppBuf2); - if (code) goto _err; + ASSERT(hdr.delimiter == TSDB_FILE_DLMT); + ASSERT(pBlockData->suid == hdr.suid); + ASSERT(pBlockData->uid == hdr.uid); - code = tBlockDataCopy(pBlockData, pBlockData2); - if (code) { - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); - goto _err; - } + pBlockData->nRow = hdr.nRow; - code = tBlockDataMerge(pBlockData1, pBlockData2, pBlockData); - if (code) { - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); - goto _err; - } - } - - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); + // uid + if (hdr.uid == 0) { + ASSERT(hdr.szUid); + code = tsdbDecmprData(p, hdr.szUid, TSDB_DATA_TYPE_BIGINT, hdr.cmprAlg, (uint8_t **)&pBlockData->aUid, + sizeof(int64_t) * hdr.nRow, &pReader->aBuf[1]); + if (code) goto _err; + } else { + ASSERT(!hdr.szUid); } + p += hdr.szUid; - tFree(pBuf1); - tFree(pBuf2); - return code; - -_err: - tsdbError("vgId:%d, tsdb read col data failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf1); - tFree(pBuf2); - return code; -} - -static int32_t tsdbReadSubBlockData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, int32_t iSubBlock, - SBlockData *pBlockData, uint8_t **ppBuf1, uint8_t **ppBuf2) { - int32_t code = 0; - uint8_t *p; - int64_t size; - int64_t n; - TdFilePtr pFD = pBlock->last ? pReader->pLastFD : pReader->pDataFD; - SSubBlock *pSubBlock = &pBlock->aSubBlock[iSubBlock]; - SArray *aBlockCol = NULL; - - tBlockDataReset(pBlockData); + // version + code = tsdbDecmprData(p, hdr.szVer, TSDB_DATA_TYPE_BIGINT, hdr.cmprAlg, (uint8_t **)&pBlockData->aVersion, + sizeof(int64_t) * hdr.nRow, &pReader->aBuf[1]); + if (code) goto _err; + p += hdr.szVer; - // realloc - code = tRealloc(ppBuf1, pSubBlock->szBlock); + // TSKEY + code = tsdbDecmprData(p, hdr.szKey, TSDB_DATA_TYPE_TIMESTAMP, hdr.cmprAlg, (uint8_t **)&pBlockData->aTSKEY, + sizeof(TSKEY) * hdr.nRow, &pReader->aBuf[1]); if (code) goto _err; + p += hdr.szKey; - // seek - n = taosLSeekFile(pFD, pSubBlock->offset, SEEK_SET); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } + ASSERT(p - pReader->aBuf[0] == pBlkInfo->szKey - sizeof(TSCKSUM)); - // read - n = taosReadFile(pFD, *ppBuf1, pSubBlock->szBlock); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } else if (n < pSubBlock->szBlock) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } + // read and decode columns + if (taosArrayGetSize(pBlockData->aIdx) == 0) goto _exit; - pBlockData->nRow = pSubBlock->nRow; + if (hdr.szBlkCol > 0) { + int64_t offset = pBlkInfo->offset + pBlkInfo->szKey; + code = tsdbReadAndCheck(pFD, offset, &pReader->aBuf[0], hdr.szBlkCol + sizeof(TSCKSUM), 1); + if (code) goto _err; + } - // TSDBKEY - p = *ppBuf1 + pSubBlock->szBlockCol + sizeof(TSCKSUM); - code = tsdbReadBlockDataKey(pBlockData, pSubBlock, p, ppBuf2); - if (code) goto _err; + SBlockCol blockCol = {.cid = 0}; + SBlockCol *pBlockCol = &blockCol; + int32_t n = 0; - // COLUMNS - aBlockCol = taosArrayInit(0, sizeof(SBlockCol)); - if (aBlockCol == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } + for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { + SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); - code = tsdbReadBlockCol(pSubBlock, *ppBuf1, aBlockCol); - if (code) goto _err; + while (pBlockCol && pBlockCol->cid < pColData->cid) { + if (n < hdr.szBlkCol) { + n += tGetBlockCol(pReader->aBuf[0] + n, pBlockCol); + } else { + ASSERT(n == hdr.szBlkCol); + pBlockCol = NULL; + } + } - for (int32_t iBlockCol = 0; iBlockCol < taosArrayGetSize(aBlockCol); iBlockCol++) { - SColData *pColData; - SBlockCol *pBlockCol = (SBlockCol *)taosArrayGet(aBlockCol, iBlockCol); + if (pBlockCol == NULL || pBlockCol->cid > pColData->cid) { + // add a lot of NONE + for (int32_t iRow = 0; iRow < hdr.nRow; iRow++) { + code = tColDataAppendValue(pColData, &COL_VAL_NONE(pColData->cid, pColData->type)); + if (code) goto _err; + } + } else { + ASSERT(pBlockCol->type == pColData->type); + ASSERT(pBlockCol->flag && pBlockCol->flag != HAS_NONE); - ASSERT(pBlockCol->flag && pBlockCol->flag != HAS_NONE); + if (pBlockCol->flag == HAS_NULL) { + // add a lot of NULL + for (int32_t iRow = 0; iRow < hdr.nRow; iRow++) { + code = tColDataAppendValue(pColData, &COL_VAL_NULL(pBlockCol->cid, pBlockCol->type)); + if (code) goto _err; + } + } else { + // decode from binary + int64_t offset = pBlkInfo->offset + pBlkInfo->szKey + hdr.szBlkCol + sizeof(TSCKSUM) + pBlockCol->offset; + int32_t size = pBlockCol->szBitmap + pBlockCol->szOffset + pBlockCol->szValue + sizeof(TSCKSUM); - code = tBlockDataAddColData(pBlockData, iBlockCol, &pColData); - if (code) goto _err; + code = tsdbReadAndCheck(pFD, offset, &pReader->aBuf[1], size, 0); + if (code) goto _err; - tColDataInit(pColData, pBlockCol->cid, pBlockCol->type, pBlockCol->smaOn); - if (pBlockCol->flag == HAS_NULL) { - for (int32_t iRow = 0; iRow < pSubBlock->nRow; iRow++) { - code = tColDataAppendValue(pColData, &COL_VAL_NULL(pBlockCol->cid, pBlockCol->type)); + code = tsdbDecmprColData(pReader->aBuf[1], pBlockCol, hdr.cmprAlg, hdr.nRow, pColData, &pReader->aBuf[2]); if (code) goto _err; } - } else { - p = *ppBuf1 + pSubBlock->szBlockCol + sizeof(TSCKSUM) + pSubBlock->szVersion + pSubBlock->szTSKEY + - sizeof(TSCKSUM) + pBlockCol->offset; - code = tsdbReadColDataImpl(pSubBlock, pBlockCol, pColData, p, ppBuf2); - if (code) goto _err; } } - taosArrayDestroy(aBlockCol); +_exit: return code; _err: - tsdbError("vgId:%d, tsdb read sub block data failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - taosArrayDestroy(aBlockCol); + tsdbError("vgId:%d tsdb read block data impl failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } -int32_t tsdbReadBlockData(SDataFReader *pReader, SBlockIdx *pBlockIdx, SBlock *pBlock, SBlockData *pBlockData, - uint8_t **ppBuf1, uint8_t **ppBuf2) { - int32_t code = 0; - TdFilePtr pFD = pBlock->last ? pReader->pLastFD : pReader->pDataFD; - uint8_t *pBuf1 = NULL; - uint8_t *pBuf2 = NULL; - int32_t iSubBlock; - - if (!ppBuf1) ppBuf1 = &pBuf1; - if (!ppBuf2) ppBuf2 = &pBuf2; +int32_t tsdbReadDataBlock(SDataFReader *pReader, SBlock *pBlock, SBlockData *pBlockData) { + int32_t code = 0; - // read the first sub-block - iSubBlock = 0; - code = tsdbReadSubBlockData(pReader, pBlockIdx, pBlock, iSubBlock, pBlockData, ppBuf1, ppBuf2); + code = tsdbReadBlockDataImpl(pReader, &pBlock->aSubBlock[0], 0, pBlockData); if (code) goto _err; - // read remain block data and do merg if (pBlock->nSubBlock > 1) { - SBlockData *pBlockData1 = &(SBlockData){0}; - SBlockData *pBlockData2 = &(SBlockData){0}; + SBlockData bData1; + SBlockData bData2; - tBlockDataInit(pBlockData1); - tBlockDataInit(pBlockData2); - for (iSubBlock = 1; iSubBlock < pBlock->nSubBlock; iSubBlock++) { - code = tsdbReadSubBlockData(pReader, pBlockIdx, pBlock, iSubBlock, pBlockData1, ppBuf1, ppBuf2); + // create + code = tBlockDataCreate(&bData1); + if (code) goto _err; + code = tBlockDataCreate(&bData2); + if (code) goto _err; + + // init + tBlockDataInitEx(&bData1, pBlockData); + tBlockDataInitEx(&bData2, pBlockData); + + for (int32_t iSubBlock = 1; iSubBlock < pBlock->nSubBlock; iSubBlock++) { + code = tsdbReadBlockDataImpl(pReader, &pBlock->aSubBlock[iSubBlock], 0, &bData1); if (code) { - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); + tBlockDataDestroy(&bData1, 1); + tBlockDataDestroy(&bData2, 1); goto _err; } - code = tBlockDataCopy(pBlockData, pBlockData2); + code = tBlockDataCopy(pBlockData, &bData2); if (code) { - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); + tBlockDataDestroy(&bData1, 1); + tBlockDataDestroy(&bData2, 1); goto _err; } - // merge two block data - code = tBlockDataMerge(pBlockData1, pBlockData2, pBlockData); + code = tBlockDataMerge(&bData1, &bData2, pBlockData); if (code) { - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); + tBlockDataDestroy(&bData1, 1); + tBlockDataDestroy(&bData2, 1); goto _err; } } - tBlockDataClear(pBlockData1, 1); - tBlockDataClear(pBlockData2, 1); + tBlockDataDestroy(&bData1, 1); + tBlockDataDestroy(&bData2, 1); } - ASSERT(pBlock->nRow == pBlockData->nRow); - ASSERT(tsdbKeyCmprFn(&pBlock->minKey, &TSDBROW_KEY(&tBlockDataFirstRow(pBlockData))) == 0); - ASSERT(tsdbKeyCmprFn(&pBlock->maxKey, &TSDBROW_KEY(&tBlockDataLastRow(pBlockData))) == 0); - - if (pBuf1) tFree(pBuf1); - if (pBuf2) tFree(pBuf2); return code; _err: - tsdbError("vgId:%d, tsdb read block data failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - if (pBuf1) tFree(pBuf1); - if (pBuf2) tFree(pBuf2); + tsdbError("vgId:%d tsdb read data block failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } -int32_t tsdbReadBlockSma(SDataFReader *pReader, SBlock *pBlock, SArray *aColumnDataAgg, uint8_t **ppBuf) { - int32_t code = 0; - TdFilePtr pFD = pReader->pSmaFD; - int64_t offset = pBlock->aSubBlock[0].sOffset; - int64_t size = pBlock->aSubBlock[0].nSma * sizeof(SColumnDataAgg) + sizeof(TSCKSUM); - uint8_t *pBuf = NULL; - int64_t n; - - ASSERT(tBlockHasSma(pBlock)); +int32_t tsdbReadLastBlock(SDataFReader *pReader, SBlockL *pBlockL, SBlockData *pBlockData) { + int32_t code = 0; - if (!ppBuf) ppBuf = &pBuf; - code = tRealloc(ppBuf, size); + code = tsdbReadBlockDataImpl(pReader, &pBlockL->bInfo, 1, pBlockData); if (code) goto _err; - // lseek - n = taosLSeekFile(pFD, offset, SEEK_SET); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } - - // read - n = taosReadFile(pFD, *ppBuf, size); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } else if (n < size) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } - - // check - if (!taosCheckChecksumWhole(*ppBuf, size)) { - code = TSDB_CODE_FILE_CORRUPTED; - goto _err; - } - - // decode - taosArrayClear(aColumnDataAgg); - for (int32_t iSma = 0; iSma < pBlock->aSubBlock[0].nSma; iSma++) { - if (taosArrayPush(aColumnDataAgg, &((SColumnDataAgg *)(*ppBuf))[iSma]) == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - } - - tFree(pBuf); return code; _err: - tsdbError("vgId:%d, read block sma failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); + tsdbError("vgId:%d tsdb read last block failed since %s", TD_VID(pReader->pTsdb->pVnode), tstrerror(code)); return code; } @@ -1225,6 +929,7 @@ int32_t tsdbDataFWriterOpen(SDataFWriter **ppWriter, STsdb *pTsdb, SDFileSet *pS code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } + if (code) goto _err; pWriter->pTsdb = pTsdb; pWriter->wSet = (SDFileSet){.diskId = pSet->diskId, .fid = pSet->fid, @@ -1357,10 +1062,11 @@ _err: int32_t tsdbDataFWriterClose(SDataFWriter **ppWriter, int8_t sync) { int32_t code = 0; - STsdb *pTsdb = (*ppWriter)->pTsdb; + STsdb *pTsdb = NULL; if (*ppWriter == NULL) goto _exit; + pTsdb = (*ppWriter)->pTsdb; if (sync) { if (taosFsyncFile((*ppWriter)->pHeadFD) < 0) { code = TAOS_SYSTEM_ERROR(errno); @@ -1403,6 +1109,9 @@ int32_t tsdbDataFWriterClose(SDataFWriter **ppWriter, int8_t sync) { goto _err; } + for (int32_t iBuf = 0; iBuf < sizeof((*ppWriter)->aBuf) / sizeof(uint8_t *); iBuf++) { + tFree((*ppWriter)->aBuf[iBuf]); + } taosMemoryFree(*ppWriter); _exit: *ppWriter = NULL; @@ -1493,38 +1202,41 @@ _err: return code; } -int32_t tsdbWriteBlockIdx(SDataFWriter *pWriter, SArray *aBlockIdx, uint8_t **ppBuf) { +int32_t tsdbWriteBlockIdx(SDataFWriter *pWriter, SArray *aBlockIdx) { int32_t code = 0; SHeadFile *pHeadFile = &pWriter->fHead; - uint8_t *pBuf = NULL; - int64_t size; + int64_t size = 0; int64_t n; - if (!ppBuf) ppBuf = &pBuf; + // check + if (taosArrayGetSize(aBlockIdx) == 0) { + pHeadFile->offset = pHeadFile->size; + goto _exit; + } // prepare - size = tPutU32(NULL, TSDB_FILE_DLMT); + size = sizeof(uint32_t); for (int32_t iBlockIdx = 0; iBlockIdx < taosArrayGetSize(aBlockIdx); iBlockIdx++) { size += tPutBlockIdx(NULL, taosArrayGet(aBlockIdx, iBlockIdx)); } size += sizeof(TSCKSUM); // alloc - code = tRealloc(ppBuf, size); + code = tRealloc(&pWriter->aBuf[0], size); if (code) goto _err; // build n = 0; - n = tPutU32(*ppBuf + n, TSDB_FILE_DLMT); + n = tPutU32(pWriter->aBuf[0] + n, TSDB_FILE_DLMT); for (int32_t iBlockIdx = 0; iBlockIdx < taosArrayGetSize(aBlockIdx); iBlockIdx++) { - n += tPutBlockIdx(*ppBuf + n, taosArrayGet(aBlockIdx, iBlockIdx)); + n += tPutBlockIdx(pWriter->aBuf[0] + n, taosArrayGet(aBlockIdx, iBlockIdx)); } - taosCalcChecksumAppend(0, *ppBuf, size); + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); ASSERT(n + sizeof(TSCKSUM) == size); // write - n = taosWriteFile(pWriter->pHeadFD, *ppBuf, size); + n = taosWriteFile(pWriter->pHeadFD, pWriter->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -1534,44 +1246,39 @@ int32_t tsdbWriteBlockIdx(SDataFWriter *pWriter, SArray *aBlockIdx, uint8_t **pp pHeadFile->offset = pHeadFile->size; pHeadFile->size += size; - tFree(pBuf); +_exit: + tsdbTrace("vgId:%d write block idx, offset:%" PRId64 " size:%" PRId64 " nBlockIdx:%d", TD_VID(pWriter->pTsdb->pVnode), + pHeadFile->offset, size, taosArrayGetSize(aBlockIdx)); return code; _err: tsdbError("vgId:%d, write block idx failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf); return code; } -int32_t tsdbWriteBlock(SDataFWriter *pWriter, SMapData *mBlock, uint8_t **ppBuf, SBlockIdx *pBlockIdx) { - int32_t code = 0; - SHeadFile *pHeadFile = &pWriter->fHead; - SBlockDataHdr hdr = {.delimiter = TSDB_FILE_DLMT, .suid = pBlockIdx->suid, .uid = pBlockIdx->uid}; - uint8_t *pBuf = NULL; - int64_t size; - int64_t n; +int32_t tsdbWriteBlock(SDataFWriter *pWriter, SMapData *mBlock, SBlockIdx *pBlockIdx) { + int32_t code = 0; + SHeadFile *pHeadFile = &pWriter->fHead; + int64_t size; + int64_t n; ASSERT(mBlock->nItem > 0); - // prepare - size = sizeof(SBlockDataHdr) + tPutMapData(NULL, mBlock) + sizeof(TSCKSUM); - // alloc - if (!ppBuf) ppBuf = &pBuf; - code = tRealloc(ppBuf, size); + size = sizeof(uint32_t) + tPutMapData(NULL, mBlock) + sizeof(TSCKSUM); + code = tRealloc(&pWriter->aBuf[0], size); if (code) goto _err; // build n = 0; - *(SBlockDataHdr *)(*ppBuf) = hdr; - n += sizeof(hdr); - n += tPutMapData(*ppBuf + n, mBlock); - taosCalcChecksumAppend(0, *ppBuf, size); + n += tPutU32(pWriter->aBuf[0] + n, TSDB_FILE_DLMT); + n += tPutMapData(pWriter->aBuf[0] + n, mBlock); + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); ASSERT(n + sizeof(TSCKSUM) == size); // write - n = taosWriteFile(pWriter->pHeadFD, *ppBuf, size); + n = taosWriteFile(pWriter->pHeadFD, pWriter->aBuf[0], size); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; @@ -1582,17 +1289,71 @@ int32_t tsdbWriteBlock(SDataFWriter *pWriter, SMapData *mBlock, uint8_t **ppBuf, pBlockIdx->size = size; pHeadFile->size += size; - tFree(pBuf); - tsdbTrace("vgId:%d, write block, offset:%" PRId64 " size:%" PRId64, TD_VID(pWriter->pTsdb->pVnode), pBlockIdx->offset, - pBlockIdx->size); + tsdbTrace("vgId:%d, write block, file ID:%d commit ID:%d suid:%" PRId64 " uid:%" PRId64 " offset:%" PRId64 + " size:%" PRId64 " nItem:%d", + TD_VID(pWriter->pTsdb->pVnode), pWriter->wSet.fid, pHeadFile->commitID, pBlockIdx->suid, pBlockIdx->uid, + pBlockIdx->offset, pBlockIdx->size, mBlock->nItem); return code; _err: - tFree(pBuf); tsdbError("vgId:%d, write block failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); return code; } +int32_t tsdbWriteBlockL(SDataFWriter *pWriter, SArray *aBlockL) { + int32_t code = 0; + SLastFile *pLastFile = &pWriter->fLast; + int64_t size; + int64_t n; + + // check + if (taosArrayGetSize(aBlockL) == 0) { + pLastFile->offset = pLastFile->size; + goto _exit; + } + + // size + size = sizeof(uint32_t); // TSDB_FILE_DLMT + for (int32_t iBlockL = 0; iBlockL < taosArrayGetSize(aBlockL); iBlockL++) { + size += tPutBlockL(NULL, taosArrayGet(aBlockL, iBlockL)); + } + size += sizeof(TSCKSUM); + + // alloc + code = tRealloc(&pWriter->aBuf[0], size); + if (code) goto _err; + + // encode + n = 0; + n += tPutU32(pWriter->aBuf[0] + n, TSDB_FILE_DLMT); + for (int32_t iBlockL = 0; iBlockL < taosArrayGetSize(aBlockL); iBlockL++) { + n += tPutBlockL(pWriter->aBuf[0] + n, taosArrayGet(aBlockL, iBlockL)); + } + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); + + ASSERT(n + sizeof(TSCKSUM) == size); + + // write + n = taosWriteFile(pWriter->pLastFD, pWriter->aBuf[0], size); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _err; + } + + // update + pLastFile->offset = pLastFile->size; + pLastFile->size += size; + +_exit: + tsdbTrace("vgId:%d tsdb write blockl, loffset:%" PRId64 " size:%" PRId64, TD_VID(pWriter->pTsdb->pVnode), + pLastFile->offset, size); + return code; + +_err: + tsdbError("vgId:%d tsdb write blockl failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); + return code; +} + static void tsdbUpdateBlockInfo(SBlockData *pBlockData, SBlock *pBlock) { for (int32_t iRow = 0; iRow < pBlockData->nRow; iRow++) { TSDBKEY key = {.ts = pBlockData->aTSKEY[iRow], .version = pBlockData->aVersion[iRow]}; @@ -1611,357 +1372,127 @@ static void tsdbUpdateBlockInfo(SBlockData *pBlockData, SBlock *pBlock) { pBlock->maxKey = key; } - pBlock->minVersion = TMIN(pBlock->minVersion, key.version); - pBlock->maxVersion = TMAX(pBlock->maxVersion, key.version); + pBlock->minVer = TMIN(pBlock->minVer, key.version); + pBlock->maxVer = TMAX(pBlock->maxVer, key.version); } pBlock->nRow += pBlockData->nRow; } -static int32_t tsdbWriteBlockDataKey(SSubBlock *pSubBlock, SBlockData *pBlockData, uint8_t **ppBuf1, int64_t *nDataP, - uint8_t **ppBuf2) { +static int32_t tsdbWriteBlockSma(SDataFWriter *pWriter, SBlockData *pBlockData, SSmaInfo *pSmaInfo) { int32_t code = 0; - int64_t size; - int64_t tsize; - - if (pSubBlock->cmprAlg == NO_COMPRESSION) { - pSubBlock->szVersion = sizeof(int64_t) * pSubBlock->nRow; - pSubBlock->szTSKEY = sizeof(TSKEY) * pSubBlock->nRow; - - code = tRealloc(ppBuf1, *nDataP + pSubBlock->szVersion + pSubBlock->szTSKEY + sizeof(TSCKSUM)); - if (code) goto _err; - - // VERSION - memcpy(*ppBuf1 + *nDataP, pBlockData->aVersion, pSubBlock->szVersion); - - // TSKEY - memcpy(*ppBuf1 + *nDataP + pSubBlock->szVersion, pBlockData->aTSKEY, pSubBlock->szTSKEY); - } else { - size = (sizeof(int64_t) + sizeof(TSKEY)) * pSubBlock->nRow + COMP_OVERFLOW_BYTES * 2; - - code = tRealloc(ppBuf1, *nDataP + size + sizeof(TSCKSUM)); - if (code) goto _err; - - tsize = sizeof(int64_t) * pSubBlock->nRow + COMP_OVERFLOW_BYTES; - if (pSubBlock->cmprAlg == TWO_STAGE_COMP) { - code = tRealloc(ppBuf2, tsize); - if (code) goto _err; - } - - // VERSION - pSubBlock->szVersion = - tsCompressBigint((char *)pBlockData->aVersion, sizeof(int64_t) * pBlockData->nRow, pBlockData->nRow, - *ppBuf1 + *nDataP, size, pSubBlock->cmprAlg, *ppBuf2, tsize); - if (pSubBlock->szVersion <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - // TSKEY - pSubBlock->szTSKEY = tsCompressTimestamp((char *)pBlockData->aTSKEY, sizeof(TSKEY) * pBlockData->nRow, - pBlockData->nRow, *ppBuf1 + *nDataP + pSubBlock->szVersion, - size - pSubBlock->szVersion, pSubBlock->cmprAlg, *ppBuf2, tsize); - if (pSubBlock->szTSKEY <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } + pSmaInfo->offset = 0; + pSmaInfo->size = 0; - ASSERT(pSubBlock->szVersion + pSubBlock->szTSKEY <= size); - } - - // checksum - size = pSubBlock->szVersion + pSubBlock->szTSKEY + sizeof(TSCKSUM); - taosCalcChecksumAppend(0, *ppBuf1 + *nDataP, size); - - *nDataP += size; - return code; - -_err: - return code; -} + // encode + for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { + SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); -static int32_t tsdbWriteColData(SColData *pColData, SBlockCol *pBlockCol, SSubBlock *pSubBlock, uint8_t **ppBuf1, - int64_t *nDataP, uint8_t **ppBuf2) { - int32_t code = 0; - int64_t size; - int64_t n = 0; + if ((!pColData->smaOn) || IS_VAR_DATA_TYPE(pColData->type)) continue; - // BITMAP - if (pColData->flag != HAS_VALUE) { - size = BIT2_SIZE(pColData->nVal) + COMP_OVERFLOW_BYTES; + SColumnDataAgg sma; + tsdbCalcColDataSMA(pColData, &sma); - code = tRealloc(ppBuf1, *nDataP + n + size); + code = tRealloc(&pWriter->aBuf[0], pSmaInfo->size + tPutColumnDataAgg(NULL, &sma)); if (code) goto _err; - - code = tRealloc(ppBuf2, size); - if (code) goto _err; - - pBlockCol->szBitmap = - tsCompressTinyint((char *)pColData->pBitMap, BIT2_SIZE(pColData->nVal), BIT2_SIZE(pColData->nVal), - *ppBuf1 + *nDataP + n, size, TWO_STAGE_COMP, *ppBuf2, size); - if (pBlockCol->szBitmap <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } - } else { - pBlockCol->szBitmap = 0; + pSmaInfo->size += tPutColumnDataAgg(pWriter->aBuf[0] + pSmaInfo->size, &sma); } - n += pBlockCol->szBitmap; - // OFFSET - if (IS_VAR_DATA_TYPE(pColData->type)) { - size = sizeof(int32_t) * pColData->nVal + COMP_OVERFLOW_BYTES; + // write + if (pSmaInfo->size) { + int32_t size = pSmaInfo->size + sizeof(TSCKSUM); - code = tRealloc(ppBuf1, *nDataP + n + size); + code = tRealloc(&pWriter->aBuf[0], size); if (code) goto _err; - code = tRealloc(ppBuf2, size); - if (code) goto _err; + taosCalcChecksumAppend(0, pWriter->aBuf[0], size); - pBlockCol->szOffset = tsCompressInt((char *)pColData->aOffset, sizeof(int32_t) * pColData->nVal, pColData->nVal, - *ppBuf1 + *nDataP + n, size, TWO_STAGE_COMP, *ppBuf2, size); - if (pBlockCol->szOffset <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; + int64_t n = taosWriteFile(pWriter->pSmaFD, pWriter->aBuf[0], size); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); goto _err; } - } else { - pBlockCol->szOffset = 0; - } - n += pBlockCol->szOffset; - // VALUE - if (pSubBlock->cmprAlg == NO_COMPRESSION) { - pBlockCol->szValue = pColData->nData; - - code = tRealloc(ppBuf1, *nDataP + n + pBlockCol->szValue + sizeof(TSCKSUM)); - if (code) goto _err; - - memcpy(*ppBuf1 + *nDataP + n, pColData->pData, pBlockCol->szValue); - } else { - size = pColData->nData + COMP_OVERFLOW_BYTES; - - code = tRealloc(ppBuf1, *nDataP + n + size + sizeof(TSCKSUM)); - if (code) goto _err; - - if (pSubBlock->cmprAlg == TWO_STAGE_COMP) { - code = tRealloc(ppBuf2, size); - if (code) goto _err; - } - - pBlockCol->szValue = - tDataTypes[pColData->type].compFunc((char *)pColData->pData, pColData->nData, pColData->nVal, - *ppBuf1 + *nDataP + n, size, pSubBlock->cmprAlg, *ppBuf2, size); - if (pBlockCol->szValue <= 0) { - code = TSDB_CODE_COMPRESS_ERROR; - goto _err; - } + pSmaInfo->offset = pWriter->fSma.size; + pWriter->fSma.size += size; } - n += pBlockCol->szValue; - pBlockCol->szOrigin = pColData->nData; - - // checksum - n += sizeof(TSCKSUM); - taosCalcChecksumAppend(0, *ppBuf1 + *nDataP, n); - - *nDataP += n; return code; _err: + tsdbError("vgId:%d tsdb write block sma failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); return code; } -static int32_t tsdbWriteBlockDataImpl(TdFilePtr pFD, SSubBlock *pSubBlock, SBlockDataHdr hdr, SArray *aBlockCol, - uint8_t *pData, int64_t nData, uint8_t **ppBuf) { +int32_t tsdbWriteBlockData(SDataFWriter *pWriter, SBlockData *pBlockData, SBlockInfo *pBlkInfo, SSmaInfo *pSmaInfo, + int8_t cmprAlg, int8_t toLast) { int32_t code = 0; - int32_t nBlockCol = taosArrayGetSize(aBlockCol); - int64_t size; - int64_t n; - // HDR + SArray - pSubBlock->szBlockCol = sizeof(hdr); - for (int32_t iBlockCol = 0; iBlockCol < nBlockCol; iBlockCol++) { - pSubBlock->szBlockCol += tPutBlockCol(NULL, taosArrayGet(aBlockCol, iBlockCol)); - } + ASSERT(pBlockData->nRow > 0); - code = tRealloc(ppBuf, pSubBlock->szBlockCol + sizeof(TSCKSUM)); - if (code) goto _err; + pBlkInfo->offset = toLast ? pWriter->fLast.size : pWriter->fData.size; + pBlkInfo->szBlock = 0; + pBlkInfo->szKey = 0; - n = 0; - memcpy(*ppBuf, &hdr, sizeof(hdr)); - n += sizeof(hdr); - for (int32_t iBlockCol = 0; iBlockCol < nBlockCol; iBlockCol++) { - n += tPutBlockCol(*ppBuf + n, taosArrayGet(aBlockCol, iBlockCol)); - } - taosCalcChecksumAppend(0, *ppBuf, pSubBlock->szBlockCol + sizeof(TSCKSUM)); + int32_t aBufN[4] = {0}; + code = tCmprBlockData(pBlockData, cmprAlg, NULL, NULL, pWriter->aBuf, aBufN); + if (code) goto _err; - ASSERT(n == pSubBlock->szBlockCol); + // write ================= + TdFilePtr pFD = toLast ? pWriter->pLastFD : pWriter->pDataFD; - n = taosWriteFile(pFD, *ppBuf, pSubBlock->szBlockCol + sizeof(TSCKSUM)); - if (n < 0) { - code = TAOS_SYSTEM_ERROR(errno); - goto _err; - } + pBlkInfo->szKey = aBufN[3] + aBufN[2]; + pBlkInfo->szBlock = aBufN[0] + aBufN[1] + aBufN[2] + aBufN[3]; - // SBlockData - n = taosWriteFile(pFD, pData, nData); + int64_t n = taosWriteFile(pFD, pWriter->aBuf[3], aBufN[3]); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; } - return code; - -_err: - return code; -} - -static int32_t tsdbWriteBlockSma(TdFilePtr pFD, SBlockData *pBlockData, SSubBlock *pSubBlock, uint8_t **ppBuf) { - int32_t code = 0; - int64_t n; - SColData *pColData; - - // prepare - pSubBlock->nSma = 0; - for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { - pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); - - if (IS_VAR_DATA_TYPE(pColData->type) || (!pColData->smaOn)) continue; - - pSubBlock->nSma++; - } - if (pSubBlock->nSma == 0) goto _exit; - - // calc - code = tRealloc(ppBuf, sizeof(SColumnDataAgg) * pSubBlock->nSma + sizeof(TSCKSUM)); - if (code) goto _err; - n = 0; - for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { - pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); - - if (IS_VAR_DATA_TYPE(pColData->type) || (!pColData->smaOn)) continue; - - tsdbCalcColDataSMA(pColData, &((SColumnDataAgg *)(*ppBuf))[n]); - n++; - } - taosCalcChecksumAppend(0, *ppBuf, sizeof(SColumnDataAgg) * pSubBlock->nSma + sizeof(TSCKSUM)); - - // write - n = taosWriteFile(pFD, *ppBuf, sizeof(SColumnDataAgg) * pSubBlock->nSma + sizeof(TSCKSUM)); + n = taosWriteFile(pFD, pWriter->aBuf[2], aBufN[2]); if (n < 0) { code = TAOS_SYSTEM_ERROR(errno); goto _err; } -_exit: - return code; - -_err: - return code; -} - -int32_t tsdbWriteBlockData(SDataFWriter *pWriter, SBlockData *pBlockData, uint8_t **ppBuf1, uint8_t **ppBuf2, - SBlockIdx *pBlockIdx, SBlock *pBlock, int8_t cmprAlg) { - int32_t code = 0; - SSubBlock *pSubBlock = &pBlock->aSubBlock[pBlock->nSubBlock++]; - SBlockCol blockCol; - SBlockCol *pBlockCol = &blockCol; - int64_t n; - TdFilePtr pFileFD = pBlock->last ? pWriter->pLastFD : pWriter->pDataFD; - SBlockDataHdr hdr = {.delimiter = TSDB_FILE_DLMT, .suid = pBlockIdx->suid, .uid = pBlockIdx->uid}; - uint8_t *p; - int64_t nData; - uint8_t *pBuf1 = NULL; - uint8_t *pBuf2 = NULL; - SArray *aBlockCol = NULL; - - if (!ppBuf1) ppBuf1 = &pBuf1; - if (!ppBuf2) ppBuf2 = &pBuf2; - - tsdbUpdateBlockInfo(pBlockData, pBlock); - - pSubBlock->nRow = pBlockData->nRow; - pSubBlock->cmprAlg = cmprAlg; - if (pBlock->last) { - pSubBlock->offset = pWriter->fLast.size; - } else { - pSubBlock->offset = pWriter->fData.size; - } - - // ======================= BLOCK DATA ======================= - // TSDBKEY - nData = 0; - code = tsdbWriteBlockDataKey(pSubBlock, pBlockData, ppBuf1, &nData, ppBuf2); - if (code) goto _err; - - // COLUMNS - aBlockCol = taosArrayInit(taosArrayGetSize(pBlockData->aIdx), sizeof(SBlockCol)); - if (aBlockCol == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - int32_t offset = 0; - for (int32_t iCol = 0; iCol < taosArrayGetSize(pBlockData->aIdx); iCol++) { - SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iCol); - - ASSERT(pColData->flag); - - if (pColData->flag == HAS_NONE) continue; - - pBlockCol->cid = pColData->cid; - pBlockCol->type = pColData->type; - pBlockCol->smaOn = pColData->smaOn; - pBlockCol->flag = pColData->flag; - - if (pColData->flag != HAS_NULL) { - code = tsdbWriteColData(pColData, pBlockCol, pSubBlock, ppBuf1, &nData, ppBuf2); - if (code) goto _err; - - pBlockCol->offset = offset; - offset = offset + pBlockCol->szBitmap + pBlockCol->szOffset + pBlockCol->szValue + sizeof(TSCKSUM); + if (aBufN[1]) { + n = taosWriteFile(pFD, pWriter->aBuf[1], aBufN[1]); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _err; } + } - if (taosArrayPush(aBlockCol, pBlockCol) == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; + if (aBufN[0]) { + n = taosWriteFile(pFD, pWriter->aBuf[0], aBufN[0]); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); goto _err; } } - // write - code = tsdbWriteBlockDataImpl(pFileFD, pSubBlock, hdr, aBlockCol, *ppBuf1, nData, ppBuf2); - if (code) goto _err; - - pSubBlock->szBlock = pSubBlock->szBlockCol + sizeof(TSCKSUM) + nData; - if (pBlock->last) { - pWriter->fLast.size += pSubBlock->szBlock; + // update info + if (toLast) { + pWriter->fLast.size += pBlkInfo->szBlock; } else { - pWriter->fData.size += pSubBlock->szBlock; + pWriter->fData.size += pBlkInfo->szBlock; } - // ======================= BLOCK SMA ======================= - pSubBlock->sOffset = 0; - pSubBlock->nSma = 0; - - if (pBlock->nSubBlock > 1 || pBlock->last || pBlock->hasDup) goto _exit; - - code = tsdbWriteBlockSma(pWriter->pSmaFD, pBlockData, pSubBlock, ppBuf1); - if (code) goto _err; - - if (pSubBlock->nSma > 0) { - pSubBlock->sOffset = pWriter->fSma.size; - pWriter->fSma.size += (sizeof(SColumnDataAgg) * pSubBlock->nSma + sizeof(TSCKSUM)); + // ================= SMA ==================== + if (pSmaInfo) { + code = tsdbWriteBlockSma(pWriter, pBlockData, pSmaInfo); + if (code) goto _err; } _exit: - tFree(pBuf1); - tFree(pBuf2); - taosArrayDestroy(aBlockCol); + tsdbTrace("vgId:%d tsdb write block data, suid:%" PRId64 " uid:%" PRId64 " nRow:%d, offset:%" PRId64 " size:%d", + TD_VID(pWriter->pTsdb->pVnode), pBlockData->suid, pBlockData->uid, pBlockData->nRow, pBlkInfo->offset, + pBlkInfo->szBlock); return code; _err: - tsdbError("vgId:%d, write block data failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); - tFree(pBuf1); - tFree(pBuf2); - taosArrayDestroy(aBlockCol); + tsdbError("vgId:%d tsdb write block data failed since %s", TD_VID(pWriter->pTsdb->pVnode), tstrerror(code)); return code; } @@ -2075,4 +1606,4 @@ int32_t tsdbDFileSetCopy(STsdb *pTsdb, SDFileSet *pSetFrom, SDFileSet *pSetTo) { _err: tsdbError("vgId:%d, tsdb DFileSet copy failed since %s", TD_VID(pTsdb->pVnode), tstrerror(code)); return code; -} \ No newline at end of file +} diff --git a/source/dnode/vnode/src/tsdb/tsdbSnapshot.c b/source/dnode/vnode/src/tsdb/tsdbSnapshot.c index c40fb98d62b8dd6b1a76c1f6ec2fd870cbff3624..ab2b2b617a3d36dbc2c86c2a2207cffac8f087f6 100644 --- a/source/dnode/vnode/src/tsdb/tsdbSnapshot.c +++ b/source/dnode/vnode/src/tsdb/tsdbSnapshot.c @@ -27,12 +27,16 @@ struct STsdbSnapReader { int32_t fid; SDataFReader* pDataFReader; SArray* aBlockIdx; // SArray - int32_t iBlockIdx; + SArray* aBlockL; // SArray SBlockIdx* pBlockIdx; - SMapData mBlock; // SMapData - int32_t iBlock; - SBlockData oBlockData; - SBlockData nBlockData; + SBlockL* pBlockL; + + int32_t iBlockIdx; + int32_t iBlockL; + SMapData mBlock; // SMapData + int32_t iBlock; + SBlockData oBlockData; + SBlockData nBlockData; // for del file int8_t delDone; SDelFReader* pDelFReader; @@ -47,114 +51,116 @@ static int32_t tsdbSnapReadData(STsdbSnapReader* pReader, uint8_t** ppData) { while (true) { if (pReader->pDataFReader == NULL) { - SDFileSet* pSet = - taosArraySearch(pReader->fs.aDFileSet, &(SDFileSet){.fid = pReader->fid}, tDFileSetCmprFn, TD_GT); - + // next + SDFileSet dFileSet = {.fid = pReader->fid}; + SDFileSet* pSet = taosArraySearch(pReader->fs.aDFileSet, &dFileSet, tDFileSetCmprFn, TD_GT); if (pSet == NULL) goto _exit; - pReader->fid = pSet->fid; - code = tsdbDataFReaderOpen(&pReader->pDataFReader, pReader->pTsdb, pSet); - if (code) goto _err; - // SBlockIdx - code = tsdbReadBlockIdx(pReader->pDataFReader, pReader->aBlockIdx, NULL); + // load + code = tsdbDataFReaderOpen(&pReader->pDataFReader, pTsdb, pSet); if (code) goto _err; - pReader->iBlockIdx = 0; - pReader->pBlockIdx = NULL; - - tsdbInfo("vgId:%d, vnode snapshot tsdb open data file to read for %s, fid:%d", TD_VID(pTsdb->pVnode), pTsdb->path, - pReader->fid); - } + code = tsdbReadBlockIdx(pReader->pDataFReader, pReader->aBlockIdx); + if (code) goto _err; - while (true) { - if (pReader->pBlockIdx == NULL) { - if (pReader->iBlockIdx >= taosArrayGetSize(pReader->aBlockIdx)) { - tsdbDataFReaderClose(&pReader->pDataFReader); - break; - } + code = tsdbReadBlockL(pReader->pDataFReader, pReader->aBlockL); + if (code) goto _err; + // init + pReader->iBlockIdx = 0; + if (pReader->iBlockIdx < taosArrayGetSize(pReader->aBlockIdx)) { pReader->pBlockIdx = (SBlockIdx*)taosArrayGet(pReader->aBlockIdx, pReader->iBlockIdx); - pReader->iBlockIdx++; - code = tsdbReadBlock(pReader->pDataFReader, pReader->pBlockIdx, &pReader->mBlock, NULL); + code = tsdbReadBlock(pReader->pDataFReader, pReader->pBlockIdx, &pReader->mBlock); if (code) goto _err; pReader->iBlock = 0; + } else { + pReader->pBlockIdx = NULL; } - SBlock block; - SBlock* pBlock = █ + pReader->iBlockL = 0; while (true) { - if (pReader->iBlock >= pReader->mBlock.nItem) { - pReader->pBlockIdx = NULL; + if (pReader->iBlockL >= taosArrayGetSize(pReader->aBlockL)) { + pReader->pBlockL = NULL; break; } - tMapDataGetItemByIdx(&pReader->mBlock, pReader->iBlock, pBlock, tGetBlock); - pReader->iBlock++; - - if (pBlock->minVersion > pReader->ever || pBlock->maxVersion < pReader->sver) continue; - - code = tsdbReadBlockData(pReader->pDataFReader, pReader->pBlockIdx, pBlock, &pReader->oBlockData, NULL, NULL); - if (code) goto _err; - - // filter - tBlockDataReset(&pReader->nBlockData); - for (int32_t iColData = 0; iColData < taosArrayGetSize(pReader->oBlockData.aIdx); iColData++) { - SColData* pColDataO = tBlockDataGetColDataByIdx(&pReader->oBlockData, iColData); - SColData* pColDataN = NULL; - - code = tBlockDataAddColData(&pReader->nBlockData, taosArrayGetSize(pReader->nBlockData.aIdx), &pColDataN); - if (code) goto _err; - - tColDataInit(pColDataN, pColDataO->cid, pColDataO->type, pColDataO->smaOn); + pReader->pBlockL = (SBlockL*)taosArrayGet(pReader->aBlockL, pReader->iBlockL); + if (pReader->pBlockL->minVer <= pReader->ever && pReader->pBlockL->maxVer >= pReader->sver) { + // TODO + break; } - for (int32_t iRow = 0; iRow < pReader->oBlockData.nRow; iRow++) { - TSDBROW row = tsdbRowFromBlockData(&pReader->oBlockData, iRow); - int64_t version = TSDBROW_VERSION(&row); + pReader->iBlockL++; + } - tsdbTrace("vgId:%d, vnode snapshot tsdb read for %s, %" PRId64 "(%" PRId64 " , %" PRId64 ")", - TD_VID(pReader->pTsdb->pVnode), pReader->pTsdb->path, version, pReader->sver, pReader->ever); + tsdbInfo("vgId:%d, vnode snapshot tsdb open data file to read for %s, fid:%d", TD_VID(pTsdb->pVnode), pTsdb->path, + pReader->fid); + } - if (version < pReader->sver || version > pReader->ever) continue; + while (true) { + if (pReader->pBlockIdx && pReader->pBlockL) { + TABLEID id = {.suid = pReader->pBlockL->suid, .uid = pReader->pBlockL->minUid}; + + ASSERT(0); + + // if (tTABLEIDCmprFn(pReader->pBlockIdx, &minId) < 0) { + // // TODO + // } else if (tTABLEIDCmprFn(pReader->pBlockIdx, &maxId) < 0) { + // // TODO + // } else { + // // TODO + // } + } else if (pReader->pBlockIdx) { + while (pReader->iBlock < pReader->mBlock.nItem) { + SBlock block; + tMapDataGetItemByIdx(&pReader->mBlock, pReader->iBlock, &block, tGetBlock); + + if (block.minVer <= pReader->ever && block.maxVer >= pReader->sver) { + // load data (todo) + } - code = tBlockDataAppendRow(&pReader->nBlockData, &row, NULL); - if (code) goto _err; + // next + pReader->iBlock++; + if (*ppData) break; } - if (pReader->nBlockData.nRow <= 0) { - continue; - } + if (pReader->iBlock >= pReader->mBlock.nItem) { + pReader->iBlockIdx++; + if (pReader->iBlockIdx < taosArrayGetSize(pReader->aBlockIdx)) { + pReader->pBlockIdx = (SBlockIdx*)taosArrayGet(pReader->aBlockIdx, pReader->iBlockIdx); - // org data - // compress data (todo) - int32_t size = sizeof(TABLEID) + tPutBlockData(NULL, &pReader->nBlockData); + code = tsdbReadBlock(pReader->pDataFReader, pReader->pBlockIdx, &pReader->mBlock); + if (code) goto _err; - *ppData = taosMemoryMalloc(sizeof(SSnapDataHdr) + size); - if (*ppData == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; + pReader->iBlock = 0; + } else { + pReader->pBlockIdx = NULL; + } } - SSnapDataHdr* pHdr = (SSnapDataHdr*)(*ppData); - pHdr->type = pReader->type; - pHdr->size = size; - - TABLEID* pId = (TABLEID*)(&pHdr[1]); - pId->suid = pReader->pBlockIdx->suid; - pId->uid = pReader->pBlockIdx->uid; - - tPutBlockData((uint8_t*)(&pId[1]), &pReader->nBlockData); + if (*ppData) goto _exit; + } else if (pReader->pBlockL) { + while (pReader->pBlockL) { + if (pReader->pBlockL->minVer <= pReader->ever && pReader->pBlockL->maxVer >= pReader->sver) { + // load data (todo) + } - tsdbInfo("vgId:%d, vnode snapshot read data for %s, fid:%d suid:%" PRId64 " uid:%" PRId64 - " iBlock:%d minVersion:%d maxVersion:%d nRow:%d out of %d size:%d", - TD_VID(pTsdb->pVnode), pTsdb->path, pReader->fid, pReader->pBlockIdx->suid, pReader->pBlockIdx->uid, - pReader->iBlock - 1, pBlock->minVersion, pBlock->maxVersion, pReader->nBlockData.nRow, pBlock->nRow, - size); + // next + pReader->iBlockL++; + if (pReader->iBlockL < taosArrayGetSize(pReader->aBlockL)) { + pReader->pBlockL = (SBlockL*)taosArrayGetSize(pReader->aBlockL); + } else { + pReader->pBlockL = NULL; + } - goto _exit; + if (*ppData) goto _exit; + } + } else { + tsdbDataFReaderClose(&pReader->pDataFReader); + break; } } } @@ -179,11 +185,11 @@ static int32_t tsdbSnapReadDel(STsdbSnapReader* pReader, uint8_t** ppData) { } // open - code = tsdbDelFReaderOpen(&pReader->pDelFReader, pDelFile, pTsdb, NULL); + code = tsdbDelFReaderOpen(&pReader->pDelFReader, pDelFile, pTsdb); if (code) goto _err; // read index - code = tsdbReadDelIdx(pReader->pDelFReader, pReader->aDelIdx, NULL); + code = tsdbReadDelIdx(pReader->pDelFReader, pReader->aDelIdx); if (code) goto _err; pReader->iDelIdx = 0; @@ -199,7 +205,7 @@ static int32_t tsdbSnapReadDel(STsdbSnapReader* pReader, uint8_t** ppData) { pReader->iDelIdx++; - code = tsdbReadDelData(pReader->pDelFReader, pDelIdx, pReader->aDelData, NULL); + code = tsdbReadDelData(pReader->pDelFReader, pDelIdx, pReader->aDelData); if (code) goto _err; int32_t size = 0; @@ -292,10 +298,15 @@ int32_t tsdbSnapReaderOpen(STsdb* pTsdb, int64_t sver, int64_t ever, int8_t type code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } + pReader->aBlockL = taosArrayInit(0, sizeof(SBlockL)); + if (pReader->aBlockL == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } pReader->mBlock = tMapDataInit(); - code = tBlockDataInit(&pReader->oBlockData); + code = tBlockDataCreate(&pReader->oBlockData); if (code) goto _err; - code = tBlockDataInit(&pReader->nBlockData); + code = tBlockDataCreate(&pReader->nBlockData); if (code) goto _err; pReader->aDelIdx = taosArrayInit(0, sizeof(SDelIdx)); @@ -327,10 +338,11 @@ int32_t tsdbSnapReaderClose(STsdbSnapReader** ppReader) { if (pReader->pDataFReader) { tsdbDataFReaderClose(&pReader->pDataFReader); } + taosArrayDestroy(pReader->aBlockL); taosArrayDestroy(pReader->aBlockIdx); tMapDataClear(&pReader->mBlock); - tBlockDataClear(&pReader->oBlockData, 1); - tBlockDataClear(&pReader->nBlockData, 1); + tBlockDataDestroy(&pReader->oBlockData, 1); + tBlockDataDestroy(&pReader->nBlockData, 1); if (pReader->pDelFReader) { tsdbDelFReaderClose(&pReader->pDelFReader); @@ -405,6 +417,7 @@ struct STsdbSnapWriter { int8_t cmprAlg; int64_t commitID; + uint8_t* aBuf[5]; // for data file SBlockData bData; @@ -418,6 +431,9 @@ struct STsdbSnapWriter { SBlockData* pBlockData; int32_t iRow; SBlockData bDataR; + SArray* aBlockL; // SArray + int32_t iBlockL; + SBlockData lDataR; SDataFWriter* pDataFWriter; SBlockIdx* pBlockIdxW; // NULL when no committing table @@ -427,6 +443,7 @@ struct STsdbSnapWriter { SMapData mBlockW; // SMapData SArray* aBlockIdxW; // SArray + SArray* aBlockLW; // SArray // for del file SDelFReader* pDelFReader; @@ -437,25 +454,6 @@ struct STsdbSnapWriter { SArray* aDelIdxW; }; -static int32_t tsdbSnapWriteAppendData(STsdbSnapWriter* pWriter, uint8_t* pData, uint32_t nData) { - int32_t code = 0; - int32_t iRow = 0; // todo - int32_t nRow = 0; // todo - SBlockData* pBlockData = NULL; // todo - - while (iRow < nRow) { - code = tBlockDataAppendRow(&pWriter->bDataW, &tsdbRowFromBlockData(pBlockData, iRow), NULL); - if (code) goto _err; - } - - return code; - -_err: - tsdbError("vgId:%d, tsdb snapshot write append data for %s failed since %s", TD_VID(pWriter->pTsdb->pVnode), - pWriter->pTsdb->path, tstrerror(code)); - return code; -} - static int32_t tsdbSnapWriteTableDataEnd(STsdbSnapWriter* pWriter) { int32_t code = 0; @@ -467,20 +465,21 @@ static int32_t tsdbSnapWriteTableDataEnd(STsdbSnapWriter* pWriter) { if (pWriter->pBlockData) { ASSERT(pWriter->iRow < pWriter->pBlockData->nRow); while (pWriter->iRow < pWriter->pBlockData->nRow) { - code = tBlockDataAppendRow(&pWriter->bDataW, &tsdbRowFromBlockData(pWriter->pBlockData, pWriter->iRow), NULL); + code = tBlockDataAppendRow(&pWriter->bDataW, &tsdbRowFromBlockData(pWriter->pBlockData, pWriter->iRow), NULL, + 0); // todo if (code) goto _err; if (pWriter->bDataW.nRow >= pWriter->maxRow * 4 / 5) { - pWriter->blockW.last = 0; - code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, - &pWriter->blockW, pWriter->cmprAlg); + // pWriter->blockW.last = 0; + // code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, + // &pWriter->blockW, pWriter->cmprAlg); if (code) goto _err; code = tMapDataPutItem(&pWriter->mBlockW, &pWriter->blockW, tPutBlock); if (code) goto _err; tBlockReset(&pWriter->blockW); - tBlockDataClearData(&pWriter->bDataW); + tBlockDataClear(&pWriter->bDataW); } pWriter->iRow++; @@ -489,16 +488,16 @@ static int32_t tsdbSnapWriteTableDataEnd(STsdbSnapWriter* pWriter) { // write remain data if has if (pWriter->bDataW.nRow > 0) { - pWriter->blockW.last = 0; + // pWriter->blockW.last = 0; if (pWriter->bDataW.nRow < pWriter->minRow) { if (pWriter->iBlock > pWriter->mBlock.nItem) { - pWriter->blockW.last = 1; + // pWriter->blockW.last = 1; } } - code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, - &pWriter->blockW, pWriter->cmprAlg); - if (code) goto _err; + // code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, + // &pWriter->blockW, pWriter->cmprAlg); + // if (code) goto _err; code = tMapDataPutItem(&pWriter->mBlockW, &pWriter->blockW, tPutBlock); if (code) goto _err; @@ -510,16 +509,16 @@ static int32_t tsdbSnapWriteTableDataEnd(STsdbSnapWriter* pWriter) { SBlock block; tMapDataGetItemByIdx(&pWriter->mBlock, pWriter->iBlock, &block, tGetBlock); - if (block.last) { - code = tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, &pWriter->bDataR, NULL, NULL); - if (code) goto _err; + // if (block.last) { + // code = tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, &pWriter->bDataR, NULL, NULL); + // if (code) goto _err; - tBlockReset(&block); - block.last = 1; - code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataR, NULL, NULL, pWriter->pBlockIdxW, &block, - pWriter->cmprAlg); - if (code) goto _err; - } + // tBlockReset(&block); + // block.last = 1; + // code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataR, NULL, NULL, pWriter->pBlockIdxW, &block, + // pWriter->cmprAlg); + // if (code) goto _err; + // } code = tMapDataPutItem(&pWriter->mBlockW, &block, tPutBlock); if (code) goto _err; @@ -528,8 +527,8 @@ static int32_t tsdbSnapWriteTableDataEnd(STsdbSnapWriter* pWriter) { } // SBlock - code = tsdbWriteBlock(pWriter->pDataFWriter, &pWriter->mBlockW, NULL, pWriter->pBlockIdxW); - if (code) goto _err; + // code = tsdbWriteBlock(pWriter->pDataFWriter, &pWriter->mBlockW, NULL, pWriter->pBlockIdxW); + // if (code) goto _err; // SBlockIdx if (taosArrayPush(pWriter->aBlockIdxW, pWriter->pBlockIdxW) == NULL) { @@ -550,7 +549,7 @@ _err: static int32_t tsdbSnapMoveWriteTableData(STsdbSnapWriter* pWriter, SBlockIdx* pBlockIdx) { int32_t code = 0; - code = tsdbReadBlock(pWriter->pDataFReader, pBlockIdx, &pWriter->mBlock, NULL); + code = tsdbReadBlock(pWriter->pDataFReader, pBlockIdx, &pWriter->mBlock); if (code) goto _err; // SBlockData @@ -559,16 +558,17 @@ static int32_t tsdbSnapMoveWriteTableData(STsdbSnapWriter* pWriter, SBlockIdx* p for (int32_t iBlock = 0; iBlock < pWriter->mBlock.nItem; iBlock++) { tMapDataGetItemByIdx(&pWriter->mBlock, iBlock, &block, tGetBlock); - if (block.last) { - code = tsdbReadBlockData(pWriter->pDataFReader, pBlockIdx, &block, &pWriter->bDataR, NULL, NULL); - if (code) goto _err; + // if (block.last) { + // code = tsdbReadBlockData(pWriter->pDataFReader, pBlockIdx, &block, &pWriter->bDataR, NULL, NULL); + // if (code) goto _err; - tBlockReset(&block); - block.last = 1; - code = - tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataR, NULL, NULL, pBlockIdx, &block, pWriter->cmprAlg); - if (code) goto _err; - } + // tBlockReset(&block); + // block.last = 1; + // code = + // tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataR, NULL, NULL, pBlockIdx, &block, + // pWriter->cmprAlg); + // if (code) goto _err; + // } code = tMapDataPutItem(&pWriter->mBlockW, &block, tPutBlock); if (code) goto _err; @@ -576,7 +576,7 @@ static int32_t tsdbSnapMoveWriteTableData(STsdbSnapWriter* pWriter, SBlockIdx* p // SBlock SBlockIdx blockIdx = {.suid = pBlockIdx->suid, .uid = pBlockIdx->uid}; - code = tsdbWriteBlock(pWriter->pDataFWriter, &pWriter->mBlockW, NULL, &blockIdx); + code = tsdbWriteBlock(pWriter->pDataFWriter, &pWriter->mBlockW, &blockIdx); if (code) goto _err; // SBlockIdx @@ -601,9 +601,9 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { TSDBROW row; TSDBROW* pRow = &row; - // correct schema - code = tBlockDataCorrectSchema(&pWriter->bDataW, pBlockData); - if (code) goto _err; + // // correct schema + // code = tBlockDataCorrectSchema(&pWriter->bDataW, pBlockData); + // if (code) goto _err; // loop to merge *pRow = tsdbRowFromBlockData(pBlockData, iRow); @@ -618,8 +618,8 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { ASSERT(c); if (c < 0) { - code = tBlockDataAppendRow(&pWriter->bDataW, pRow, NULL); - if (code) goto _err; + // code = tBlockDataAppendRow(&pWriter->bDataW, pRow, NULL); + // if (code) goto _err; iRow++; if (iRow < pWriter->pBlockData->nRow) { @@ -628,8 +628,8 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { pRow = NULL; } } else if (c > 0) { - code = tBlockDataAppendRow(&pWriter->bDataW, &tsdbRowFromBlockData(pWriter->pBlockData, pWriter->iRow), NULL); - if (code) goto _err; + // code = tBlockDataAppendRow(&pWriter->bDataW, &tsdbRowFromBlockData(pWriter->pBlockData, pWriter->iRow), + // NULL); if (code) goto _err; pWriter->iRow++; if (pWriter->iRow >= pWriter->pBlockData->nRow) { @@ -647,16 +647,15 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { tMapDataGetItemByIdx(&pWriter->mBlock, pWriter->iBlock, &block, tGetBlock); - if (block.last) { - pWriter->pBlockData = &pWriter->bDataR; + // if (block.last) { + // pWriter->pBlockData = &pWriter->bDataR; - code = tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, pWriter->pBlockData, NULL, NULL); - if (code) goto _err; - pWriter->iRow = 0; + // code = tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, pWriter->pBlockData, NULL, + // NULL); if (code) goto _err; pWriter->iRow = 0; - pWriter->iBlock++; - break; - } + // pWriter->iBlock++; + // break; + // } c = tsdbKeyCmprFn(&block.maxKey, &key); @@ -664,16 +663,16 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { if (c < 0) { if (pWriter->bDataW.nRow) { - pWriter->blockW.last = 0; - code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, - &pWriter->blockW, pWriter->cmprAlg); - if (code) goto _err; + // pWriter->blockW.last = 0; + // code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, + // &pWriter->blockW, pWriter->cmprAlg); + // if (code) goto _err; code = tMapDataPutItem(&pWriter->mBlockW, &pWriter->blockW, tPutBlock); if (code) goto _err; tBlockReset(&pWriter->blockW); - tBlockDataClearData(&pWriter->bDataW); + tBlockDataClear(&pWriter->bDataW); } code = tMapDataPutItem(&pWriter->mBlockW, &block, tPutBlock); @@ -687,9 +686,10 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { if (c > 0) { pWriter->pBlockData = &pWriter->bDataR; - code = - tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, pWriter->pBlockData, NULL, NULL); - if (code) goto _err; + // code = + // tsdbReadBlockData(pWriter->pDataFReader, pWriter->pBlockIdx, &block, pWriter->pBlockData, NULL, + // NULL); + // if (code) goto _err; pWriter->iRow = 0; pWriter->iBlock++; @@ -700,8 +700,8 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { if (pWriter->pBlockData) continue; - code = tBlockDataAppendRow(&pWriter->bDataW, pRow, NULL); - if (code) goto _err; + // code = tBlockDataAppendRow(&pWriter->bDataW, pRow, NULL); + // if (code) goto _err; iRow++; if (iRow < pBlockData->nRow) { @@ -715,15 +715,15 @@ static int32_t tsdbSnapWriteTableDataImpl(STsdbSnapWriter* pWriter) { if (pWriter->bDataW.nRow < pWriter->maxRow * 4 / 5) continue; _write_block: - code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, - &pWriter->blockW, pWriter->cmprAlg); - if (code) goto _err; + // code = tsdbWriteBlockData(pWriter->pDataFWriter, &pWriter->bDataW, NULL, NULL, pWriter->pBlockIdxW, + // &pWriter->blockW, pWriter->cmprAlg); + // if (code) goto _err; code = tMapDataPutItem(&pWriter->mBlockW, &pWriter->blockW, tPutBlock); if (code) goto _err; tBlockReset(&pWriter->blockW); - tBlockDataClearData(&pWriter->bDataW); + tBlockDataClear(&pWriter->bDataW); } return code; @@ -789,7 +789,7 @@ static int32_t tsdbSnapWriteTableData(STsdbSnapWriter* pWriter, TABLEID id) { } if (pWriter->pBlockIdx) { - code = tsdbReadBlock(pWriter->pDataFReader, pWriter->pBlockIdx, &pWriter->mBlock, NULL); + code = tsdbReadBlock(pWriter->pDataFReader, pWriter->pBlockIdx, &pWriter->mBlock); if (code) goto _err; } else { tMapDataReset(&pWriter->mBlock); @@ -831,9 +831,11 @@ static int32_t tsdbSnapWriteDataEnd(STsdbSnapWriter* pWriter) { if (pWriter->pDataFWriter == NULL) goto _exit; + // finish current table code = tsdbSnapWriteTableDataEnd(pWriter); if (code) goto _err; + // move remain table while (pWriter->iBlockIdx < taosArrayGetSize(pWriter->aBlockIdx)) { code = tsdbSnapMoveWriteTableData(pWriter, (SBlockIdx*)taosArrayGet(pWriter->aBlockIdx, pWriter->iBlockIdx)); if (code) goto _err; @@ -841,8 +843,16 @@ static int32_t tsdbSnapWriteDataEnd(STsdbSnapWriter* pWriter) { pWriter->iBlockIdx++; } - code = tsdbWriteBlockIdx(pWriter->pDataFWriter, pWriter->aBlockIdxW, NULL); - if (code) goto _err; + // write remain stuff + if (taosArrayGetSize(pWriter->aBlockLW) > 0) { + code = tsdbWriteBlockL(pWriter->pDataFWriter, pWriter->aBlockIdxW); + if (code) goto _err; + } + + if (taosArrayGetSize(pWriter->aBlockIdx) > 0) { + code = tsdbWriteBlockIdx(pWriter->pDataFWriter, pWriter->aBlockIdxW); + if (code) goto _err; + } code = tsdbFSUpsertFSet(&pWriter->fs, &pWriter->pDataFWriter->wSet); if (code) goto _err; @@ -866,19 +876,22 @@ _err: } static int32_t tsdbSnapWriteData(STsdbSnapWriter* pWriter, uint8_t* pData, uint32_t nData) { - int32_t code = 0; - STsdb* pTsdb = pWriter->pTsdb; - TABLEID id = *(TABLEID*)(pData + sizeof(SSnapDataHdr)); - int64_t n; + int32_t code = 0; + STsdb* pTsdb = pWriter->pTsdb; + SSnapDataHdr* pHdr = (SSnapDataHdr*)pData; + TABLEID id = *(TABLEID*)(pData + sizeof(SSnapDataHdr)); + int64_t n; // decode SBlockData* pBlockData = &pWriter->bData; - n = tGetBlockData(pData + sizeof(SSnapDataHdr) + sizeof(TABLEID), pBlockData); - ASSERT(n + sizeof(SSnapDataHdr) + sizeof(TABLEID) == nData); + code = tDecmprBlockData(pData + sizeof(SSnapDataHdr) + sizeof(TABLEID), pHdr->size - sizeof(TABLEID), pBlockData, + pWriter->aBuf); + if (code) goto _err; // open file - TSDBKEY keyFirst = tBlockDataFirstKey(pBlockData); - TSDBKEY keyLast = tBlockDataLastKey(pBlockData); + TSDBKEY keyFirst = {.version = pBlockData->aVersion[0], .ts = pBlockData->aTSKEY[0]}; + TSDBKEY keyLast = {.version = pBlockData->aVersion[pBlockData->nRow - 1], + .ts = pBlockData->aTSKEY[pBlockData->nRow - 1]}; int32_t fid = tsdbKeyFid(keyFirst.ts, pWriter->minutes, pWriter->precision); ASSERT(fid == tsdbKeyFid(keyLast.ts, pWriter->minutes, pWriter->precision)); @@ -895,11 +908,15 @@ static int32_t tsdbSnapWriteData(STsdbSnapWriter* pWriter, uint8_t* pData, uint3 code = tsdbDataFReaderOpen(&pWriter->pDataFReader, pTsdb, pSet); if (code) goto _err; - code = tsdbReadBlockIdx(pWriter->pDataFReader, pWriter->aBlockIdx, NULL); + code = tsdbReadBlockIdx(pWriter->pDataFReader, pWriter->aBlockIdx); + if (code) goto _err; + + code = tsdbReadBlockL(pWriter->pDataFReader, pWriter->aBlockL); if (code) goto _err; } else { ASSERT(pWriter->pDataFReader == NULL); taosArrayClear(pWriter->aBlockIdx); + taosArrayClear(pWriter->aBlockL); } pWriter->iBlockIdx = 0; pWriter->pBlockIdx = NULL; @@ -907,7 +924,9 @@ static int32_t tsdbSnapWriteData(STsdbSnapWriter* pWriter, uint8_t* pData, uint3 pWriter->iBlock = 0; pWriter->pBlockData = NULL; pWriter->iRow = 0; + pWriter->iBlockL = 0; tBlockDataReset(&pWriter->bDataR); + tBlockDataReset(&pWriter->lDataR); // write SHeadFile fHead; @@ -928,7 +947,7 @@ static int32_t tsdbSnapWriteData(STsdbSnapWriter* pWriter, uint8_t* pData, uint3 wSet.fid = fid; fHead = (SHeadFile){.commitID = pWriter->commitID, .offset = 0, .size = 0}; fData = (SDataFile){.commitID = pWriter->commitID, .size = 0}; - fLast = (SLastFile){.commitID = pWriter->commitID, .size = 0}; + fLast = (SLastFile){.commitID = pWriter->commitID, .size = 0, .offset = 0}; fSma = (SSmaFile){.commitID = pWriter->commitID, .size = 0}; } @@ -936,6 +955,7 @@ static int32_t tsdbSnapWriteData(STsdbSnapWriter* pWriter, uint8_t* pData, uint3 if (code) goto _err; taosArrayClear(pWriter->aBlockIdxW); + taosArrayClear(pWriter->aBlockLW); tMapDataReset(&pWriter->mBlockW); pWriter->pBlockIdxW = NULL; tBlockDataReset(&pWriter->bDataW); @@ -963,10 +983,10 @@ static int32_t tsdbSnapWriteDel(STsdbSnapWriter* pWriter, uint8_t* pData, uint32 // reader if (pDelFile) { - code = tsdbDelFReaderOpen(&pWriter->pDelFReader, pDelFile, pTsdb, NULL); + code = tsdbDelFReaderOpen(&pWriter->pDelFReader, pDelFile, pTsdb); if (code) goto _err; - code = tsdbReadDelIdx(pWriter->pDelFReader, pWriter->aDelIdxR, NULL); + code = tsdbReadDelIdx(pWriter->pDelFReader, pWriter->aDelIdxR); if (code) goto _err; } @@ -980,60 +1000,57 @@ static int32_t tsdbSnapWriteDel(STsdbSnapWriter* pWriter, uint8_t* pData, uint32 TABLEID id = *(TABLEID*)(pData + sizeof(SSnapDataHdr)); while (true) { - SDelIdx* pDelIdx = NULL; - int64_t n = sizeof(SSnapDataHdr) + sizeof(TABLEID); - SDelData delData; - SDelIdx delIdx; - int8_t toBreak = 0; + if (pWriter->iDelIdx >= taosArrayGetSize(pWriter->aDelIdxR)) break; + if (tTABLEIDCmprFn(taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx), &id) >= 0) break; - if (pWriter->iDelIdx < taosArrayGetSize(pWriter->aDelIdxR)) { - pDelIdx = (SDelIdx*)taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx); - } + SDelIdx* pDelIdx = (SDelIdx*)taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx); - if (pDelIdx) { - int32_t c = tTABLEIDCmprFn(&id, pDelIdx); - if (c < 0) { - goto _new_del; - } else { - code = tsdbReadDelData(pWriter->pDelFReader, pDelIdx, pWriter->aDelData, NULL); - if (code) goto _err; + code = tsdbReadDelData(pWriter->pDelFReader, pDelIdx, pWriter->aDelData); + if (code) goto _err; - pWriter->iDelIdx++; - if (c == 0) { - toBreak = 1; - delIdx = (SDelIdx){.suid = id.suid, .uid = id.uid}; - goto _merge_del; - } else { - delIdx = (SDelIdx){.suid = pDelIdx->suid, .uid = pDelIdx->uid}; - goto _write_del; - } - } + SDelIdx delIdx = *pDelIdx; + code = tsdbWriteDelData(pWriter->pDelFWriter, pWriter->aDelData, &delIdx); + if (code) goto _err; + + if (taosArrayPush(pWriter->aDelIdxW, &delIdx) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; } - _new_del: - toBreak = 1; - delIdx = (SDelIdx){.suid = id.suid, .uid = id.uid}; - taosArrayClear(pWriter->aDelData); + pWriter->iDelIdx++; + } - _merge_del: - while (n < nData) { - n += tGetDelData(pData + n, &delData); - if (taosArrayPush(pWriter->aDelData, &delData) == NULL) { - code = TSDB_CODE_OUT_OF_MEMORY; - goto _err; - } - } + if (pWriter->iDelIdx < taosArrayGetSize(pWriter->aDelIdxR) && + tTABLEIDCmprFn(taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx), &id) == 0) { + SDelIdx* pDelIdx = (SDelIdx*)taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx); - _write_del: - code = tsdbWriteDelData(pWriter->pDelFWriter, pWriter->aDelData, NULL, &delIdx); + code = tsdbReadDelData(pWriter->pDelFReader, pDelIdx, pWriter->aDelData); if (code) goto _err; - if (taosArrayPush(pWriter->aDelIdxW, &delIdx) == NULL) { + pWriter->iDelIdx++; + } else { + taosArrayClear(pWriter->aDelData); + } + + int64_t n = sizeof(SSnapDataHdr) + sizeof(TABLEID); + while (n < nData) { + SDelData delData; + + n += tGetDelData(pData + n, &delData); + + if (taosArrayPush(pWriter->aDelData, &delData) == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } + } - if (toBreak) break; + SDelIdx delIdx = {.suid = id.suid, .uid = id.uid}; + code = tsdbWriteDelData(pWriter->pDelFWriter, pWriter->aDelData, &delIdx); + if (code) goto _err; + + if (taosArrayPush(pWriter->aDelIdxW, &delIdx) == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; } _exit: @@ -1054,11 +1071,11 @@ static int32_t tsdbSnapWriteDelEnd(STsdbSnapWriter* pWriter) { for (; pWriter->iDelIdx < taosArrayGetSize(pWriter->aDelIdxR); pWriter->iDelIdx++) { SDelIdx* pDelIdx = (SDelIdx*)taosArrayGet(pWriter->aDelIdxR, pWriter->iDelIdx); - code = tsdbReadDelData(pWriter->pDelFReader, pDelIdx, pWriter->aDelData, NULL); + code = tsdbReadDelData(pWriter->pDelFReader, pDelIdx, pWriter->aDelData); if (code) goto _err; - SDelIdx delIdx = (SDelIdx){.suid = pDelIdx->suid, .uid = pDelIdx->uid}; - code = tsdbWriteDelData(pWriter->pDelFWriter, pWriter->aDelData, NULL, &delIdx); + SDelIdx delIdx = *pDelIdx; + code = tsdbWriteDelData(pWriter->pDelFWriter, pWriter->aDelData, &delIdx); if (code) goto _err; if (taosArrayPush(pWriter->aDelIdxR, &delIdx) == NULL) { @@ -1117,7 +1134,7 @@ int32_t tsdbSnapWriterOpen(STsdb* pTsdb, int64_t sver, int64_t ever, STsdbSnapWr pWriter->commitID = pTsdb->pVnode->state.commitID; // for data file - code = tBlockDataInit(&pWriter->bData); + code = tBlockDataCreate(&pWriter->bData); if (code) goto _err; pWriter->aBlockIdx = taosArrayInit(0, sizeof(SBlockIdx)); @@ -1125,17 +1142,29 @@ int32_t tsdbSnapWriterOpen(STsdb* pTsdb, int64_t sver, int64_t ever, STsdbSnapWr code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } - code = tBlockDataInit(&pWriter->bDataR); + code = tBlockDataCreate(&pWriter->bDataR); if (code) goto _err; + pWriter->aBlockL = taosArrayInit(0, sizeof(SBlockL)); + if (pWriter->aBlockL == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + pWriter->aBlockIdxW = taosArrayInit(0, sizeof(SBlockIdx)); if (pWriter->aBlockIdxW == NULL) { code = TSDB_CODE_OUT_OF_MEMORY; goto _err; } - code = tBlockDataInit(&pWriter->bDataW); + code = tBlockDataCreate(&pWriter->bDataW); if (code) goto _err; + pWriter->aBlockLW = taosArrayInit(0, sizeof(SBlockL)); + if (pWriter->aBlockLW == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _err; + } + // for del file pWriter->aDelIdxR = taosArrayInit(0, sizeof(SDelIdx)); if (pWriter->aDelIdxR == NULL) { @@ -1186,6 +1215,10 @@ int32_t tsdbSnapWriterClose(STsdbSnapWriter** ppWriter, int8_t rollback) { if (code) goto _err; } + for (int32_t iBuf = 0; iBuf < sizeof(pWriter->aBuf) / sizeof(uint8_t*); iBuf++) { + tFree(pWriter->aBuf[iBuf]); + } + tsdbInfo("vgId:%d, vnode snapshot tsdb writer close for %s", TD_VID(pWriter->pTsdb->pVnode), pWriter->pTsdb->path); taosMemoryFree(pWriter); *ppWriter = NULL; @@ -1224,6 +1257,7 @@ int32_t tsdbSnapWrite(STsdbSnapWriter* pWriter, uint8_t* pData, uint32_t nData) _exit: tsdbDebug("vgId:%d, tsdb snapshot write for %s succeed", TD_VID(pWriter->pTsdb->pVnode), pWriter->pTsdb->path); + return code; _err: diff --git a/source/dnode/vnode/src/tsdb/tsdbUtil.c b/source/dnode/vnode/src/tsdb/tsdbUtil.c index 76af751196d3877873eee7265321f34c39f7026f..6db9d5e6f40c5d35e52d90dd86b28f4cb7a94676 100644 --- a/source/dnode/vnode/src/tsdb/tsdbUtil.c +++ b/source/dnode/vnode/src/tsdb/tsdbUtil.c @@ -152,25 +152,6 @@ int32_t tTABLEIDCmprFn(const void *p1, const void *p2) { return 0; } -// TSDBKEY ====================================================== -static FORCE_INLINE int32_t tPutTSDBKEY(uint8_t *p, TSDBKEY *pKey) { - int32_t n = 0; - - n += tPutI64v(p ? p + n : p, pKey->version); - n += tPutI64(p ? p + n : p, pKey->ts); - - return n; -} - -static FORCE_INLINE int32_t tGetTSDBKEY(uint8_t *p, TSDBKEY *pKey) { - int32_t n = 0; - - n += tGetI64v(p + n, &pKey->version); - n += tGetI64(p + n, &pKey->ts); - - return n; -} - // SBlockIdx ====================================================== int32_t tPutBlockIdx(uint8_t *p, void *ph) { int32_t n = 0; @@ -215,34 +196,51 @@ int32_t tCmprBlockIdx(void const *lhs, void const *rhs) { return 0; } +int32_t tCmprBlockL(void const *lhs, void const *rhs) { + SBlockIdx *lBlockIdx = (SBlockIdx *)lhs; + SBlockL *rBlockL = (SBlockL *)rhs; + + if (lBlockIdx->suid < rBlockL->suid) { + return -1; + } else if (lBlockIdx->suid > rBlockL->suid) { + return 1; + } + + if (lBlockIdx->uid < rBlockL->minUid) { + return -1; + } else if (lBlockIdx->uid > rBlockL->maxUid) { + return 1; + } + + return 0; +} + // SBlock ====================================================== void tBlockReset(SBlock *pBlock) { - *pBlock = - (SBlock){.minKey = TSDBKEY_MAX, .maxKey = TSDBKEY_MIN, .minVersion = VERSION_MAX, .maxVersion = VERSION_MIN}; + *pBlock = (SBlock){.minKey = TSDBKEY_MAX, .maxKey = TSDBKEY_MIN, .minVer = VERSION_MAX, .maxVer = VERSION_MIN}; } int32_t tPutBlock(uint8_t *p, void *ph) { int32_t n = 0; SBlock *pBlock = (SBlock *)ph; - n += tPutTSDBKEY(p ? p + n : p, &pBlock->minKey); - n += tPutTSDBKEY(p ? p + n : p, &pBlock->maxKey); - n += tPutI64v(p ? p + n : p, pBlock->minVersion); - n += tPutI64v(p ? p + n : p, pBlock->maxVersion); + n += tPutI64v(p ? p + n : p, pBlock->minKey.version); + n += tPutI64v(p ? p + n : p, pBlock->minKey.ts); + n += tPutI64v(p ? p + n : p, pBlock->maxKey.version); + n += tPutI64v(p ? p + n : p, pBlock->maxKey.ts); + n += tPutI64v(p ? p + n : p, pBlock->minVer); + n += tPutI64v(p ? p + n : p, pBlock->maxVer); n += tPutI32v(p ? p + n : p, pBlock->nRow); - n += tPutI8(p ? p + n : p, pBlock->last); n += tPutI8(p ? p + n : p, pBlock->hasDup); n += tPutI8(p ? p + n : p, pBlock->nSubBlock); for (int8_t iSubBlock = 0; iSubBlock < pBlock->nSubBlock; iSubBlock++) { - n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].nRow); - n += tPutI8(p ? p + n : p, pBlock->aSubBlock[iSubBlock].cmprAlg); n += tPutI64v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].offset); - n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].szBlockCol); - n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].szVersion); - n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].szTSKEY); n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].szBlock); - n += tPutI64v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].sOffset); - n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].nSma); + n += tPutI32v(p ? p + n : p, pBlock->aSubBlock[iSubBlock].szKey); + } + if (pBlock->nSubBlock == 1 && !pBlock->hasDup) { + n += tPutI64v(p ? p + n : p, pBlock->smaInfo.offset); + n += tPutI32v(p ? p + n : p, pBlock->smaInfo.size); } return n; @@ -252,24 +250,26 @@ int32_t tGetBlock(uint8_t *p, void *ph) { int32_t n = 0; SBlock *pBlock = (SBlock *)ph; - n += tGetTSDBKEY(p + n, &pBlock->minKey); - n += tGetTSDBKEY(p + n, &pBlock->maxKey); - n += tGetI64v(p + n, &pBlock->minVersion); - n += tGetI64v(p + n, &pBlock->maxVersion); + n += tGetI64v(p + n, &pBlock->minKey.version); + n += tGetI64v(p + n, &pBlock->minKey.ts); + n += tGetI64v(p + n, &pBlock->maxKey.version); + n += tGetI64v(p + n, &pBlock->maxKey.ts); + n += tGetI64v(p + n, &pBlock->minVer); + n += tGetI64v(p + n, &pBlock->maxVer); n += tGetI32v(p + n, &pBlock->nRow); - n += tGetI8(p + n, &pBlock->last); n += tGetI8(p + n, &pBlock->hasDup); n += tGetI8(p + n, &pBlock->nSubBlock); for (int8_t iSubBlock = 0; iSubBlock < pBlock->nSubBlock; iSubBlock++) { - n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].nRow); - n += tGetI8(p + n, &pBlock->aSubBlock[iSubBlock].cmprAlg); n += tGetI64v(p + n, &pBlock->aSubBlock[iSubBlock].offset); - n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].szBlockCol); - n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].szVersion); - n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].szTSKEY); n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].szBlock); - n += tGetI64v(p + n, &pBlock->aSubBlock[iSubBlock].sOffset); - n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].nSma); + n += tGetI32v(p + n, &pBlock->aSubBlock[iSubBlock].szKey); + } + if (pBlock->nSubBlock == 1 && !pBlock->hasDup) { + n += tGetI64v(p + n, &pBlock->smaInfo.offset); + n += tGetI32v(p + n, &pBlock->smaInfo.size); + } else { + pBlock->smaInfo.offset = 0; + pBlock->smaInfo.size = 0; } return n; @@ -290,10 +290,48 @@ int32_t tBlockCmprFn(const void *p1, const void *p2) { bool tBlockHasSma(SBlock *pBlock) { if (pBlock->nSubBlock > 1) return false; - if (pBlock->last) return false; if (pBlock->hasDup) return false; - return pBlock->aSubBlock[0].nSma > 0; + return pBlock->smaInfo.size > 0; +} + +// SBlockL ====================================================== +int32_t tPutBlockL(uint8_t *p, void *ph) { + int32_t n = 0; + SBlockL *pBlockL = (SBlockL *)ph; + + n += tPutI64(p ? p + n : p, pBlockL->suid); + n += tPutI64(p ? p + n : p, pBlockL->minUid); + n += tPutI64(p ? p + n : p, pBlockL->maxUid); + n += tPutI64v(p ? p + n : p, pBlockL->minKey); + n += tPutI64v(p ? p + n : p, pBlockL->maxKey); + n += tPutI64v(p ? p + n : p, pBlockL->minVer); + n += tPutI64v(p ? p + n : p, pBlockL->maxVer); + n += tPutI32v(p ? p + n : p, pBlockL->nRow); + n += tPutI64v(p ? p + n : p, pBlockL->bInfo.offset); + n += tPutI32v(p ? p + n : p, pBlockL->bInfo.szBlock); + n += tPutI32v(p ? p + n : p, pBlockL->bInfo.szKey); + + return n; +} + +int32_t tGetBlockL(uint8_t *p, void *ph) { + int32_t n = 0; + SBlockL *pBlockL = (SBlockL *)ph; + + n += tGetI64(p + n, &pBlockL->suid); + n += tGetI64(p + n, &pBlockL->minUid); + n += tGetI64(p + n, &pBlockL->maxUid); + n += tGetI64v(p + n, &pBlockL->minKey); + n += tGetI64v(p + n, &pBlockL->maxKey); + n += tGetI64v(p + n, &pBlockL->minVer); + n += tGetI64v(p + n, &pBlockL->maxVer); + n += tGetI32v(p + n, &pBlockL->nRow); + n += tGetI64v(p + n, &pBlockL->bInfo.offset); + n += tGetI32v(p + n, &pBlockL->bInfo.szBlock); + n += tGetI32v(p + n, &pBlockL->bInfo.szKey); + + return n; } // SBlockCol ====================================================== @@ -307,15 +345,25 @@ int32_t tPutBlockCol(uint8_t *p, void *ph) { n += tPutI8(p ? p + n : p, pBlockCol->type); n += tPutI8(p ? p + n : p, pBlockCol->smaOn); n += tPutI8(p ? p + n : p, pBlockCol->flag); + n += tPutI32v(p ? p + n : p, pBlockCol->szOrigin); if (pBlockCol->flag != HAS_NULL) { + if (pBlockCol->flag != HAS_VALUE) { + n += tPutI32v(p ? p + n : p, pBlockCol->szBitmap); + } + + if (IS_VAR_DATA_TYPE(pBlockCol->type)) { + n += tPutI32v(p ? p + n : p, pBlockCol->szOffset); + } + + if (pBlockCol->flag != (HAS_NULL | HAS_NONE)) { + n += tPutI32v(p ? p + n : p, pBlockCol->szValue); + } + n += tPutI32v(p ? p + n : p, pBlockCol->offset); - n += tPutI32v(p ? p + n : p, pBlockCol->szBitmap); - n += tPutI32v(p ? p + n : p, pBlockCol->szOffset); - n += tPutI32v(p ? p + n : p, pBlockCol->szValue); - n += tPutI32v(p ? p + n : p, pBlockCol->szOrigin); } +_exit: return n; } @@ -327,15 +375,29 @@ int32_t tGetBlockCol(uint8_t *p, void *ph) { n += tGetI8(p + n, &pBlockCol->type); n += tGetI8(p + n, &pBlockCol->smaOn); n += tGetI8(p + n, &pBlockCol->flag); + n += tGetI32v(p + n, &pBlockCol->szOrigin); ASSERT(pBlockCol->flag && (pBlockCol->flag != HAS_NONE)); + pBlockCol->szBitmap = 0; + pBlockCol->szOffset = 0; + pBlockCol->szValue = 0; + pBlockCol->offset = 0; + if (pBlockCol->flag != HAS_NULL) { + if (pBlockCol->flag != HAS_VALUE) { + n += tGetI32v(p + n, &pBlockCol->szBitmap); + } + + if (IS_VAR_DATA_TYPE(pBlockCol->type)) { + n += tGetI32v(p + n, &pBlockCol->szOffset); + } + + if (pBlockCol->flag != (HAS_NULL | HAS_NONE)) { + n += tGetI32v(p + n, &pBlockCol->szValue); + } + n += tGetI32v(p + n, &pBlockCol->offset); - n += tGetI32v(p + n, &pBlockCol->szBitmap); - n += tGetI32v(p + n, &pBlockCol->szOffset); - n += tGetI32v(p + n, &pBlockCol->szValue); - n += tGetI32v(p + n, &pBlockCol->szOrigin); } return n; @@ -866,6 +928,9 @@ int32_t tColDataAppendValue(SColData *pColData, SColVal *pColVal) { size = BIT2_SIZE(pColData->nVal + 1); code = tRealloc(&pColData->pBitMap, size); if (code) goto _exit; + if ((pColData->nVal & 3) == 0) { + pColData->pBitMap[pColData->nVal >> 2] = 0; + } // put value if (pColVal->isNone) { @@ -910,13 +975,14 @@ int32_t tColDataCopy(SColData *pColDataSrc, SColData *pColDataDest) { int32_t size; ASSERT(pColDataSrc->nVal > 0); + ASSERT(pColDataDest->cid = pColDataSrc->cid); + ASSERT(pColDataDest->type = pColDataSrc->type); - pColDataDest->cid = pColDataSrc->cid; - pColDataDest->type = pColDataSrc->type; pColDataDest->smaOn = pColDataSrc->smaOn; pColDataDest->nVal = pColDataSrc->nVal; pColDataDest->flag = pColDataSrc->flag; + // bitmap if (pColDataSrc->flag != HAS_NONE && pColDataSrc->flag != HAS_NULL && pColDataSrc->flag != HAS_VALUE) { size = BIT2_SIZE(pColDataSrc->nVal); code = tRealloc(&pColDataDest->pBitMap, size); @@ -924,6 +990,7 @@ int32_t tColDataCopy(SColData *pColDataSrc, SColData *pColDataDest) { memcpy(pColDataDest->pBitMap, pColDataSrc->pBitMap, size); } + // offset if (IS_VAR_DATA_TYPE(pColDataDest->type)) { size = sizeof(int32_t) * pColDataSrc->nVal; @@ -933,9 +1000,10 @@ int32_t tColDataCopy(SColData *pColDataSrc, SColData *pColDataDest) { memcpy(pColDataDest->aOffset, pColDataSrc->aOffset, size); } + // value + pColDataDest->nData = pColDataSrc->nData; code = tRealloc(&pColDataDest->pData, pColDataSrc->nData); if (code) goto _exit; - pColDataDest->nData = pColDataSrc->nData; memcpy(pColDataDest->pData, pColDataSrc->pData, pColDataDest->nData); _exit: @@ -1068,10 +1136,13 @@ static FORCE_INLINE int32_t tColDataCmprFn(const void *p1, const void *p2) { } // SBlockData ====================================================== -int32_t tBlockDataInit(SBlockData *pBlockData) { +int32_t tBlockDataCreate(SBlockData *pBlockData) { int32_t code = 0; + pBlockData->suid = 0; + pBlockData->uid = 0; pBlockData->nRow = 0; + pBlockData->aUid = NULL; pBlockData->aVersion = NULL; pBlockData->aTSKEY = NULL; pBlockData->aIdx = taosArrayInit(0, sizeof(int32_t)); @@ -1090,42 +1161,77 @@ _exit: return code; } -void tBlockDataReset(SBlockData *pBlockData) { - pBlockData->nRow = 0; - taosArrayClear(pBlockData->aIdx); -} - -void tBlockDataClear(SBlockData *pBlockData, int8_t deepClear) { +void tBlockDataDestroy(SBlockData *pBlockData, int8_t deepClear) { + tFree((uint8_t *)pBlockData->aUid); tFree((uint8_t *)pBlockData->aVersion); tFree((uint8_t *)pBlockData->aTSKEY); taosArrayDestroy(pBlockData->aIdx); taosArrayDestroyEx(pBlockData->aColData, deepClear ? tColDataClear : NULL); - pBlockData->aColData = NULL; - pBlockData->aIdx = NULL; - pBlockData->aTSKEY = NULL; + pBlockData->aUid = NULL; pBlockData->aVersion = NULL; + pBlockData->aTSKEY = NULL; + pBlockData->aIdx = NULL; + pBlockData->aColData = NULL; } -int32_t tBlockDataSetSchema(SBlockData *pBlockData, STSchema *pTSchema) { - int32_t code = 0; - SColData *pColData; - STColumn *pTColumn; +int32_t tBlockDataInit(SBlockData *pBlockData, int64_t suid, int64_t uid, STSchema *pTSchema) { + int32_t code = 0; + + ASSERT(suid || uid); + + pBlockData->suid = suid; + pBlockData->uid = uid; + pBlockData->nRow = 0; - tBlockDataReset(pBlockData); + taosArrayClear(pBlockData->aIdx); for (int32_t iColumn = 1; iColumn < pTSchema->numOfCols; iColumn++) { - pTColumn = &pTSchema->columns[iColumn]; + STColumn *pTColumn = &pTSchema->columns[iColumn]; + SColData *pColData; code = tBlockDataAddColData(pBlockData, iColumn - 1, &pColData); if (code) goto _exit; - tColDataInit(pColData, pTColumn->colId, pTColumn->type, (pTColumn->flags & COL_SMA_ON) != 0); + tColDataInit(pColData, pTColumn->colId, pTColumn->type, (pTColumn->flags & COL_SMA_ON) ? 1 : 0); } _exit: return code; } -void tBlockDataClearData(SBlockData *pBlockData) { +int32_t tBlockDataInitEx(SBlockData *pBlockData, SBlockData *pBlockDataFrom) { + int32_t code = 0; + + ASSERT(pBlockDataFrom->suid || pBlockDataFrom->uid); + + pBlockData->suid = pBlockDataFrom->suid; + pBlockData->uid = pBlockDataFrom->uid; + pBlockData->nRow = 0; + + taosArrayClear(pBlockData->aIdx); + for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockDataFrom->aIdx); iColData++) { + SColData *pColDataFrom = tBlockDataGetColDataByIdx(pBlockDataFrom, iColData); + + SColData *pColData; + code = tBlockDataAddColData(pBlockData, iColData, &pColData); + if (code) goto _exit; + + tColDataInit(pColData, pColDataFrom->cid, pColDataFrom->type, pColDataFrom->smaOn); + } + +_exit: + return code; +} + +void tBlockDataReset(SBlockData *pBlockData) { + pBlockData->suid = 0; + pBlockData->uid = 0; + pBlockData->nRow = 0; + taosArrayClear(pBlockData->aIdx); +} + +void tBlockDataClear(SBlockData *pBlockData) { + ASSERT(pBlockData->suid || pBlockData->uid); + pBlockData->nRow = 0; for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); @@ -1159,52 +1265,47 @@ _err: return code; } -int32_t tBlockDataAppendRow(SBlockData *pBlockData, TSDBROW *pRow, STSchema *pTSchema) { +int32_t tBlockDataAppendRow(SBlockData *pBlockData, TSDBROW *pRow, STSchema *pTSchema, int64_t uid) { int32_t code = 0; - // TSDBKEY + ASSERT(pBlockData->suid || pBlockData->uid); + + // uid + if (pBlockData->uid == 0) { + ASSERT(uid); + code = tRealloc((uint8_t **)&pBlockData->aUid, sizeof(int64_t) * (pBlockData->nRow + 1)); + if (code) goto _err; + pBlockData->aUid[pBlockData->nRow] = uid; + } + // version code = tRealloc((uint8_t **)&pBlockData->aVersion, sizeof(int64_t) * (pBlockData->nRow + 1)); if (code) goto _err; + pBlockData->aVersion[pBlockData->nRow] = TSDBROW_VERSION(pRow); + // timestamp code = tRealloc((uint8_t **)&pBlockData->aTSKEY, sizeof(TSKEY) * (pBlockData->nRow + 1)); if (code) goto _err; - pBlockData->aVersion[pBlockData->nRow] = TSDBROW_VERSION(pRow); pBlockData->aTSKEY[pBlockData->nRow] = TSDBROW_TS(pRow); // OTHER - int32_t iColData = 0; - int32_t nColData = taosArrayGetSize(pBlockData->aIdx); - SRowIter iter = {0}; - SRowIter *pIter = &iter; - SColData *pColData; - SColVal *pColVal; - - if (nColData == 0) goto _exit; - - tRowIterInit(pIter, pRow, pTSchema); - pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); - pColVal = tRowIterNext(pIter); - - while (pColData) { - if (pColVal) { - if (pColData->cid == pColVal->cid) { - code = tColDataAppendValue(pColData, pColVal); - if (code) goto _err; - - pColVal = tRowIterNext(pIter); - pColData = ((++iColData) < nColData) ? tBlockDataGetColDataByIdx(pBlockData, iColData) : NULL; - } else if (pColData->cid < pColVal->cid) { - code = tColDataAppendValue(pColData, &COL_VAL_NONE(pColData->cid, pColData->type)); - if (code) goto _err; - - pColData = ((++iColData) < nColData) ? tBlockDataGetColDataByIdx(pBlockData, iColData) : NULL; - } else { - pColVal = tRowIterNext(pIter); - } - } else { + SRowIter rIter = {0}; + SColVal *pColVal; + + tRowIterInit(&rIter, pRow, pTSchema); + pColVal = tRowIterNext(&rIter); + for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { + SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); + + while (pColVal && pColVal->cid < pColData->cid) { + pColVal = tRowIterNext(&rIter); + } + + if (pColVal == NULL || pColVal->cid > pColData->cid) { code = tColDataAppendValue(pColData, &COL_VAL_NONE(pColData->cid, pColData->type)); if (code) goto _err; - - pColData = ((++iColData) < nColData) ? tBlockDataGetColDataByIdx(pBlockData, iColData) : NULL; + } else { + code = tColDataAppendValue(pColData, pColVal); + if (code) goto _err; + pColVal = tRowIterNext(&rIter); } } @@ -1259,128 +1360,111 @@ _exit: int32_t tBlockDataMerge(SBlockData *pBlockData1, SBlockData *pBlockData2, SBlockData *pBlockData) { int32_t code = 0; - // set target - int32_t iColData1 = 0; - int32_t nColData1 = taosArrayGetSize(pBlockData1->aIdx); - int32_t iColData2 = 0; - int32_t nColData2 = taosArrayGetSize(pBlockData2->aIdx); - SColData *pColData1; - SColData *pColData2; - SColData *pColData; - - tBlockDataReset(pBlockData); - while (iColData1 < nColData1 && iColData2 < nColData2) { - pColData1 = tBlockDataGetColDataByIdx(pBlockData1, iColData1); - pColData2 = tBlockDataGetColDataByIdx(pBlockData2, iColData2); - - if (pColData1->cid == pColData2->cid) { - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _exit; - tColDataInit(pColData, pColData2->cid, pColData2->type, pColData2->smaOn); - - iColData1++; - iColData2++; - } else if (pColData1->cid < pColData2->cid) { - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _exit; - tColDataInit(pColData, pColData1->cid, pColData1->type, pColData1->smaOn); - - iColData1++; - } else { - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _exit; - tColDataInit(pColData, pColData2->cid, pColData2->type, pColData2->smaOn); - - iColData2++; - } - } - - while (iColData1 < nColData1) { - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _exit; - tColDataInit(pColData, pColData1->cid, pColData1->type, pColData1->smaOn); + ASSERT(pBlockData->suid == pBlockData1->suid); + ASSERT(pBlockData->uid == pBlockData1->uid); + ASSERT(pBlockData1->nRow > 0); + ASSERT(pBlockData2->nRow > 0); - iColData1++; - } + tBlockDataClear(pBlockData); - while (iColData2 < nColData2) { - code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); - if (code) goto _exit; - tColDataInit(pColData, pColData2->cid, pColData2->type, pColData2->smaOn); + TSDBROW row1 = tsdbRowFromBlockData(pBlockData1, 0); + TSDBROW row2 = tsdbRowFromBlockData(pBlockData2, 0); + TSDBROW *pRow1 = &row1; + TSDBROW *pRow2 = &row2; - iColData2++; - } - - // loop to merge - int32_t iRow1 = 0; - int32_t nRow1 = pBlockData1->nRow; - int32_t iRow2 = 0; - int32_t nRow2 = pBlockData2->nRow; - TSDBROW row1; - TSDBROW row2; - int32_t c; + while (pRow1 && pRow2) { + int32_t c = tsdbRowCmprFn(pRow1, pRow2); - while (iRow1 < nRow1 && iRow2 < nRow2) { - row1 = tsdbRowFromBlockData(pBlockData1, iRow1); - row2 = tsdbRowFromBlockData(pBlockData2, iRow2); - - c = tsdbKeyCmprFn(&TSDBROW_KEY(&row1), &TSDBROW_KEY(&row2)); if (c < 0) { - code = tBlockDataAppendRow(pBlockData, &row1, NULL); + code = tBlockDataAppendRow(pBlockData, pRow1, NULL, + pBlockData1->uid ? pBlockData1->uid : pBlockData1->aUid[pRow1->iRow]); if (code) goto _exit; - iRow1++; + + pRow1->iRow++; + if (pRow1->iRow < pBlockData1->nRow) { + *pRow1 = tsdbRowFromBlockData(pBlockData1, pRow1->iRow); + } else { + pRow1 = NULL; + } } else if (c > 0) { - code = tBlockDataAppendRow(pBlockData, &row2, NULL); + code = tBlockDataAppendRow(pBlockData, pRow2, NULL, + pBlockData2->uid ? pBlockData2->uid : pBlockData2->aUid[pRow2->iRow]); if (code) goto _exit; - iRow2++; + + pRow2->iRow++; + if (pRow2->iRow < pBlockData2->nRow) { + *pRow2 = tsdbRowFromBlockData(pBlockData2, pRow2->iRow); + } else { + pRow2 = NULL; + } } else { ASSERT(0); } } - while (iRow1 < nRow1) { - row1 = tsdbRowFromBlockData(pBlockData1, iRow1); - code = tBlockDataAppendRow(pBlockData, &row1, NULL); + while (pRow1) { + code = tBlockDataAppendRow(pBlockData, pRow1, NULL, + pBlockData1->uid ? pBlockData1->uid : pBlockData1->aUid[pRow1->iRow]); if (code) goto _exit; - iRow1++; + + pRow1->iRow++; + if (pRow1->iRow < pBlockData1->nRow) { + *pRow1 = tsdbRowFromBlockData(pBlockData1, pRow1->iRow); + } else { + pRow1 = NULL; + } } - while (iRow2 < nRow2) { - row2 = tsdbRowFromBlockData(pBlockData2, iRow2); - code = tBlockDataAppendRow(pBlockData, &row2, NULL); + while (pRow2) { + code = tBlockDataAppendRow(pBlockData, pRow2, NULL, + pBlockData2->uid ? pBlockData2->uid : pBlockData2->aUid[pRow2->iRow]); if (code) goto _exit; - iRow2++; + + pRow2->iRow++; + if (pRow2->iRow < pBlockData2->nRow) { + *pRow2 = tsdbRowFromBlockData(pBlockData2, pRow2->iRow); + } else { + pRow2 = NULL; + } } _exit: return code; } -int32_t tBlockDataCopy(SBlockData *pBlockDataSrc, SBlockData *pBlockDataDest) { - int32_t code = 0; - SColData *pColDataSrc; - SColData *pColDataDest; +int32_t tBlockDataCopy(SBlockData *pSrc, SBlockData *pDest) { + int32_t code = 0; + + tBlockDataClear(pDest); - ASSERT(pBlockDataSrc->nRow > 0); + ASSERT(pDest->suid == pSrc->suid); + ASSERT(pDest->uid == pSrc->uid); + ASSERT(taosArrayGetSize(pSrc->aIdx) == taosArrayGetSize(pDest->aIdx)); - tBlockDataReset(pBlockDataDest); + pDest->nRow = pSrc->nRow; + + if (pSrc->uid == 0) { + code = tRealloc((uint8_t **)&pDest->aUid, sizeof(int64_t) * pDest->nRow); + if (code) goto _exit; + memcpy(pDest->aUid, pSrc->aUid, sizeof(int64_t) * pDest->nRow); + } - pBlockDataDest->nRow = pBlockDataSrc->nRow; - // TSDBKEY - code = tRealloc((uint8_t **)&pBlockDataDest->aVersion, sizeof(int64_t) * pBlockDataSrc->nRow); + code = tRealloc((uint8_t **)&pDest->aVersion, sizeof(int64_t) * pDest->nRow); if (code) goto _exit; - code = tRealloc((uint8_t **)&pBlockDataDest->aTSKEY, sizeof(TSKEY) * pBlockDataSrc->nRow); + memcpy(pDest->aVersion, pSrc->aVersion, sizeof(int64_t) * pDest->nRow); + + code = tRealloc((uint8_t **)&pDest->aTSKEY, sizeof(TSKEY) * pDest->nRow); if (code) goto _exit; - memcpy(pBlockDataDest->aVersion, pBlockDataSrc->aVersion, sizeof(int64_t) * pBlockDataSrc->nRow); - memcpy(pBlockDataDest->aTSKEY, pBlockDataSrc->aTSKEY, sizeof(TSKEY) * pBlockDataSrc->nRow); + memcpy(pDest->aTSKEY, pSrc->aTSKEY, sizeof(TSKEY) * pDest->nRow); - // other - for (size_t iColData = 0; iColData < taosArrayGetSize(pBlockDataSrc->aIdx); iColData++) { - pColDataSrc = tBlockDataGetColDataByIdx(pBlockDataSrc, iColData); - code = tBlockDataAddColData(pBlockDataDest, iColData, &pColDataDest); - if (code) goto _exit; + for (int32_t iColData = 0; iColData < taosArrayGetSize(pSrc->aIdx); iColData++) { + SColData *pColSrc = tBlockDataGetColDataByIdx(pSrc, iColData); + SColData *pColDest = tBlockDataGetColDataByIdx(pDest, iColData); - code = tColDataCopy(pColDataSrc, pColDataDest); + ASSERT(pColSrc->cid == pColDest->cid); + ASSERT(pColSrc->type == pColDest->type); + + code = tColDataCopy(pColSrc, pColDest); if (code) goto _exit; } @@ -1416,53 +1500,241 @@ void tBlockDataGetColData(SBlockData *pBlockData, int16_t cid, SColData **ppColD *ppColData = NULL; } -int32_t tPutBlockData(uint8_t *p, SBlockData *pBlockData) { - int32_t n = 0; +int32_t tCmprBlockData(SBlockData *pBlockData, int8_t cmprAlg, uint8_t **ppOut, int32_t *szOut, uint8_t *aBuf[], + int32_t aBufN[]) { + int32_t code = 0; - n += tPutI32v(p ? p + n : p, pBlockData->nRow); - if (p) { - memcpy(p + n, pBlockData->aVersion, sizeof(int64_t) * pBlockData->nRow); + SDiskDataHdr hdr = {.delimiter = TSDB_FILE_DLMT, + .fmtVer = 0, + .suid = pBlockData->suid, + .uid = pBlockData->uid, + .nRow = pBlockData->nRow, + .cmprAlg = cmprAlg}; + + // encode ================= + // columns AND SBlockCol + aBufN[0] = 0; + for (int32_t iColData = 0; iColData < taosArrayGetSize(pBlockData->aIdx); iColData++) { + SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iColData); + + ASSERT(pColData->flag); + + if (pColData->flag == HAS_NONE) continue; + + SBlockCol blockCol = {.cid = pColData->cid, + .type = pColData->type, + .smaOn = pColData->smaOn, + .flag = pColData->flag, + .szOrigin = pColData->nData}; + + if (pColData->flag != HAS_NULL) { + code = tsdbCmprColData(pColData, cmprAlg, &blockCol, &aBuf[0], aBufN[0], &aBuf[2]); + if (code) goto _exit; + + blockCol.offset = aBufN[0]; + aBufN[0] = aBufN[0] + blockCol.szBitmap + blockCol.szOffset + blockCol.szValue + sizeof(TSCKSUM); + } + + code = tRealloc(&aBuf[1], hdr.szBlkCol + tPutBlockCol(NULL, &blockCol)); + if (code) goto _exit; + hdr.szBlkCol += tPutBlockCol(aBuf[1] + hdr.szBlkCol, &blockCol); } - n = n + sizeof(int64_t) * pBlockData->nRow; - if (p) { - memcpy(p + n, pBlockData->aTSKEY, sizeof(TSKEY) * pBlockData->nRow); + + aBufN[1] = 0; + if (hdr.szBlkCol > 0) { + aBufN[1] = hdr.szBlkCol + sizeof(TSCKSUM); + + code = tRealloc(&aBuf[1], aBufN[1]); + if (code) goto _exit; + + taosCalcChecksumAppend(0, aBuf[1], aBufN[1]); } - n = n + sizeof(TSKEY) * pBlockData->nRow; - int32_t nCol = taosArrayGetSize(pBlockData->aIdx); - n += tPutI32v(p ? p + n : p, nCol); - for (int32_t iCol = 0; iCol < nCol; iCol++) { - SColData *pColData = tBlockDataGetColDataByIdx(pBlockData, iCol); - n += tPutColData(p ? p + n : p, pColData); + // uid + version + tskey + aBufN[2] = 0; + if (pBlockData->uid == 0) { + code = tsdbCmprData((uint8_t *)pBlockData->aUid, sizeof(int64_t) * pBlockData->nRow, TSDB_DATA_TYPE_BIGINT, cmprAlg, + &aBuf[2], aBufN[2], &hdr.szUid, &aBuf[3]); + if (code) goto _exit; } + aBufN[2] += hdr.szUid; - return n; + code = tsdbCmprData((uint8_t *)pBlockData->aVersion, sizeof(int64_t) * pBlockData->nRow, TSDB_DATA_TYPE_BIGINT, + cmprAlg, &aBuf[2], aBufN[2], &hdr.szVer, &aBuf[3]); + if (code) goto _exit; + aBufN[2] += hdr.szVer; + + code = tsdbCmprData((uint8_t *)pBlockData->aTSKEY, sizeof(TSKEY) * pBlockData->nRow, TSDB_DATA_TYPE_TIMESTAMP, + cmprAlg, &aBuf[2], aBufN[2], &hdr.szKey, &aBuf[3]); + if (code) goto _exit; + aBufN[2] += hdr.szKey; + + aBufN[2] += sizeof(TSCKSUM); + code = tRealloc(&aBuf[2], aBufN[2]); + if (code) goto _exit; + + // hdr + aBufN[3] = tPutDiskDataHdr(NULL, &hdr); + code = tRealloc(&aBuf[3], aBufN[3]); + if (code) goto _exit; + tPutDiskDataHdr(aBuf[3], &hdr); + taosCalcChecksumAppend(taosCalcChecksum(0, aBuf[3], aBufN[3]), aBuf[2], aBufN[2]); + + // aggragate + if (ppOut) { + *szOut = aBufN[0] + aBufN[1] + aBufN[2] + aBufN[3]; + code = tRealloc(ppOut, *szOut); + if (code) goto _exit; + + memcpy(*ppOut, aBuf[3], aBufN[3]); + memcpy(*ppOut + aBufN[3], aBuf[2], aBufN[2]); + if (aBufN[1]) { + memcpy(*ppOut + aBufN[3] + aBufN[2], aBuf[1], aBufN[1]); + } + if (aBufN[0]) { + memcpy(*ppOut + aBufN[3] + aBufN[2] + aBufN[1], aBuf[0], aBufN[0]); + } + } + +_exit: + return code; } -int32_t tGetBlockData(uint8_t *p, SBlockData *pBlockData) { - int32_t n = 0; +int32_t tDecmprBlockData(uint8_t *pIn, int32_t szIn, SBlockData *pBlockData, uint8_t *aBuf[]) { + int32_t code = 0; + + tBlockDataClear(pBlockData); + + int32_t n = 0; + SDiskDataHdr hdr = {0}; + + // SDiskDataHdr + n += tGetDiskDataHdr(pIn + n, &hdr); + if (!taosCheckChecksumWhole(pIn, n + hdr.szUid + hdr.szVer + hdr.szKey + sizeof(TSCKSUM))) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _exit; + } + ASSERT(hdr.delimiter == TSDB_FILE_DLMT); + + pBlockData->suid = hdr.suid; + pBlockData->uid = hdr.uid; + pBlockData->nRow = hdr.nRow; - tBlockDataReset(pBlockData); + // uid + if (hdr.uid == 0) { + ASSERT(hdr.szUid); + code = tsdbDecmprData(pIn + n, hdr.szUid, TSDB_DATA_TYPE_BIGINT, hdr.cmprAlg, (uint8_t **)&pBlockData->aUid, + sizeof(int64_t) * hdr.nRow, &aBuf[0]); + if (code) goto _exit; + } else { + ASSERT(!hdr.szUid); + } + n += hdr.szUid; - n += tGetI32v(p + n, &pBlockData->nRow); - pBlockData->aVersion = (int64_t *)(p + n); - n = n + sizeof(int64_t) * pBlockData->nRow; - pBlockData->aTSKEY = (TSKEY *)(p + n); - n = n + sizeof(TSKEY) * pBlockData->nRow; + // version + code = tsdbDecmprData(pIn + n, hdr.szVer, TSDB_DATA_TYPE_BIGINT, hdr.cmprAlg, (uint8_t **)&pBlockData->aVersion, + sizeof(int64_t) * hdr.nRow, &aBuf[0]); + if (code) goto _exit; + n += hdr.szVer; + + // TSKEY + code = tsdbDecmprData(pIn + n, hdr.szKey, TSDB_DATA_TYPE_TIMESTAMP, hdr.cmprAlg, (uint8_t **)&pBlockData->aTSKEY, + sizeof(TSKEY) * hdr.nRow, &aBuf[0]); + if (code) goto _exit; + n = n + hdr.szKey + sizeof(TSCKSUM); + + // loop to decode each column data + if (hdr.szBlkCol == 0) goto _exit; + + int32_t nt = 0; + while (nt < hdr.szBlkCol) { + SBlockCol blockCol = {0}; + nt += tGetBlockCol(pIn + n + nt, &blockCol); + ASSERT(nt <= hdr.szBlkCol); - int32_t nCol; - n += tGetI32v(p + n, &nCol); - for (int32_t iCol = 0; iCol < nCol; iCol++) { SColData *pColData; + code = tBlockDataAddColData(pBlockData, taosArrayGetSize(pBlockData->aIdx), &pColData); + if (code) goto _exit; - if (tBlockDataAddColData(pBlockData, iCol, &pColData)) return -1; - n += tGetColData(p + n, pColData); + tColDataInit(pColData, blockCol.cid, blockCol.type, blockCol.smaOn); + if (blockCol.flag == HAS_NULL) { + for (int32_t iRow = 0; iRow < hdr.nRow; iRow++) { + code = tColDataAppendValue(pColData, &COL_VAL_NULL(blockCol.cid, blockCol.type)); + if (code) goto _exit; + } + } else { + code = tsdbDecmprColData(pIn + n + hdr.szBlkCol + sizeof(TSCKSUM) + blockCol.offset, &blockCol, hdr.cmprAlg, + hdr.nRow, pColData, &aBuf[0]); + if (code) goto _exit; + } } +_exit: + return code; +} + +// SDiskDataHdr ============================== +int32_t tPutDiskDataHdr(uint8_t *p, void *ph) { + int32_t n = 0; + SDiskDataHdr *pHdr = (SDiskDataHdr *)ph; + + n += tPutU32(p ? p + n : p, pHdr->delimiter); + n += tPutU32v(p ? p + n : p, pHdr->fmtVer); + n += tPutI64(p ? p + n : p, pHdr->suid); + n += tPutI64(p ? p + n : p, pHdr->uid); + n += tPutI32v(p ? p + n : p, pHdr->szUid); + n += tPutI32v(p ? p + n : p, pHdr->szVer); + n += tPutI32v(p ? p + n : p, pHdr->szKey); + n += tPutI32v(p ? p + n : p, pHdr->szBlkCol); + n += tPutI32v(p ? p + n : p, pHdr->nRow); + n += tPutI8(p ? p + n : p, pHdr->cmprAlg); + + return n; +} + +int32_t tGetDiskDataHdr(uint8_t *p, void *ph) { + int32_t n = 0; + SDiskDataHdr *pHdr = (SDiskDataHdr *)ph; + + n += tGetU32(p + n, &pHdr->delimiter); + n += tGetU32v(p + n, &pHdr->fmtVer); + n += tGetI64(p + n, &pHdr->suid); + n += tGetI64(p + n, &pHdr->uid); + n += tGetI32v(p + n, &pHdr->szUid); + n += tGetI32v(p + n, &pHdr->szVer); + n += tGetI32v(p + n, &pHdr->szKey); + n += tGetI32v(p + n, &pHdr->szBlkCol); + n += tGetI32v(p + n, &pHdr->nRow); + n += tGetI8(p + n, &pHdr->cmprAlg); + return n; } // ALGORITHM ============================== +int32_t tPutColumnDataAgg(uint8_t *p, SColumnDataAgg *pColAgg) { + int32_t n = 0; + + n += tPutI16v(p ? p + n : p, pColAgg->colId); + n += tPutI16v(p ? p + n : p, pColAgg->numOfNull); + n += tPutI64(p ? p + n : p, pColAgg->sum); + n += tPutI64(p ? p + n : p, pColAgg->max); + n += tPutI64(p ? p + n : p, pColAgg->min); + + return n; +} + +int32_t tGetColumnDataAgg(uint8_t *p, SColumnDataAgg *pColAgg) { + int32_t n = 0; + + n += tGetI16v(p + n, &pColAgg->colId); + n += tGetI16v(p + n, &pColAgg->numOfNull); + n += tGetI64(p + n, &pColAgg->sum); + n += tGetI64(p + n, &pColAgg->max); + n += tGetI64(p + n, &pColAgg->min); + + return n; +} + void tsdbCalcColDataSMA(SColData *pColData, SColumnDataAgg *pColAgg) { SColVal colVal; SColVal *pColVal = &colVal; @@ -1532,25 +1804,25 @@ void tsdbCalcColDataSMA(SColData *pColData, SColumnDataAgg *pColAgg) { break; } case TSDB_DATA_TYPE_FLOAT: { - *(double*)(&pColAgg->sum) += colVal.value.f; - if (!minAssigned || *(double*)(&pColAgg->min) > colVal.value.f) { - *(double*)(&pColAgg->min) = colVal.value.f; + *(double *)(&pColAgg->sum) += colVal.value.f; + if (!minAssigned || *(double *)(&pColAgg->min) > colVal.value.f) { + *(double *)(&pColAgg->min) = colVal.value.f; minAssigned = true; } - if (!maxAssigned || *(double*)(&pColAgg->max) < colVal.value.f) { - *(double*)(&pColAgg->max) = colVal.value.f; + if (!maxAssigned || *(double *)(&pColAgg->max) < colVal.value.f) { + *(double *)(&pColAgg->max) = colVal.value.f; maxAssigned = true; } break; } case TSDB_DATA_TYPE_DOUBLE: { - *(double*)(&pColAgg->sum) += colVal.value.d; - if (!minAssigned || *(double*)(&pColAgg->min) > colVal.value.d) { - *(double*)(&pColAgg->min) = colVal.value.d; + *(double *)(&pColAgg->sum) += colVal.value.d; + if (!minAssigned || *(double *)(&pColAgg->min) > colVal.value.d) { + *(double *)(&pColAgg->min) = colVal.value.d; minAssigned = true; } - if (!maxAssigned || *(double*)(&pColAgg->max) < colVal.value.d) { - *(double*)(&pColAgg->max) = colVal.value.d; + if (!maxAssigned || *(double *)(&pColAgg->max) < colVal.value.d) { + *(double *)(&pColAgg->max) = colVal.value.d; maxAssigned = true; } break; @@ -1634,3 +1906,268 @@ void tsdbCalcColDataSMA(SColData *pColData, SColumnDataAgg *pColAgg) { } } } + +int32_t tsdbCmprData(uint8_t *pIn, int32_t szIn, int8_t type, int8_t cmprAlg, uint8_t **ppOut, int32_t nOut, + int32_t *szOut, uint8_t **ppBuf) { + int32_t code = 0; + + ASSERT(szIn > 0 && ppOut); + + if (cmprAlg == NO_COMPRESSION) { + code = tRealloc(ppOut, nOut + szIn); + if (code) goto _exit; + + memcpy(*ppOut + nOut, pIn, szIn); + *szOut = szIn; + } else { + int32_t size = szIn + COMP_OVERFLOW_BYTES; + + code = tRealloc(ppOut, nOut + size); + if (code) goto _exit; + + if (cmprAlg == TWO_STAGE_COMP) { + ASSERT(ppBuf); + code = tRealloc(ppBuf, size); + if (code) goto _exit; + } + + *szOut = + tDataTypes[type].compFunc(pIn, szIn, szIn / tDataTypes[type].bytes, *ppOut + nOut, size, cmprAlg, *ppBuf, size); + if (*szOut <= 0) { + code = TSDB_CODE_COMPRESS_ERROR; + goto _exit; + } + } + +_exit: + return code; +} + +int32_t tsdbDecmprData(uint8_t *pIn, int32_t szIn, int8_t type, int8_t cmprAlg, uint8_t **ppOut, int32_t szOut, + uint8_t **ppBuf) { + int32_t code = 0; + + code = tRealloc(ppOut, szOut); + if (code) goto _exit; + + if (cmprAlg == NO_COMPRESSION) { + ASSERT(szIn == szOut); + memcpy(*ppOut, pIn, szOut); + } else { + if (cmprAlg == TWO_STAGE_COMP) { + code = tRealloc(ppBuf, szOut + COMP_OVERFLOW_BYTES); + if (code) goto _exit; + } + + int32_t size = tDataTypes[type].decompFunc(pIn, szIn, szOut / tDataTypes[type].bytes, *ppOut, szOut, cmprAlg, + *ppBuf, szOut + COMP_OVERFLOW_BYTES); + if (size <= 0) { + code = TSDB_CODE_COMPRESS_ERROR; + goto _exit; + } + + ASSERT(size == szOut); + } + +_exit: + return code; +} + +int32_t tsdbCmprColData(SColData *pColData, int8_t cmprAlg, SBlockCol *pBlockCol, uint8_t **ppOut, int32_t nOut, + uint8_t **ppBuf) { + int32_t code = 0; + + ASSERT(pColData->flag && (pColData->flag != HAS_NONE) && (pColData->flag != HAS_NULL)); + + pBlockCol->szBitmap = 0; + pBlockCol->szOffset = 0; + pBlockCol->szValue = 0; + + int32_t size = 0; + // bitmap + if (pColData->flag != HAS_VALUE) { + uint8_t *pBitMap = pColData->pBitMap; + int32_t szBitMap = BIT2_SIZE(pColData->nVal); + + // BIT2 to BIT1 + if (pColData->flag != (HAS_VALUE | HAS_NULL | HAS_NONE)) { + szBitMap = BIT1_SIZE(pColData->nVal); + pBitMap = taosMemoryCalloc(1, szBitMap); + if (pBitMap == NULL) { + code = TSDB_CODE_OUT_OF_MEMORY; + goto _exit; + } + + for (int32_t iVal = 0; iVal < pColData->nVal; iVal++) { + uint8_t v = GET_BIT2(pColData->pBitMap, iVal); + switch (pColData->flag) { + case (HAS_NULL | HAS_NONE): + SET_BIT1(pBitMap, iVal, v); + break; + case (HAS_VALUE | HAS_NONE): + if (v) { + SET_BIT1(pBitMap, iVal, 1); + } else { + SET_BIT1(pBitMap, iVal, 0); + } + break; + case (HAS_VALUE | HAS_NULL): + SET_BIT1(pBitMap, iVal, v - 1); + break; + default: + ASSERT(0); + } + } + } + + code = tsdbCmprData(pBitMap, szBitMap, TSDB_DATA_TYPE_TINYINT, cmprAlg, ppOut, nOut + size, &pBlockCol->szBitmap, + ppBuf); + if (code) goto _exit; + + if (pColData->flag != (HAS_VALUE | HAS_NULL | HAS_NONE)) { + taosMemoryFree(pBitMap); + } + } + size += pBlockCol->szBitmap; + + // offset + if (IS_VAR_DATA_TYPE(pColData->type)) { + code = tsdbCmprData((uint8_t *)pColData->aOffset, sizeof(int32_t) * pColData->nVal, TSDB_DATA_TYPE_INT, cmprAlg, + ppOut, nOut + size, &pBlockCol->szOffset, ppBuf); + if (code) goto _exit; + } + size += pBlockCol->szOffset; + + // value + if (pColData->flag != (HAS_NULL | HAS_NONE)) { + code = tsdbCmprData((uint8_t *)pColData->pData, pColData->nData, pColData->type, cmprAlg, ppOut, nOut + size, + &pBlockCol->szValue, ppBuf); + if (code) goto _exit; + } + size += pBlockCol->szValue; + + // checksum + size += sizeof(TSCKSUM); + code = tRealloc(ppOut, nOut + size); + if (code) goto _exit; + taosCalcChecksumAppend(0, *ppOut + nOut, size); + +_exit: + return code; +} + +int32_t tsdbDecmprColData(uint8_t *pIn, SBlockCol *pBlockCol, int8_t cmprAlg, int32_t nVal, SColData *pColData, + uint8_t **ppBuf) { + int32_t code = 0; + + int32_t size = pBlockCol->szBitmap + pBlockCol->szOffset + pBlockCol->szValue + sizeof(TSCKSUM); + if (!taosCheckChecksumWhole(pIn, size)) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _exit; + } + + ASSERT(pColData->cid == pBlockCol->cid); + ASSERT(pColData->type == pBlockCol->type); + pColData->smaOn = pBlockCol->smaOn; + pColData->flag = pBlockCol->flag; + pColData->nVal = nVal; + pColData->nData = pBlockCol->szOrigin; + + uint8_t *p = pIn; + // bitmap + if (pBlockCol->szBitmap) { + if (pBlockCol->flag != (HAS_VALUE | HAS_NULL | HAS_NONE)) { + uint8_t *pBitMap = NULL; + code = tsdbDecmprData(p, pBlockCol->szBitmap, TSDB_DATA_TYPE_TINYINT, cmprAlg, &pBitMap, + BIT1_SIZE(pColData->nVal), ppBuf); + if (code) goto _exit; + + code = tRealloc(&pColData->pBitMap, BIT2_SIZE(pColData->nVal)); + if (code) { + tFree(pBitMap); + goto _exit; + } + + // BIT1 to BIT2 + for (int32_t iVal = 0; iVal < nVal; iVal++) { + uint8_t v = GET_BIT1(pBitMap, iVal); + switch (pBlockCol->flag) { + case (HAS_NULL | HAS_NONE): + SET_BIT2(pColData->pBitMap, iVal, v); + break; + case (HAS_VALUE | HAS_NONE): + if (v) { + SET_BIT2(pColData->pBitMap, iVal, 2); + } else { + SET_BIT2(pColData->pBitMap, iVal, 0); + } + break; + case (HAS_VALUE | HAS_NULL): + SET_BIT2(pColData->pBitMap, iVal, v + 1); + break; + default: + ASSERT(0); + } + } + + tFree(pBitMap); + } else { + code = tsdbDecmprData(p, pBlockCol->szBitmap, TSDB_DATA_TYPE_TINYINT, cmprAlg, &pColData->pBitMap, + BIT2_SIZE(pColData->nVal), ppBuf); + if (code) goto _exit; + } + } + p += pBlockCol->szBitmap; + + // offset + if (pBlockCol->szOffset) { + code = tsdbDecmprData(p, pBlockCol->szOffset, TSDB_DATA_TYPE_INT, cmprAlg, (uint8_t **)&pColData->aOffset, + sizeof(int32_t) * pColData->nVal, ppBuf); + if (code) goto _exit; + } + p += pBlockCol->szOffset; + + // value + if (pBlockCol->szValue) { + code = tsdbDecmprData(p, pBlockCol->szValue, pColData->type, cmprAlg, &pColData->pData, pColData->nData, ppBuf); + if (code) goto _exit; + } + p += pBlockCol->szValue; + +_exit: + return code; +} + +int32_t tsdbReadAndCheck(TdFilePtr pFD, int64_t offset, uint8_t **ppOut, int32_t size, int8_t toCheck) { + int32_t code = 0; + + // alloc + code = tRealloc(ppOut, size); + if (code) goto _exit; + + // seek + int64_t n = taosLSeekFile(pFD, offset, SEEK_SET); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _exit; + } + + // read + n = taosReadFile(pFD, *ppOut, size); + if (n < 0) { + code = TAOS_SYSTEM_ERROR(errno); + goto _exit; + } else if (n < size) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _exit; + } + + // check + if (toCheck && !taosCheckChecksumWhole(*ppOut, size)) { + code = TSDB_CODE_FILE_CORRUPTED; + goto _exit; + } + +_exit: + return code; +} diff --git a/source/dnode/vnode/src/vnd/vnodeCommit.c b/source/dnode/vnode/src/vnd/vnodeCommit.c index 64f223b974199b99b88867e9d93ddfed6d9d3bd1..8c73499229b89d67edbc21f86ff7f76b570a4275 100644 --- a/source/dnode/vnode/src/vnd/vnodeCommit.c +++ b/source/dnode/vnode/src/vnd/vnodeCommit.c @@ -220,13 +220,6 @@ int vnodeCommit(SVnode *pVnode) { vInfo("vgId:%d, start to commit, commit ID:%" PRId64 " version:%" PRId64, TD_VID(pVnode), pVnode->state.commitID, pVnode->state.applied); - // preCommit - // smaSyncPreCommit(pVnode->pSma); - smaAsyncPreCommit(pVnode->pSma); - - vnodeBufPoolUnRef(pVnode->inUse); - pVnode->inUse = NULL; - pVnode->state.commitTerm = pVnode->state.applyTerm; // save info @@ -241,6 +234,16 @@ int vnodeCommit(SVnode *pVnode) { } walBeginSnapshot(pVnode->pWal, pVnode->state.applied); + // preCommit + // smaSyncPreCommit(pVnode->pSma); + if(smaAsyncPreCommit(pVnode->pSma) < 0){ + ASSERT(0); + return -1; + } + + vnodeBufPoolUnRef(pVnode->inUse); + pVnode->inUse = NULL; + // commit each sub-system if (metaCommit(pVnode->pMeta) < 0) { ASSERT(0); @@ -248,7 +251,10 @@ int vnodeCommit(SVnode *pVnode) { } if (VND_IS_RSMA(pVnode)) { - smaAsyncCommit(pVnode->pSma); + if (smaAsyncCommit(pVnode->pSma) < 0) { + ASSERT(0); + return -1; + } if (tsdbCommit(VND_RSMA0(pVnode)) < 0) { ASSERT(0); @@ -285,7 +291,10 @@ int vnodeCommit(SVnode *pVnode) { // postCommit // smaSyncPostCommit(pVnode->pSma); - smaAsyncPostCommit(pVnode->pSma); + if (smaAsyncPostCommit(pVnode->pSma) < 0) { + ASSERT(0); + return -1; + } // apply the commit (TODO) walEndSnapshot(pVnode->pWal); diff --git a/source/dnode/vnode/src/vnd/vnodeOpen.c b/source/dnode/vnode/src/vnd/vnodeOpen.c index a4fd984fb762a934b48489da254b3af1aa4dc908..dcfbd33b903c9fcd55e216bd1b24c73f2845af7b 100644 --- a/source/dnode/vnode/src/vnd/vnodeOpen.c +++ b/source/dnode/vnode/src/vnd/vnodeOpen.c @@ -87,6 +87,7 @@ SVnode *vnodeOpen(const char *path, STfs *pTfs, SMsgCb msgCb) { pVnode->msgCb = msgCb; taosThreadMutexInit(&pVnode->lock, NULL); pVnode->blocked = false; + pVnode->inClose = false; tsem_init(&pVnode->syncSem, 0, 0); tsem_init(&(pVnode->canCommit), 0, 1); @@ -181,6 +182,8 @@ _err: void vnodePreClose(SVnode *pVnode) { if (pVnode) { syncLeaderTransfer(pVnode->sync); + pVnode->inClose = true; + smaPreClose(pVnode->pSma); } } diff --git a/source/dnode/vnode/src/vnd/vnodeSvr.c b/source/dnode/vnode/src/vnd/vnodeSvr.c index c73d2ccfd5216179607df145a50354f1387d7ab0..7a8d168f4f1a2cb4b1379e9d5794ff58e83841bf 100644 --- a/source/dnode/vnode/src/vnd/vnodeSvr.c +++ b/source/dnode/vnode/src/vnd/vnodeSvr.c @@ -301,8 +301,6 @@ int32_t vnodeProcessQueryMsg(SVnode *pVnode, SRpcMsg *pMsg) { return qWorkerProcessQueryMsg(&handle, pVnode->pQuery, pMsg, 0); case TDMT_SCH_QUERY_CONTINUE: return qWorkerProcessCQueryMsg(&handle, pVnode->pQuery, pMsg, 0); - case TDMT_VND_FETCH_RSMA: - return smaProcessFetch(pVnode->pSma, pMsg); case TDMT_VND_EXEC_RSMA: return smaProcessExec(pVnode->pSma, pMsg); default: diff --git a/source/libs/executor/inc/executorimpl.h b/source/libs/executor/inc/executorimpl.h index 311d82c8a28bccfbc92344e75d85099a69f94289..fb4eac991f4d64e2c5477e3b102395ad6c83550b 100644 --- a/source/libs/executor/inc/executorimpl.h +++ b/source/libs/executor/inc/executorimpl.h @@ -205,7 +205,7 @@ typedef struct SExprSupp { } SExprSupp; typedef struct SOperatorInfo { - uint8_t operatorType; + uint16_t operatorType; bool blocking; // block operator or not uint8_t status; // denote if current operator is completed char* name; // name, for debug purpose @@ -434,7 +434,7 @@ typedef struct SStreamAggSupporter { typedef struct SessionWindowSupporter { SStreamAggSupporter* pStreamAggSup; int64_t gap; - uint8_t parentType; + uint16_t parentType; SAggSupporter* pIntervalAggSup; } SessionWindowSupporter; @@ -860,8 +860,8 @@ int32_t handleLimitOffset(SOperatorInfo *pOperator, SLimitInfo* pLimitInfo, SSDa bool hasLimitOffsetInfo(SLimitInfo* pLimitInfo); void initLimitInfo(const SNode* pLimit, const SNode* pSLimit, SLimitInfo* pLimitInfo); -void doApplyFunctions(SExecTaskInfo* taskInfo, SqlFunctionCtx* pCtx, STimeWindow* pWin, SColumnInfoData* pTimeWindowData, int32_t offset, - int32_t forwardStep, TSKEY* tsCol, int32_t numOfTotal, int32_t numOfOutput, int32_t order); +void doApplyFunctions(SExecTaskInfo* taskInfo, SqlFunctionCtx* pCtx, SColumnInfoData* pTimeWindowData, int32_t offset, + int32_t forwardStep, int32_t numOfTotal, int32_t numOfOutput); int32_t extractDataBlockFromFetchRsp(SSDataBlock* pRes, char* pData, int32_t numOfOutput, SArray* pColList, char** pNextStart); void updateLoadRemoteInfo(SLoadRemoteDataInfo *pInfo, int32_t numOfRows, int32_t dataLen, int64_t startTs, @@ -924,9 +924,6 @@ SOperatorInfo* createMergeAlignedIntervalOperatorInfo(SOperatorInfo* downstream, SOperatorInfo* createStreamFinalIntervalOperatorInfo(SOperatorInfo* downstream, SPhysiNode* pPhyNode, SExecTaskInfo* pTaskInfo, int32_t numOfChild); -SOperatorInfo* createStreamIntervalOperatorInfo(SOperatorInfo* downstream, SExprInfo* pExprInfo, int32_t numOfCols, - SSDataBlock* pResBlock, SInterval* pInterval, int32_t primaryTsSlotId, - STimeWindowAggSupp *pTwAggSupp, SExecTaskInfo* pTaskInfo); SOperatorInfo* createSessionAggOperatorInfo(SOperatorInfo* downstream, SSessionWinodwPhysiNode* pSessionNode, SExecTaskInfo* pTaskInfo); SOperatorInfo* createGroupOperatorInfo(SOperatorInfo* downstream, SExprInfo* pExprInfo, int32_t numOfCols, diff --git a/source/libs/executor/src/executil.c b/source/libs/executor/src/executil.c index baeb972e0505630950e5a441f54fc8e4f3859160..bf969bf2e4855a12a9abfbe5e67301c5df5f3702 100644 --- a/source/libs/executor/src/executil.c +++ b/source/libs/executor/src/executil.c @@ -429,7 +429,9 @@ static SColumnInfoData* getColInfoResult(void* metaHandle, uint64_t suid, SArray for (int32_t i = 0; i < rows; i++) { int64_t* uid = taosArrayGet(uidList, i); void* tag = taosHashGet(tags, uid, sizeof(int64_t)); - ASSERT(tag); + if (suid != 0) { + ASSERT(tag); + } for(int32_t j = 0; j < taosArrayGetSize(pResBlock->pDataBlock); j++){ SColumnInfoData* pColInfo = (SColumnInfoData*)taosArrayGet(pResBlock->pDataBlock, j); diff --git a/source/libs/executor/src/executor.c b/source/libs/executor/src/executor.c index 4c3d5cf7affbc291bcc9f62adcff688e149a414a..d8f63cb0082f4822a7b025adf1733510e2cf91a5 100644 --- a/source/libs/executor/src/executor.c +++ b/source/libs/executor/src/executor.c @@ -348,7 +348,7 @@ int32_t qCreateExecTask(SReadHandle* readHandle, int32_t vgId, uint64_t taskId, taosThreadOnce(&initPoolOnce, initRefPool); atexit(cleanupRefPool); - qDebug("start to create subplan task, TID:0x%"PRIx64 " QID:0x%"PRIx64, taskId, pSubplan->id.queryId); + qDebug("start to create subplan task, TID:0x%" PRIx64 " QID:0x%" PRIx64, taskId, pSubplan->id.queryId); int32_t code = createExecTaskInfoImpl(pSubplan, pTask, readHandle, taskId, sql, model); if (code != TSDB_CODE_SUCCESS) { @@ -374,7 +374,7 @@ int32_t qCreateExecTask(SReadHandle* readHandle, int32_t vgId, uint64_t taskId, } } - qDebug("subplan task create completed, TID:0x%"PRIx64 " QID:0x%"PRIx64, taskId, pSubplan->id.queryId); + qDebug("subplan task create completed, TID:0x%" PRIx64 " QID:0x%" PRIx64, taskId, pSubplan->id.queryId); _error: // if failed to add ref for all tables in this query, abort current query @@ -427,7 +427,7 @@ int waitMoment(SQInfo* pQInfo) { #endif static void freeBlock(void* param) { - SSDataBlock* pBlock = *(SSDataBlock**) param; + SSDataBlock* pBlock = *(SSDataBlock**)param; blockDataDestroy(pBlock); } @@ -467,12 +467,12 @@ int32_t qExecTaskOpt(qTaskInfo_t tinfo, SArray* pResList, uint64_t* useconds) { qDebug("%s execTask is launched", GET_TASKID(pTaskInfo)); - int32_t current = 0; + int32_t current = 0; SSDataBlock* pRes = NULL; int64_t st = taosGetTimestampUs(); - while((pRes = pTaskInfo->pRoot->fpSet.getNextFn(pTaskInfo->pRoot)) != NULL) { + while ((pRes = pTaskInfo->pRoot->fpSet.getNextFn(pTaskInfo->pRoot)) != NULL) { SSDataBlock* p = createOneDataBlock(pRes, true); current += p->info.rows; ASSERT(p->info.rows > 0); @@ -494,7 +494,7 @@ int32_t qExecTaskOpt(qTaskInfo_t tinfo, SArray* pResList, uint64_t* useconds) { uint64_t total = pTaskInfo->pRoot->resultInfo.totalRows; qDebug("%s task suspended, %d rows in %d blocks returned, total:%" PRId64 " rows, in sinkNode:%d, elapsed:%.2f ms", - GET_TASKID(pTaskInfo), current, (int32_t) taosArrayGetSize(pResList), total, 0, el / 1000.0); + GET_TASKID(pTaskInfo), current, (int32_t)taosArrayGetSize(pResList), total, 0, el / 1000.0); atomic_store_64(&pTaskInfo->owner, 0); return pTaskInfo->code; @@ -632,7 +632,7 @@ int32_t qExtractStreamScanner(qTaskInfo_t tinfo, void** scanner) { SOperatorInfo* pOperator = pTaskInfo->pRoot; while (1) { - uint8_t type = pOperator->operatorType; + uint16_t type = pOperator->operatorType; if (type == QUERY_NODE_PHYSICAL_PLAN_STREAM_SCAN) { *scanner = pOperator->info; return 0; @@ -691,7 +691,7 @@ int32_t qStreamPrepareScan(qTaskInfo_t tinfo, const STqOffsetVal* pOffset) { pTaskInfo->streamInfo.prepareStatus = *pOffset; if (!tOffsetEqual(pOffset, &pTaskInfo->streamInfo.lastStatus)) { while (1) { - uint8_t type = pOperator->operatorType; + uint16_t type = pOperator->operatorType; pOperator->status = OP_OPENED; // TODO add more check if (type != QUERY_NODE_PHYSICAL_PLAN_STREAM_SCAN) { diff --git a/source/libs/executor/src/executorimpl.c b/source/libs/executor/src/executorimpl.c index 9ff5b5d75935aee0673c0f4268abfbcf6cd61553..16a1cb898fae0874141aaf8913ee090647e0756d 100644 --- a/source/libs/executor/src/executorimpl.c +++ b/source/libs/executor/src/executorimpl.c @@ -76,6 +76,12 @@ static UNUSED_FUNC void* u_realloc(void* p, size_t __size) { #define realloc u_realloc #endif +#define T_LONG_JMP(_obj, _c) \ + do { \ + assert((_c) != -1); \ + longjmp((_obj), (_c)); \ + } while (0); + #define CLEAR_QUERY_STATUS(q, st) ((q)->status &= (~(st))) #define QUERY_IS_INTERVAL_QUERY(_q) ((_q)->interval.interval > 0) @@ -372,15 +378,30 @@ void initExecTimeWindowInfo(SColumnInfoData* pColData, STimeWindow* pQueryWindow void cleanupExecTimeWindowInfo(SColumnInfoData* pColData) { colDataDestroy(pColData); } -void doApplyFunctions(SExecTaskInfo* taskInfo, SqlFunctionCtx* pCtx, STimeWindow* pWin, - SColumnInfoData* pTimeWindowData, int32_t offset, int32_t forwardStep, TSKEY* tsCol, - int32_t numOfTotal, int32_t numOfOutput, int32_t order) { +typedef struct { + bool hasAgg; + int32_t numOfRows; + int32_t startOffset; +} SFunctionCtxStatus; + +static void functionCtxSave(SqlFunctionCtx* pCtx, SFunctionCtxStatus* pStatus) { + pStatus->hasAgg = pCtx->input.colDataAggIsSet; + pStatus->numOfRows = pCtx->input.numOfRows; + pStatus->startOffset = pCtx->input.startRowIndex; +} + +static void functionCtxRestore(SqlFunctionCtx* pCtx, SFunctionCtxStatus* pStatus) { + pCtx->input.colDataAggIsSet = pStatus->hasAgg; + pCtx->input.numOfRows = pStatus->numOfRows; + pCtx->input.startRowIndex = pStatus->startOffset; +} + +void doApplyFunctions(SExecTaskInfo* taskInfo, SqlFunctionCtx* pCtx, SColumnInfoData* pTimeWindowData, int32_t offset, + int32_t forwardStep, int32_t numOfTotal, int32_t numOfOutput) { for (int32_t k = 0; k < numOfOutput; ++k) { // keep it temporarily - // todo no need this?? - bool hasAgg = pCtx[k].input.colDataAggIsSet; - int32_t numOfRows = pCtx[k].input.numOfRows; - int32_t startOffset = pCtx[k].input.startRowIndex; + SFunctionCtxStatus status = {0}; + functionCtxSave(&pCtx[k], &status); pCtx[k].input.startRowIndex = offset; pCtx[k].input.numOfRows = forwardStep; @@ -418,9 +439,7 @@ void doApplyFunctions(SExecTaskInfo* taskInfo, SqlFunctionCtx* pCtx, STimeWindow } // restore it - pCtx[k].input.colDataAggIsSet = hasAgg; - pCtx[k].input.startRowIndex = startOffset; - pCtx[k].input.numOfRows = numOfRows; + functionCtxRestore(&pCtx[k], &status); } } } @@ -3131,6 +3150,7 @@ int32_t aggDecodeResultRow(SOperatorInfo* pOperator, char* result) { initResultRow(resultRow); pInfo->resultRowInfo.cur = (SResultRowPosition){.pageId = resultRow->pageId, .offset = resultRow->offset}; + // releaseBufPage(pSup->pResultBuf, getBufPage(pSup->pResultBuf, pageId)); } if (offset != length) { diff --git a/source/libs/executor/src/groupoperator.c b/source/libs/executor/src/groupoperator.c index 507719e0aac3d8a8224828433cc5c66445dea0c1..05dffc658b29bb5eb6675edae62d04bb6442cc48 100644 --- a/source/libs/executor/src/groupoperator.c +++ b/source/libs/executor/src/groupoperator.c @@ -277,7 +277,7 @@ static void doHashGroupbyAgg(SOperatorInfo* pOperator, SSDataBlock* pBlock) { } int32_t rowIndex = j - num; - doApplyFunctions(pTaskInfo, pCtx, &w, NULL, rowIndex, num, NULL, pBlock->info.rows, pOperator->exprSupp.numOfExprs, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pCtx, NULL, rowIndex, num, pBlock->info.rows, pOperator->exprSupp.numOfExprs); // assign the group keys or user input constant values if required doAssignGroupKeys(pCtx, pOperator->exprSupp.numOfExprs, pBlock->info.rows, rowIndex); @@ -295,7 +295,7 @@ static void doHashGroupbyAgg(SOperatorInfo* pOperator, SSDataBlock* pBlock) { } int32_t rowIndex = pBlock->info.rows - num; - doApplyFunctions(pTaskInfo, pCtx, &w, NULL, rowIndex, num, NULL, pBlock->info.rows, pOperator->exprSupp.numOfExprs, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pCtx, NULL, rowIndex, num, pBlock->info.rows, pOperator->exprSupp.numOfExprs); doAssignGroupKeys(pCtx, pOperator->exprSupp.numOfExprs, pBlock->info.rows, rowIndex); } } diff --git a/source/libs/executor/src/timewindowoperator.c b/source/libs/executor/src/timewindowoperator.c index 02be01cdb831d2bb6c91a6ffbe8b80fae9eb55ee..9eaab6963346b788de4a886812f57db6063f511e 100644 --- a/source/libs/executor/src/timewindowoperator.c +++ b/source/libs/executor/src/timewindowoperator.c @@ -641,8 +641,7 @@ static void doInterpUnclosedTimeWindow(SOperatorInfo* pOperatorInfo, int32_t num setResultRowInterpo(pResult, RESULT_ROW_END_INTERP); setNotInterpoWindowKey(pSup->pCtx, numOfExprs, RESULT_ROW_START_INTERP); - doApplyFunctions(pTaskInfo, pSup->pCtx, &w, &pInfo->twAggSup.timeWindowData, startPos, 0, tsCols, pBlock->info.rows, - numOfExprs, pInfo->inputOrder); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, startPos, 0, pBlock->info.rows, numOfExprs); if (isResultRowInterpolated(pResult, RESULT_ROW_END_INTERP)) { closeResultRow(pr); @@ -871,7 +870,6 @@ static int32_t saveWinResult(int64_t ts, int32_t pageId, int32_t offset, uint64_ static int32_t saveWinResultRow(SResultRow* result, uint64_t groupId, SHashObj* pUpdatedMap) { return saveWinResult(result->win.skey, result->pageId, result->offset, groupId, pUpdatedMap); - ; } static int32_t saveResultRow(SResultRow* result, uint64_t groupId, SArray* pUpdated) { @@ -910,11 +908,11 @@ int32_t compareWinRes(void* pKey, void* data, int32_t index) { } static void removeDeleteResults(SHashObj* pUpdatedMap, SArray* pDelWins) { - if (!pUpdatedMap || taosHashGetSize(pUpdatedMap) == 0) { + int32_t delSize = taosArrayGetSize(pDelWins); + if (taosHashGetSize(pUpdatedMap) == 0 || delSize == 0) { return; } - int32_t delSize = taosArrayGetSize(pDelWins); - void* pIte = NULL; + void* pIte = NULL; while ((pIte = taosHashIterate(pUpdatedMap, pIte)) != NULL) { SResKeyPos* pResKey = (SResKeyPos*)pIte; int32_t index = binarySearchCom(pDelWins, delSize, pResKey, TSDB_ORDER_DESC, compareWinRes); @@ -987,8 +985,8 @@ static void hashIntervalAgg(SOperatorInfo* pOperatorInfo, SResultRowInfo* pResul if ((!pInfo->ignoreExpiredData || !isCloseWindow(&win, &pInfo->twAggSup)) && inSlidingWindow(&pInfo->interval, &win, &pBlock->info)) { updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &win, true); - doApplyFunctions(pTaskInfo, pSup->pCtx, &win, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, tsCols, - pBlock->info.rows, numOfOutput, pInfo->inputOrder); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, + pBlock->info.rows, numOfOutput); } doCloseWindow(pResultRowInfo, pInfo, pResult); @@ -1027,8 +1025,8 @@ static void hashIntervalAgg(SOperatorInfo* pOperatorInfo, SResultRowInfo* pResul doWindowBorderInterpolation(pInfo, pBlock, pResult, &nextWin, startPos, forwardRows, pSup); updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &nextWin, true); - doApplyFunctions(pTaskInfo, pSup->pCtx, &nextWin, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, tsCols, - pBlock->info.rows, numOfOutput, pInfo->inputOrder); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, + pBlock->info.rows, numOfOutput); doCloseWindow(pResultRowInfo, pInfo, pResult); } @@ -1191,8 +1189,8 @@ static void doStateWindowAggImpl(SOperatorInfo* pOperator, SStateWindowOperatorI } updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &window, false); - doApplyFunctions(pTaskInfo, pSup->pCtx, &window, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, - pRowSup->numOfRows, NULL, pBlock->info.rows, numOfOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, + pRowSup->numOfRows, pBlock->info.rows, numOfOutput); // here we start a new session window doKeepNewWindowStartInfo(pRowSup, tsList, j, gid); @@ -1216,8 +1214,8 @@ static void doStateWindowAggImpl(SOperatorInfo* pOperator, SStateWindowOperatorI } updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &pRowSup->win, false); - doApplyFunctions(pTaskInfo, pSup->pCtx, &pRowSup->win, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, - pRowSup->numOfRows, NULL, pBlock->info.rows, numOfOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, + pRowSup->numOfRows, pBlock->info.rows, numOfOutput); } static SSDataBlock* doStateWindowAgg(SOperatorInfo* pOperator) { @@ -1595,7 +1593,7 @@ static SSDataBlock* doStreamIntervalAgg(SOperatorInfo* pOperator) { SOperatorInfo* downstream = pOperator->pDownstream[0]; SArray* pUpdated = taosArrayInit(4, POINTER_BYTES); // SResKeyPos - _hash_fn_t hashFn = taosGetDefaultHashFunction(TSDB_DATA_TYPE_TIMESTAMP); + _hash_fn_t hashFn = taosGetDefaultHashFunction(TSDB_DATA_TYPE_BINARY); SHashObj* pUpdatedMap = taosHashInit(1024, hashFn, false, HASH_NO_LOCK); while (1) { SSDataBlock* pBlock = downstream->fpSet.getNextFn(downstream); @@ -1785,7 +1783,7 @@ void increaseTs(SqlFunctionCtx* pCtx) { } } -void initIntervalDownStream(SOperatorInfo* downstream, uint8_t type, SAggSupporter* pSup) { +void initIntervalDownStream(SOperatorInfo* downstream, uint16_t type, SAggSupporter* pSup) { if (downstream->operatorType != QUERY_NODE_PHYSICAL_PLAN_STREAM_SCAN) { // Todo(liuyao) support partition by column return; @@ -1886,62 +1884,6 @@ _error: return NULL; } -SOperatorInfo* createStreamIntervalOperatorInfo(SOperatorInfo* downstream, SExprInfo* pExprInfo, int32_t numOfCols, - SSDataBlock* pResBlock, SInterval* pInterval, int32_t primaryTsSlotId, - STimeWindowAggSupp* pTwAggSupp, SExecTaskInfo* pTaskInfo) { - SIntervalAggOperatorInfo* pInfo = taosMemoryCalloc(1, sizeof(SIntervalAggOperatorInfo)); - SOperatorInfo* pOperator = taosMemoryCalloc(1, sizeof(SOperatorInfo)); - if (pInfo == NULL || pOperator == NULL) { - goto _error; - } - - pOperator->pTaskInfo = pTaskInfo; - pInfo->inputOrder = TSDB_ORDER_ASC; - pInfo->interval = *pInterval; - pInfo->execModel = OPTR_EXEC_MODEL_STREAM; - pInfo->win = pTaskInfo->window; - pInfo->twAggSup = *pTwAggSupp; - pInfo->primaryTsIndex = primaryTsSlotId; - - int32_t numOfRows = 4096; - size_t keyBufSize = sizeof(int64_t) + sizeof(int64_t) + POINTER_BYTES; - - initResultSizeInfo(&pOperator->resultInfo, numOfRows); - int32_t code = initAggInfo(&pOperator->exprSupp, &pInfo->aggSup, pExprInfo, numOfCols, keyBufSize, pTaskInfo->id.str); - initBasicInfo(&pInfo->binfo, pResBlock); - initExecTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &pInfo->win); - - if (code != TSDB_CODE_SUCCESS) { - goto _error; - } - - initResultRowInfo(&pInfo->binfo.resultRowInfo); - - pOperator->name = "StreamTimeIntervalAggOperator"; - pOperator->operatorType = QUERY_NODE_PHYSICAL_PLAN_HASH_INTERVAL; - pOperator->blocking = true; - pOperator->status = OP_NOT_OPENED; - pOperator->exprSupp.pExprInfo = pExprInfo; - pOperator->exprSupp.numOfExprs = numOfCols; - pOperator->info = pInfo; - - pOperator->fpSet = createOperatorFpSet(doOpenIntervalAgg, doStreamIntervalAgg, doStreamIntervalAgg, NULL, - destroyIntervalOperatorInfo, aggEncodeResultRow, aggDecodeResultRow, NULL); - - code = appendDownstream(pOperator, &downstream, 1); - if (code != TSDB_CODE_SUCCESS) { - goto _error; - } - - return pOperator; - -_error: - destroyIntervalOperatorInfo(pInfo, numOfCols); - taosMemoryFreeClear(pOperator); - pTaskInfo->code = code; - return NULL; -} - // todo handle multiple timeline cases. assume no timeline interweaving static void doSessionWindowAggImpl(SOperatorInfo* pOperator, SSessionAggOperatorInfo* pInfo, SSDataBlock* pBlock) { SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo; @@ -1991,8 +1933,8 @@ static void doSessionWindowAggImpl(SOperatorInfo* pOperator, SSessionAggOperator // pInfo->numOfRows data belong to the current session window updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &window, false); - doApplyFunctions(pTaskInfo, pSup->pCtx, &window, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, - pRowSup->numOfRows, NULL, pBlock->info.rows, numOfOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, + pRowSup->numOfRows, pBlock->info.rows, numOfOutput); // here we start a new session window doKeepNewWindowStartInfo(pRowSup, tsList, j, gid); @@ -2009,8 +1951,8 @@ static void doSessionWindowAggImpl(SOperatorInfo* pOperator, SSessionAggOperator } updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &pRowSup->win, false); - doApplyFunctions(pTaskInfo, pSup->pCtx, &pRowSup->win, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, - pRowSup->numOfRows, NULL, pBlock->info.rows, numOfOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, pRowSup->startRowIndex, + pRowSup->numOfRows, pBlock->info.rows, numOfOutput); } static SSDataBlock* doSessionWindowAgg(SOperatorInfo* pOperator) { @@ -2413,11 +2355,11 @@ static SSDataBlock* doTimeslice(SOperatorInfo* pOperator) { break; } } + } - if (pSliceInfo->current > pSliceInfo->win.ekey) { - doSetOperatorCompleted(pOperator); - break; - } + if (pSliceInfo->current > pSliceInfo->win.ekey) { + doSetOperatorCompleted(pOperator); + break; } if (ts == pSliceInfo->current) { @@ -3009,8 +2951,8 @@ static void doHashInterval(SOperatorInfo* pOperatorInfo, SSDataBlock* pSDataBloc setResultBufPageDirty(pInfo->aggSup.pResultBuf, &pResultRowInfo->cur); } updateTimeWindowInfo(&pInfo->twAggSup.timeWindowData, &nextWin, true); - doApplyFunctions(pTaskInfo, pSup->pCtx, &nextWin, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, tsCols, - pSDataBlock->info.rows, numOfOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, &pInfo->twAggSup.timeWindowData, startPos, forwardRows, + pSDataBlock->info.rows, numOfOutput); int32_t prevEndPos = (forwardRows - 1) * step + startPos; ASSERT(pSDataBlock->info.window.skey > 0 && pSDataBlock->info.window.ekey > 0); startPos = getNextQualifiedWindow(&pInfo->interval, &nextWin, &pSDataBlock->info, tsCols, prevEndPos, pInfo->order); @@ -3109,11 +3051,12 @@ void processPullOver(SSDataBlock* pBlock, SHashObj* pMap) { static SSDataBlock* doStreamFinalIntervalAgg(SOperatorInfo* pOperator) { SStreamFinalIntervalOperatorInfo* pInfo = pOperator->info; - SOperatorInfo* downstream = pOperator->pDownstream[0]; - SArray* pUpdated = taosArrayInit(4, POINTER_BYTES); - _hash_fn_t hashFn = taosGetDefaultHashFunction(TSDB_DATA_TYPE_TIMESTAMP); - SHashObj* pUpdatedMap = taosHashInit(1024, hashFn, false, HASH_NO_LOCK); - TSKEY maxTs = INT64_MIN; + + SOperatorInfo* downstream = pOperator->pDownstream[0]; + SArray* pUpdated = taosArrayInit(4, POINTER_BYTES); + _hash_fn_t hashFn = taosGetDefaultHashFunction(TSDB_DATA_TYPE_BINARY); + SHashObj* pUpdatedMap = taosHashInit(1024, hashFn, false, HASH_NO_LOCK); + TSKEY maxTs = INT64_MIN; SExprSupp* pSup = &pOperator->exprSupp; @@ -3546,7 +3489,7 @@ void initDummyFunction(SqlFunctionCtx* pDummy, SqlFunctionCtx* pCtx, int32_t num } void initDownStream(SOperatorInfo* downstream, SStreamAggSupporter* pAggSup, int64_t gap, int64_t waterMark, - uint8_t type) { + uint16_t type) { ASSERT(downstream->operatorType == QUERY_NODE_PHYSICAL_PLAN_STREAM_SCAN); SStreamScanInfo* pScanInfo = downstream->info; pScanInfo->sessionSup = (SessionWindowSupporter){.pStreamAggSup = pAggSup, .gap = gap, .parentType = type}; @@ -3832,8 +3775,7 @@ static int32_t doOneWindowAggImpl(int32_t tsColId, SOptrBasicInfo* pBinfo, SStre return TSDB_CODE_QRY_OUT_OF_MEMORY; } updateTimeWindowInfo(pTimeWindowData, &pCurWin->win, false); - doApplyFunctions(pTaskInfo, pSup->pCtx, &pCurWin->win, pTimeWindowData, startIndex, winRows, tsCols, - pSDataBlock->info.rows, numOutput, TSDB_ORDER_ASC); + doApplyFunctions(pTaskInfo, pSup->pCtx, pTimeWindowData, startIndex, winRows, pSDataBlock->info.rows, numOutput); SFilePage* bufPage = getBufPage(pAggSup->pResultBuf, pCurWin->pos.pageId); setBufPageDirty(bufPage, true); releaseBufPage(pAggSup->pResultBuf, bufPage); @@ -4994,8 +4936,8 @@ static void doMergeAlignedIntervalAggImpl(SOperatorInfo* pOperatorInfo, SResultR } updateTimeWindowInfo(&iaInfo->twAggSup.timeWindowData, &currWin, true); - doApplyFunctions(pTaskInfo, pSup->pCtx, &currWin, &iaInfo->twAggSup.timeWindowData, startPos, currPos - startPos, - tsCols, pBlock->info.rows, numOfOutput, iaInfo->inputOrder); + doApplyFunctions(pTaskInfo, pSup->pCtx, &iaInfo->twAggSup.timeWindowData, startPos, currPos - startPos, + pBlock->info.rows, numOfOutput); outputMergeAlignedIntervalResult(pOperatorInfo, tableGroupId, pResultBlock, miaInfo->curTs); miaInfo->curTs = tsCols[currPos]; @@ -5016,8 +4958,8 @@ static void doMergeAlignedIntervalAggImpl(SOperatorInfo* pOperatorInfo, SResultR } updateTimeWindowInfo(&iaInfo->twAggSup.timeWindowData, &currWin, true); - doApplyFunctions(pTaskInfo, pSup->pCtx, &currWin, &iaInfo->twAggSup.timeWindowData, startPos, currPos - startPos, - tsCols, pBlock->info.rows, numOfOutput, iaInfo->inputOrder); + doApplyFunctions(pTaskInfo, pSup->pCtx, &iaInfo->twAggSup.timeWindowData, startPos, currPos - startPos, + pBlock->info.rows, numOfOutput); } static void doMergeAlignedIntervalAgg(SOperatorInfo* pOperator) { @@ -5309,8 +5251,8 @@ static void doMergeIntervalAggImpl(SOperatorInfo* pOperatorInfo, SResultRowInfo* } updateTimeWindowInfo(&iaInfo->twAggSup.timeWindowData, &win, true); - doApplyFunctions(pTaskInfo, pExprSup->pCtx, &win, &iaInfo->twAggSup.timeWindowData, startPos, forwardRows, tsCols, - pBlock->info.rows, numOfOutput, iaInfo->inputOrder); + doApplyFunctions(pTaskInfo, pExprSup->pCtx, &iaInfo->twAggSup.timeWindowData, startPos, forwardRows, + pBlock->info.rows, numOfOutput); doCloseWindow(pResultRowInfo, iaInfo, pResult); // output previous interval results after this interval (&win) is closed @@ -5341,8 +5283,8 @@ static void doMergeIntervalAggImpl(SOperatorInfo* pOperatorInfo, SResultRowInfo* doWindowBorderInterpolation(iaInfo, pBlock, pResult, &nextWin, startPos, forwardRows, pExprSup); updateTimeWindowInfo(&iaInfo->twAggSup.timeWindowData, &nextWin, true); - doApplyFunctions(pTaskInfo, pExprSup->pCtx, &nextWin, &iaInfo->twAggSup.timeWindowData, startPos, forwardRows, - tsCols, pBlock->info.rows, numOfOutput, iaInfo->inputOrder); + doApplyFunctions(pTaskInfo, pExprSup->pCtx, &iaInfo->twAggSup.timeWindowData, startPos, forwardRows, + pBlock->info.rows, numOfOutput); doCloseWindow(pResultRowInfo, iaInfo, pResult); // output previous interval results after this interval (&nextWin) is closed diff --git a/source/libs/function/src/builtins.c b/source/libs/function/src/builtins.c index c19b459cdb9e2a435a73f1bebb1f6de5e9b07fb3..ed82e4cb50cd2ce72ab3e9965b7ef1481fe2ccfa 100644 --- a/source/libs/function/src/builtins.c +++ b/source/libs/function/src/builtins.c @@ -1503,11 +1503,17 @@ static int32_t translateInterp(SFunctionNode* pFunc, char* pErrBuf, int32_t len) return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName); } + uint8_t nodeType = nodeType(nodesListGetNode(pFunc->pParameterList, 0)); + uint8_t paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, 0))->resType.type; + if (!IS_NUMERIC_TYPE(paraType) || QUERY_NODE_VALUE == nodeType) { + return invaildFuncParaTypeErrMsg(pErrBuf, len, pFunc->functionName); + } + if (3 <= numOfParams) { int64_t timeVal[2] = {0}; for (int32_t i = 1; i < 3; ++i) { - uint8_t nodeType = nodeType(nodesListGetNode(pFunc->pParameterList, i)); - uint8_t paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, i))->resType.type; + nodeType = nodeType(nodesListGetNode(pFunc->pParameterList, i)); + paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, i))->resType.type; if (!IS_VAR_DATA_TYPE(paraType) || QUERY_NODE_VALUE != nodeType) { return invaildFuncParaTypeErrMsg(pErrBuf, len, pFunc->functionName); } @@ -1525,8 +1531,8 @@ static int32_t translateInterp(SFunctionNode* pFunc, char* pErrBuf, int32_t len) } if (4 == numOfParams) { - uint8_t nodeType = nodeType(nodesListGetNode(pFunc->pParameterList, 3)); - uint8_t paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, 3))->resType.type; + nodeType = nodeType(nodesListGetNode(pFunc->pParameterList, 3)); + paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, 3))->resType.type; if (!IS_INTEGER_TYPE(paraType) || QUERY_NODE_VALUE != nodeType) { return invaildFuncParaTypeErrMsg(pErrBuf, len, pFunc->functionName); } diff --git a/source/libs/function/src/builtinsimpl.c b/source/libs/function/src/builtinsimpl.c index bf4a07f8e2c110c88ec3fe79b2fba557123fa7ab..5051dcd65c9b494f3d4ffff0f5a930dd5acee3e8 100644 --- a/source/libs/function/src/builtinsimpl.c +++ b/source/libs/function/src/builtinsimpl.c @@ -5574,6 +5574,7 @@ int32_t twaFunction(SqlFunctionCtx* pCtx) { if (pCtx->end.key != INT64_MIN) { pInfo->dOutput += twa_get_area(pInfo->p, pCtx->end); pInfo->p = pCtx->end; + numOfElems += 1; } pInfo->win.ekey = pInfo->p.key; diff --git a/source/libs/nodes/src/nodesCodeFuncs.c b/source/libs/nodes/src/nodesCodeFuncs.c index a6546f3299e5bb5e4a99f5fe829492f42bc3b4b5..0f32001c476c2e98465a0ce3f07fb705c3f53244 100644 --- a/source/libs/nodes/src/nodesCodeFuncs.c +++ b/source/libs/nodes/src/nodesCodeFuncs.c @@ -4673,7 +4673,6 @@ static int32_t jsonToNode(const SJson* pJson, void* pObj) { int32_t code; tjsonGetNumberValue(pJson, jkNodeType, pNode->type, code); - ; if (TSDB_CODE_SUCCESS == code) { code = tjsonToObject(pJson, nodesNodeName(pNode->type), jsonToSpecificNode, pNode); if (TSDB_CODE_SUCCESS != code) { diff --git a/source/libs/nodes/src/nodesToSQLFuncs.c b/source/libs/nodes/src/nodesToSQLFuncs.c index 3b129740e81b4743dbae25dd82da36ac2a49407e..e521c57c3d80eac9455ab9affa5a4b053983ef84 100644 --- a/source/libs/nodes/src/nodesToSQLFuncs.c +++ b/source/libs/nodes/src/nodesToSQLFuncs.c @@ -21,36 +21,89 @@ #include "taoserror.h" #include "thash.h" -char *gOperatorStr[] = {NULL, - "+", - "-", - "*", - "/", - "%", - "-", - "&", - "|", - ">", - ">=", - "<", - "<=", - "=", - "<>", - "IN", - "NOT IN", - "LIKE", - "NOT LIKE", - "MATCH", - "NMATCH", - "IS NULL", - "IS NOT NULL", - "IS TRUE", - "IS FALSE", - "IS UNKNOWN", - "IS NOT TRUE", - "IS NOT FALSE", - "IS NOT UNKNOWN"}; -char *gLogicConditionStr[] = {"AND", "OR", "NOT"}; +const char *operatorTypeStr(EOperatorType type) { + switch (type) { + case OP_TYPE_ADD: + return "+"; + case OP_TYPE_SUB: + return "-"; + case OP_TYPE_MULTI: + return "*"; + case OP_TYPE_DIV: + return "/"; + case OP_TYPE_REM: + return "%"; + case OP_TYPE_MINUS: + return "-"; + case OP_TYPE_BIT_AND: + return "&"; + case OP_TYPE_BIT_OR: + return "|"; + case OP_TYPE_GREATER_THAN: + return ">"; + case OP_TYPE_GREATER_EQUAL: + return ">="; + case OP_TYPE_LOWER_THAN: + return "<"; + case OP_TYPE_LOWER_EQUAL: + return "<="; + case OP_TYPE_EQUAL: + return "="; + case OP_TYPE_NOT_EQUAL: + return "<>"; + case OP_TYPE_IN: + return "IN"; + case OP_TYPE_NOT_IN: + return "NOT IN"; + case OP_TYPE_LIKE: + return "LIKE"; + case OP_TYPE_NOT_LIKE: + return "NOT LIKE"; + case OP_TYPE_MATCH: + return "MATCH"; + case OP_TYPE_NMATCH: + return "NMATCH"; + case OP_TYPE_IS_NULL: + return "IS NULL"; + case OP_TYPE_IS_NOT_NULL: + return "IS NOT NULL"; + case OP_TYPE_IS_TRUE: + return "IS TRUE"; + case OP_TYPE_IS_FALSE: + return "IS FALSE"; + case OP_TYPE_IS_UNKNOWN: + return "IS UNKNOWN"; + case OP_TYPE_IS_NOT_TRUE: + return "IS NOT TRUE"; + case OP_TYPE_IS_NOT_FALSE: + return "IS NOT FALSE"; + case OP_TYPE_IS_NOT_UNKNOWN: + return "IS NOT UNKNOWN"; + case OP_TYPE_JSON_GET_VALUE: + return "=>"; + case OP_TYPE_JSON_CONTAINS: + return "CONTAINS"; + case OP_TYPE_ASSIGN: + return "="; + default: + break; + } + return "UNKNOWN"; +} + +const char *logicConditionTypeStr(ELogicConditionType type) { + switch (type) { + case LOGIC_COND_TYPE_AND: + return "AND"; + case LOGIC_COND_TYPE_OR: + return "OR"; + case LOGIC_COND_TYPE_NOT: + return "NOT"; + default: + break; + } + return "UNKNOWN"; +} int32_t nodesNodeToSQL(SNode *pNode, char *buf, int32_t bufSize, int32_t *len) { switch (pNode->type) { @@ -94,12 +147,7 @@ int32_t nodesNodeToSQL(SNode *pNode, char *buf, int32_t bufSize, int32_t *len) { NODES_ERR_RET(nodesNodeToSQL(pOpNode->pLeft, buf, bufSize, len)); } - if (pOpNode->opType >= (sizeof(gOperatorStr) / sizeof(gOperatorStr[0]))) { - nodesError("unknown operation type:%d", pOpNode->opType); - NODES_ERR_RET(TSDB_CODE_QRY_APP_ERROR); - } - - *len += snprintf(buf + *len, bufSize - *len, " %s ", gOperatorStr[pOpNode->opType]); + *len += snprintf(buf + *len, bufSize - *len, " %s ", operatorTypeStr(pOpNode->opType)); if (pOpNode->pRight) { NODES_ERR_RET(nodesNodeToSQL(pOpNode->pRight, buf, bufSize, len)); @@ -118,7 +166,7 @@ int32_t nodesNodeToSQL(SNode *pNode, char *buf, int32_t bufSize, int32_t *len) { FOREACH(node, pLogicNode->pParameterList) { if (!first) { - *len += snprintf(buf + *len, bufSize - *len, " %s ", gLogicConditionStr[pLogicNode->condType]); + *len += snprintf(buf + *len, bufSize - *len, " %s ", logicConditionTypeStr(pLogicNode->condType)); } NODES_ERR_RET(nodesNodeToSQL(node, buf, bufSize, len)); first = false; diff --git a/source/libs/parser/src/parInsert.c b/source/libs/parser/src/parInsert.c index 31ae35e7177d55397a6000a86712b977962942dd..0922cdb6b9cdee6e5d2d1f6f315d84a8bafacc70 100644 --- a/source/libs/parser/src/parInsert.c +++ b/source/libs/parser/src/parInsert.c @@ -143,9 +143,9 @@ static int32_t createSName(SName* pName, SToken* pTableName, int32_t acctId, con } char name[TSDB_DB_FNAME_LEN] = {0}; strncpy(name, pTableName->z, dbLen); - dbLen = strdequote(name); + int32_t actualDbLen = strdequote(name); - code = tNameSetDbName(pName, acctId, name, dbLen); + code = tNameSetDbName(pName, acctId, name, actualDbLen); if (code != TSDB_CODE_SUCCESS) { return buildInvalidOperationMsg(pMsgBuf, msg1); } diff --git a/source/libs/parser/src/parTranslater.c b/source/libs/parser/src/parTranslater.c index 1c7446ad6f2bad5bdb00aca710708e57bca07695..3c0d9a5f63bbf008469618589c206bc9eb9f447c 100644 --- a/source/libs/parser/src/parTranslater.c +++ b/source/libs/parser/src/parTranslater.c @@ -2822,6 +2822,29 @@ static int32_t createDefaultFillNode(STranslateContext* pCxt, SNode** pOutput) { return TSDB_CODE_SUCCESS; } +static int32_t checkEvery(STranslateContext* pCxt, SValueNode* pInterval) { + int32_t len = strlen(pInterval->literal); + + char *unit = &pInterval->literal[len - 1]; + if (*unit == 'n' || *unit == 'y') { + return generateSyntaxErrMsgExt(&pCxt->msgBuf, TSDB_CODE_PAR_WRONG_VALUE_TYPE, + "Unsupported time unit in EVERY clause"); + } + + return TSDB_CODE_SUCCESS; +} + +static int32_t translateInterpEvery(STranslateContext* pCxt, SNode** pEvery) { + int32_t code = TSDB_CODE_SUCCESS; + + code = checkEvery(pCxt, (SValueNode *)(*pEvery)); + if (TSDB_CODE_SUCCESS == code) { + code = translateExpr(pCxt, pEvery); + } + + return code; +} + static int32_t translateInterpFill(STranslateContext* pCxt, SSelectStmt* pSelect) { int32_t code = TSDB_CODE_SUCCESS; @@ -2856,7 +2879,7 @@ static int32_t translateInterp(STranslateContext* pCxt, SSelectStmt* pSelect) { int32_t code = translateExpr(pCxt, &pSelect->pRange); if (TSDB_CODE_SUCCESS == code) { - code = translateExpr(pCxt, &pSelect->pEvery); + code = translateInterpEvery(pCxt, &pSelect->pEvery); } if (TSDB_CODE_SUCCESS == code) { code = translateInterpFill(pCxt, pSelect); diff --git a/source/libs/scalar/inc/filterInt.h b/source/libs/scalar/inc/filterInt.h index 54e873065b20c6317d96c6a56240c7920371a663..23693c785aa17921e5ba4420fe6477fa72b27392 100644 --- a/source/libs/scalar/inc/filterInt.h +++ b/source/libs/scalar/inc/filterInt.h @@ -350,7 +350,6 @@ struct SFilterInfo { extern bool filterDoCompare(__compar_fn_t func, uint8_t optr, void *left, void *right); extern __compar_fn_t filterGetCompFunc(int32_t type, int32_t optr); -extern OptrStr gOptrStr[]; #ifdef __cplusplus } diff --git a/source/libs/scalar/src/filter.c b/source/libs/scalar/src/filter.c index 1664a4d61283e0e393862de666bb90990355227a..4377dbf14ec55dae53d41859af8480886f4cce51 100644 --- a/source/libs/scalar/src/filter.c +++ b/source/libs/scalar/src/filter.c @@ -24,46 +24,6 @@ #include "ttime.h" #include "functionMgt.h" -OptrStr gOptrStr[] = { - {0, "invalid"}, - {OP_TYPE_ADD, "+"}, - {OP_TYPE_SUB, "-"}, - {OP_TYPE_MULTI, "*"}, - {OP_TYPE_DIV, "/"}, - {OP_TYPE_REM, "%"}, - {OP_TYPE_MINUS, "minus"}, - {OP_TYPE_ASSIGN, "assign"}, - // bit operator - {OP_TYPE_BIT_AND, "&"}, - {OP_TYPE_BIT_OR, "|"}, - - // comparison operator - {OP_TYPE_GREATER_THAN, ">"}, - {OP_TYPE_GREATER_EQUAL, ">="}, - {OP_TYPE_LOWER_THAN, "<"}, - {OP_TYPE_LOWER_EQUAL, "<="}, - {OP_TYPE_EQUAL, "=="}, - {OP_TYPE_NOT_EQUAL, "!="}, - {OP_TYPE_IN, "in"}, - {OP_TYPE_NOT_IN, "not in"}, - {OP_TYPE_LIKE, "like"}, - {OP_TYPE_NOT_LIKE, "not like"}, - {OP_TYPE_MATCH, "match"}, - {OP_TYPE_NMATCH, "nmatch"}, - {OP_TYPE_IS_NULL, "is null"}, - {OP_TYPE_IS_NOT_NULL, "not null"}, - {OP_TYPE_IS_TRUE, "is true"}, - {OP_TYPE_IS_FALSE, "is false"}, - {OP_TYPE_IS_UNKNOWN, "is unknown"}, - {OP_TYPE_IS_NOT_TRUE, "not true"}, - {OP_TYPE_IS_NOT_FALSE, "not false"}, - {OP_TYPE_IS_NOT_UNKNOWN, "not unknown"}, - - // json operator - {OP_TYPE_JSON_GET_VALUE, "->"}, - {OP_TYPE_JSON_CONTAINS, "json contains"} -}; - bool filterRangeCompGi (const void *minv, const void *maxv, const void *minr, const void *maxr, __compar_fn_t cfunc) { int32_t result = cfunc(maxv, minr); return result >= 0; @@ -986,7 +946,7 @@ int32_t filterAddUnit(SFilterInfo *info, uint8_t optr, SFilterFieldId *left, SFi } else { int32_t paramNum = scalarGetOperatorParamNum(optr); if (1 != paramNum) { - fltError("invalid right field in unit, operator:%s, rightType:%d", gOptrStr[optr].str, u->right.type); + fltError("invalid right field in unit, operator:%s, rightType:%d", operatorTypeStr(optr), u->right.type); return TSDB_CODE_QRY_APP_ERROR; } } @@ -1517,7 +1477,7 @@ void filterDumpInfoToString(SFilterInfo *info, const char *msg, int32_t options) SFilterField *left = FILTER_UNIT_LEFT_FIELD(info, unit); SColumnNode *refNode = (SColumnNode *)left->desc; if (unit->compare.optr >= 0 && unit->compare.optr <= OP_TYPE_JSON_CONTAINS){ - len = sprintf(str, "UNIT[%d] => [%d][%d] %s [", i, refNode->dataBlockId, refNode->slotId, gOptrStr[unit->compare.optr].str); + len = sprintf(str, "UNIT[%d] => [%d][%d] %s [", i, refNode->dataBlockId, refNode->slotId, operatorTypeStr(unit->compare.optr)); } if (unit->right.type == FLD_TYPE_VALUE && FILTER_UNIT_OPTR(unit) != OP_TYPE_IN) { @@ -1536,7 +1496,7 @@ void filterDumpInfoToString(SFilterInfo *info, const char *msg, int32_t options) if (unit->compare.optr2) { strcat(str, " && "); if (unit->compare.optr2 >= 0 && unit->compare.optr2 <= OP_TYPE_JSON_CONTAINS){ - sprintf(str + strlen(str), "[%d][%d] %s [", refNode->dataBlockId, refNode->slotId, gOptrStr[unit->compare.optr2].str); + sprintf(str + strlen(str), "[%d][%d] %s [", refNode->dataBlockId, refNode->slotId, operatorTypeStr(unit->compare.optr2)); } if (unit->right2.type == FLD_TYPE_VALUE && FILTER_UNIT_OPTR(unit) != OP_TYPE_IN) { diff --git a/source/libs/scalar/test/scalar/scalarTests.cpp b/source/libs/scalar/test/scalar/scalarTests.cpp index 9b40f0a4659ee9a8a85131c9daf017b8c3b17ce2..7229fdec386b7df5bb626ea0524144d4f42dd995 100644 --- a/source/libs/scalar/test/scalar/scalarTests.cpp +++ b/source/libs/scalar/test/scalar/scalarTests.cpp @@ -1089,16 +1089,16 @@ void makeCalculate(void *json, void *key, int32_t rightType, void *rightData, do }else if(opType == OP_TYPE_ADD || opType == OP_TYPE_SUB || opType == OP_TYPE_MULTI || opType == OP_TYPE_DIV || opType == OP_TYPE_REM || opType == OP_TYPE_MINUS){ - printf("op:%s,1result:%f,except:%f\n", gOptrStr[opType].str, *((double *)colDataGetData(column, 0)), exceptValue); + printf("op:%s,1result:%f,except:%f\n", operatorTypeStr(opType), *((double *)colDataGetData(column, 0)), exceptValue); ASSERT_TRUE(fabs(*((double *)colDataGetData(column, 0)) - exceptValue) < 0.0001); }else if(opType == OP_TYPE_BIT_AND || opType == OP_TYPE_BIT_OR){ - printf("op:%s,2result:%" PRId64 ",except:%f\n", gOptrStr[opType].str, *((int64_t *)colDataGetData(column, 0)), exceptValue); + printf("op:%s,2result:%" PRId64 ",except:%f\n", operatorTypeStr(opType), *((int64_t *)colDataGetData(column, 0)), exceptValue); ASSERT_EQ(*((int64_t *)colDataGetData(column, 0)), exceptValue); }else if(opType == OP_TYPE_GREATER_THAN || opType == OP_TYPE_GREATER_EQUAL || opType == OP_TYPE_LOWER_THAN || opType == OP_TYPE_LOWER_EQUAL || opType == OP_TYPE_EQUAL || opType == OP_TYPE_NOT_EQUAL || opType == OP_TYPE_IS_NULL || opType == OP_TYPE_IS_NOT_NULL || opType == OP_TYPE_IS_TRUE || opType == OP_TYPE_LIKE || opType == OP_TYPE_NOT_LIKE || opType == OP_TYPE_MATCH || opType == OP_TYPE_NMATCH){ - printf("op:%s,3result:%d,except:%f\n", gOptrStr[opType].str, *((bool *)colDataGetData(column, 0)), exceptValue); + printf("op:%s,3result:%d,except:%f\n", operatorTypeStr(opType), *((bool *)colDataGetData(column, 0)), exceptValue); ASSERT_EQ(*((bool *)colDataGetData(column, 0)), exceptValue); } diff --git a/source/libs/scheduler/src/schRemote.c b/source/libs/scheduler/src/schRemote.c index 8c9003a9b2af371ebdb50620f53d382c04609a03..ecd9daf1bcd3b83803754017aec27c1ebe62becf 100644 --- a/source/libs/scheduler/src/schRemote.c +++ b/source/libs/scheduler/src/schRemote.c @@ -20,7 +20,7 @@ #include "tmsg.h" #include "tref.h" #include "trpc.h" - +// clang-format off int32_t schValidateRspMsgType(SSchJob *pJob, SSchTask *pTask, int32_t msgType) { int32_t lastMsgType = pTask->lastMsgType; int32_t taskStatus = SCH_GET_TASK_STATUS(pTask); @@ -402,7 +402,7 @@ int32_t schHandleDropCallback(void *param, SDataBuf *pMsg, int32_t code) { qDebug("QID:0x%" PRIx64 ",TID:0x%" PRIx64 " drop task rsp received, code:0x%x", pParam->queryId, pParam->taskId, code); if (pMsg) { - taosMemoryFree(pMsg->pData); + taosMemoryFree(pMsg->pData); } return TSDB_CODE_SUCCESS; } @@ -415,7 +415,7 @@ int32_t schHandleLinkBrokenCallback(void *param, SDataBuf *pMsg, int32_t code) { if (head->isHbParam) { taosMemoryFree(pMsg->pData); - + SSchHbCallbackParam *hbParam = (SSchHbCallbackParam *)param; SSchTrans trans = {.pTrans = hbParam->pTrans, .pHandle = NULL}; SCH_ERR_RET(schUpdateHbConnection(&hbParam->nodeEpId, &trans)); @@ -1104,7 +1104,7 @@ int32_t schBuildAndSendMsg(SSchJob *pJob, SSchTask *pTask, SQueryNodeAddr *addr, #if 1 SSchTrans trans = {.pTrans = pJob->conn.pTrans, .pHandle = SCH_GET_TASK_HANDLE(pTask)}; - schAsyncSendMsg(pJob, pTask, &trans, addr, msgType, msg, msgSize, persistHandle, (rpcCtx.args ? &rpcCtx : NULL)); + code = schAsyncSendMsg(pJob, pTask, &trans, addr, msgType, msg, msgSize, persistHandle, (rpcCtx.args ? &rpcCtx : NULL)); msg = NULL; SCH_ERR_JRET(code); @@ -1114,7 +1114,7 @@ int32_t schBuildAndSendMsg(SSchJob *pJob, SSchTask *pTask, SQueryNodeAddr *addr, #else if (TDMT_VND_SUBMIT != msgType) { SSchTrans trans = {.pTrans = pJob->conn.pTrans, .pHandle = SCH_GET_TASK_HANDLE(pTask)}; - schAsyncSendMsg(pJob, pTask, &trans, addr, msgType, msg, msgSize, persistHandle, (rpcCtx.args ? &rpcCtx : NULL)); + code = schAsyncSendMsg(pJob, pTask, &trans, addr, msgType, msg, msgSize, persistHandle, (rpcCtx.args ? &rpcCtx : NULL)); msg = NULL; SCH_ERR_JRET(code); @@ -1136,3 +1136,4 @@ _return: taosMemoryFreeClear(msg); SCH_RET(code); } +// clang-format on diff --git a/source/libs/tdb/src/db/tdbPCache.c b/source/libs/tdb/src/db/tdbPCache.c index 22229ea0e8407cf57a7e8aa3ed40ca47cf255c12..ab9b21dc3fc01158ef5c504d69530b30eab6d79a 100644 --- a/source/libs/tdb/src/db/tdbPCache.c +++ b/source/libs/tdb/src/db/tdbPCache.c @@ -199,10 +199,20 @@ static SPage *tdbPCacheFetchImpl(SPCache *pCache, const SPgid *pPgid, TXN *pTxn) if (pPageH) { // copy the page content memcpy(&(pPage->pgid), pPgid, sizeof(*pPgid)); + + for (int nLoops = 0;;) { + if (pPageH->pPager) break; + if (++nLoops > 1000) { + sched_yield(); + nLoops = 0; + } + } + pPage->pLruNext = NULL; pPage->pPager = pPageH->pPager; memcpy(pPage->pData, pPageH->pData, pPage->pageSize); + tdbDebug("pcache/pPageH: %p %d %p %p", pPageH, pPageH->pPageHdr - pPageH->pData, pPageH->xCellSize, pPage); tdbPageInit(pPage, pPageH->pPageHdr - pPageH->pData, pPageH->xCellSize); pPage->kLen = pPageH->kLen; pPage->vLen = pPageH->vLen; diff --git a/source/libs/transport/src/thttp.c b/source/libs/transport/src/thttp.c index 4e49e9ca13a7924e43dc106ec9b495b4d8fa4015..7cfb188ac9c688a073cea236d8019eb4ab17469c 100644 --- a/source/libs/transport/src/thttp.c +++ b/source/libs/transport/src/thttp.c @@ -21,11 +21,14 @@ #include "taoserror.h" #include "tlog.h" + +#define HTTP_RECV_BUF_SIZE 1024 typedef struct SHttpClient { uv_connect_t conn; uv_tcp_t tcp; uv_write_t req; - uv_buf_t* buf; + uv_buf_t* wbuf; + char *rbuf; char* addr; uint16_t port; } SHttpClient; @@ -122,14 +125,30 @@ _OVER: } static void destroyHttpClient(SHttpClient* cli) { - taosMemoryFree(cli->buf); + taosMemoryFree(cli->wbuf); + taosMemoryFree(cli->rbuf); taosMemoryFree(cli->addr); taosMemoryFree(cli); + } static void clientCloseCb(uv_handle_t* handle) { SHttpClient* cli = handle->data; destroyHttpClient(cli); } +static void clientAllocBuffCb(uv_handle_t *handle, size_t suggested_size, uv_buf_t *buf) { + SHttpClient* cli = handle->data; + buf->base = cli->rbuf; + buf->len = HTTP_RECV_BUF_SIZE; +} +static void clientRecvCb(uv_stream_t* handle, ssize_t nread, const uv_buf_t *buf) { + SHttpClient* cli = handle->data; + if (nread < 0) { + uError("http-report read error:%s", uv_err_name(nread)); + } else { + uInfo("http-report succ to read %d bytes, just ignore it", nread); + } + uv_close((uv_handle_t*)&cli->tcp, clientCloseCb); +} static void clientSentCb(uv_write_t* req, int32_t status) { SHttpClient* cli = req->data; if (status != 0) { @@ -138,7 +157,7 @@ static void clientSentCb(uv_write_t* req, int32_t status) { } else { uInfo("http-report succ to send data"); } - uv_close((uv_handle_t*)&cli->tcp, clientCloseCb); + uv_read_start((uv_stream_t *)&cli->tcp, clientAllocBuffCb, clientRecvCb); } static void clientConnCb(uv_connect_t* req, int32_t status) { SHttpClient* cli = req->data; @@ -148,7 +167,7 @@ static void clientConnCb(uv_connect_t* req, int32_t status) { uv_close((uv_handle_t*)&cli->tcp, clientCloseCb); return; } - uv_write(&cli->req, (uv_stream_t*)&cli->tcp, cli->buf, 2, clientSentCb); + uv_write(&cli->req, (uv_stream_t*)&cli->tcp, cli->wbuf, 2, clientSentCb); } static int32_t taosBuildDstAddr(const char* server, uint16_t port, struct sockaddr_in* dest) { @@ -189,7 +208,8 @@ int32_t taosSendHttpReport(const char* server, uint16_t port, char* pCont, int32 cli->conn.data = cli; cli->tcp.data = cli; cli->req.data = cli; - cli->buf = wb; + cli->wbuf = wb; + cli->rbuf = taosMemoryCalloc(1, HTTP_RECV_BUF_SIZE); cli->addr = tstrdup(server); cli->port = port; @@ -199,7 +219,6 @@ int32_t taosSendHttpReport(const char* server, uint16_t port, char* pCont, int32 int32_t fd = taosCreateSocketWithTimeout(5); uv_tcp_open((uv_tcp_t*)&cli->tcp, fd); - int32_t ret = uv_tcp_connect(&cli->conn, &cli->tcp, (const struct sockaddr*)&dest, clientConnCb); if (ret != 0) { uError("http-report failed to connect to server, reason:%s, dst:%s:%d", uv_strerror(ret), cli->addr, cli->port); diff --git a/source/libs/transport/src/transCli.c b/source/libs/transport/src/transCli.c index 763483cbf6245448afa6d70e7952862ce88cbfee..7052b0b915137678d6aff528a26540a973cd74f5 100644 --- a/source/libs/transport/src/transCli.c +++ b/source/libs/transport/src/transCli.c @@ -1432,7 +1432,7 @@ int transSendRequest(void* shandle, const SEpSet* pEpSet, STransMsg* pReq, STran if (pThrd == NULL && valid == false) { transFreeMsg(pReq->pCont); transReleaseExHandle(transGetInstMgt(), (int64_t)shandle); - return -1; + return TSDB_CODE_RPC_BROKEN_LINK; } TRACE_SET_MSGID(&pReq->info.traceId, tGenIdPI64()); @@ -1477,7 +1477,7 @@ int transSendRecv(void* shandle, const SEpSet* pEpSet, STransMsg* pReq, STransMs if (pThrd == NULL && valid == false) { transFreeMsg(pReq->pCont); transReleaseExHandle(transGetInstMgt(), (int64_t)shandle); - return -1; + return TSDB_CODE_RPC_BROKEN_LINK; } tsem_t* sem = taosMemoryCalloc(1, sizeof(tsem_t)); diff --git a/source/libs/transport/src/transSvr.c b/source/libs/transport/src/transSvr.c index db05aefe7b1930b286e3a2cc8aa92a251077939d..447db7613656613255369230138979a7596754a9 100644 --- a/source/libs/transport/src/transSvr.c +++ b/source/libs/transport/src/transSvr.c @@ -275,16 +275,15 @@ void uvOnRecvCb(uv_stream_t* cli, ssize_t nread, const uv_buf_t* buf) { if (pBuf->len <= TRANS_PACKET_LIMIT) { while (transReadComplete(pBuf)) { tTrace("%s conn %p alread read complete packet", transLabel(pTransInst), conn); - if (pBuf->invalid) { - tTrace("%s conn %p alread read invalid packet", transLabel(pTransInst), conn); + if (true == pBuf->invalid || false == uvHandleReq(conn)) { + tError("%s conn %p read invalid packet", transLabel(pTransInst), conn); destroyConn(conn, true); return; - } else { - if (false == uvHandleReq(conn)) break; } } return; } else { + tError("%s conn %p read invalid packet, exceed limit", transLabel(pTransInst), conn); destroyConn(conn, true); return; } diff --git a/source/util/src/tconfig.c b/source/util/src/tconfig.c index fdb397561dbbddba2015a2a7e149ec0a71b6be2d..2a28ec66d2f58225b7c4dde5766d3a1522ca1a3d 100644 --- a/source/util/src/tconfig.c +++ b/source/util/src/tconfig.c @@ -33,6 +33,8 @@ int32_t cfgLoadFromEnvCmd(SConfig *pConfig, const char **envCmd); int32_t cfgLoadFromApollUrl(SConfig *pConfig, const char *url); int32_t cfgSetItem(SConfig *pConfig, const char *name, const char *value, ECfgSrcType stype); +extern char **environ; + SConfig *cfgInit() { SConfig *pCfg = taosMemoryCalloc(1, sizeof(SConfig)); if (pCfg == NULL) { @@ -627,24 +629,17 @@ void cfgDumpCfg(SConfig *pCfg, bool tsc, bool dump) { } int32_t cfgLoadFromEnvVar(SConfig *pConfig) { - char *line = NULL, *name, *value, *value2, *value3; + char line[1024], *name, *value, *value2, *value3; int32_t olen, vlen, vlen2, vlen3; int32_t code = 0; - ssize_t _bytes = 0; - TdCmdPtr pCmd = taosOpenCmd("set"); - if (pCmd == NULL) { - terrno = TAOS_SYSTEM_ERROR(errno); - return -1; - } - while (!taosEOFCmd(pCmd)) { + char **pEnv = environ; + line[1023] = 0; + while(*pEnv != NULL) { name = value = value2 = value3 = NULL; olen = vlen = vlen2 = vlen3 = 0; - _bytes = taosGetLineCmd(pCmd, &line); - if (_bytes < 0) { - break; - } - if(line[_bytes - 1] == '\n') line[_bytes - 1] = 0; + strncpy(line, *pEnv, sizeof(line)-1); + pEnv++; taosEnvToCfg(line, line); paGetToken(line, &name, &olen); @@ -671,9 +666,6 @@ int32_t cfgLoadFromEnvVar(SConfig *pConfig) { } } - taosCloseCmd(&pCmd); - if (line != NULL) taosMemoryFreeClear(line); - uInfo("load from env variables cfg success"); return 0; } @@ -1040,34 +1032,25 @@ int32_t cfgGetApollUrl(const char **envCmd, const char *envFile, char* apolloUrl index++; } - char *line = NULL; - ssize_t _bytes = 0; - TdCmdPtr pCmd = taosOpenCmd("set"); - if (pCmd != NULL) { - while (!taosEOFCmd(pCmd)) { - _bytes = taosGetLineCmd(pCmd, &line); - if (_bytes < 0) { - break; - } - if(line[_bytes - 1] == '\n') line[_bytes - 1] = 0; - if (strncmp(line, "TAOS_APOLLO_URL", 14) == 0) { - char *p = strchr(line, '='); - if (p != NULL) { + char line[1024]; + char **pEnv = environ; + line[1023] = 0; + while(*pEnv != NULL) { + strncpy(line, *pEnv, sizeof(line)-1); + pEnv++; + if (strncmp(line, "TAOS_APOLLO_URL", 14) == 0) { + char *p = strchr(line, '='); + if (p != NULL) { + p++; + if (*p == '\'') { p++; - if (*p == '\'') { - p++; - p[strlen(p)-1] = '\0'; - } - memcpy(apolloUrl, p, TMIN(strlen(p)+1,PATH_MAX)); - uInfo("get apollo url from env variables success, apolloUrl=%s",apolloUrl); - taosCloseCmd(&pCmd); - if (line != NULL) taosMemoryFreeClear(line); - return 0; + p[strlen(p)-1] = '\0'; } + memcpy(apolloUrl, p, TMIN(strlen(p)+1,PATH_MAX)); + uInfo("get apollo url from env variables success, apolloUrl=%s",apolloUrl); + return 0; } } - taosCloseCmd(&pCmd); - if (line != NULL) taosMemoryFreeClear(line); } const char *filepath = ".env"; @@ -1083,10 +1066,11 @@ int32_t cfgGetApollUrl(const char **envCmd, const char *envFile, char* apolloUrl return 0; } } + int64_t _bytes; TdFilePtr pFile = taosOpenFile(filepath, TD_FILE_READ | TD_FILE_STREAM); if (pFile != NULL) { while (!taosEOFFile(pFile)) { - _bytes = taosGetLineFile(pFile, &line); + _bytes = taosGetsFile(pFile, sizeof(line) - 1, line); if (_bytes <= 0) { break; } @@ -1101,14 +1085,12 @@ int32_t cfgGetApollUrl(const char **envCmd, const char *envFile, char* apolloUrl } memcpy(apolloUrl, p, TMIN(strlen(p)+1,PATH_MAX)); taosCloseFile(&pFile); - if (line != NULL) taosMemoryFreeClear(line); uInfo("get apollo url from env file success"); return 0; } } } taosCloseFile(&pFile); - if (line != NULL) taosMemoryFreeClear(line); } uInfo("fail get apollo url from cmd env file"); diff --git a/source/util/src/tqueue.c b/source/util/src/tqueue.c index 50beba8a9b67e191d402e580a2964f954c454869..eb70002680cd8df2849ffa4fcb6c7c27ddf330d4 100644 --- a/source/util/src/tqueue.c +++ b/source/util/src/tqueue.c @@ -298,7 +298,8 @@ int32_t taosGetQitem(STaosQall *qall, void **ppItem) { return num; } -void taosResetQitems(STaosQall *qall) { qall->current = qall->start; } +void taosResetQitems(STaosQall *qall) { qall->current = qall->start; } +int32_t taosQallItemSize(STaosQall *qall) { return qall->numOfItems; } STaosQset *taosOpenQset() { STaosQset *qset = taosMemoryCalloc(sizeof(STaosQset), 1); diff --git a/source/util/src/trbtree.c b/source/util/src/trbtree.c new file mode 100644 index 0000000000000000000000000000000000000000..0970485dade90bb8719a2fa39facb047e07bcfff --- /dev/null +++ b/source/util/src/trbtree.c @@ -0,0 +1,210 @@ +/* + * Copyright (c) 2019 TAOS Data, Inc. + * + * This program is free software: you can use, redistribute, and/or modify + * it under the terms of the GNU Affero General Public License, version 3 + * or later ("AGPL"), as published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. + * + * You should have received a copy of the GNU Affero General Public License + * along with this program. If not, see . + */ + +#include "os.h" + +typedef int32_t (*tRBTreeCmprFn)(void *, void *); + +typedef struct SRBTree SRBTree; +typedef struct SRBTreeNode SRBTreeNode; +typedef struct SRBTreeIter SRBTreeIter; + +struct SRBTreeNode { + enum { RED, BLACK } color; + SRBTreeNode *parent; + SRBTreeNode *left; + SRBTreeNode *right; + uint8_t payload[]; +}; + +struct SRBTree { + tRBTreeCmprFn cmprFn; + SRBTreeNode *root; +}; + +struct SRBTreeIter { + SRBTree *pTree; +}; + +#define RBTREE_NODE_COLOR(N) ((N) ? (N)->color : BLACK) + +// APIs ================================================ +static void tRBTreeRotateLeft(SRBTree *pTree, SRBTreeNode *pNode) { + SRBTreeNode *right = pNode->right; + + pNode->right = right->left; + if (pNode->right) { + pNode->right->parent = pNode; + } + + right->parent = pNode->parent; + if (pNode->parent == NULL) { + pTree->root = right; + } else if (pNode == pNode->parent->left) { + pNode->parent->left = right; + } else { + pNode->parent->right = right; + } + + right->left = pNode; + pNode->parent = right; +} + +static void tRBTreeRotateRight(SRBTree *pTree, SRBTreeNode *pNode) { + SRBTreeNode *left = pNode->left; + + pNode->left = left->right; + if (pNode->left) { + pNode->left->parent = pNode; + } + + left->parent = pNode->parent; + if (pNode->parent == NULL) { + pTree->root = left; + } else if (pNode == pNode->parent->left) { + pNode->parent->left = left; + } else { + pNode->parent->right = left; + } + + left->right = pNode; + pNode->parent = left; +} + +#define tRBTreeCreate(compare) \ + (SRBTree) { .cmprFn = (compare), .root = NULL } + +SRBTreeNode *tRBTreePut(SRBTree *pTree, SRBTreeNode *pNew) { + pNew->left = NULL; + pNew->right = NULL; + pNew->color = RED; + + // insert + if (pTree->root == NULL) { + pNew->parent = NULL; + pTree->root = pNew; + } else { + SRBTreeNode *pNode = pTree->root; + while (true) { + ASSERT(pNode); + + int32_t c = pTree->cmprFn(pNew->payload, pNode->payload); + if (c < 0) { + if (pNode->left) { + pNode = pNode->left; + } else { + pNew->parent = pNode; + pNode->left = pNew; + break; + } + } else if (c > 0) { + if (pNode->right) { + pNode = pNode->right; + } else { + pNew->parent = pNode; + pNode->right = pNew; + break; + } + } else { + return NULL; + } + } + } + + // fix + SRBTreeNode *pNode = pNew; + while (pNode->parent && pNode->parent->color == RED) { + SRBTreeNode *p = pNode->parent; + SRBTreeNode *g = p->parent; + + if (p == g->left) { + SRBTreeNode *u = g->right; + + if (RBTREE_NODE_COLOR(u) == RED) { + p->color = BLACK; + u->color = BLACK; + g->color = RED; + pNode = g; + } else { + if (pNode == p->right) { + pNode = p; + tRBTreeRotateLeft(pTree, pNode); + } + pNode->parent->color = BLACK; + pNode->parent->parent->color = RED; + tRBTreeRotateRight(pTree, pNode->parent->parent); + } + } else { + SRBTreeNode *u = g->left; + + if (RBTREE_NODE_COLOR(u) == RED) { + p->color = BLACK; + u->color = BLACK; + g->color = RED; + } else { + if (pNode == p->left) { + pNode = p; + tRBTreeRotateRight(pTree, pNode); + } + pNode->parent->color = BLACK; + pNode->parent->parent->color = RED; + tRBTreeRotateLeft(pTree, pNode->parent->parent); + } + } + } + + pTree->root->color = BLACK; + return pNew; +} + +SRBTreeNode *tRBTreeDrop(SRBTree *pTree, void *pKey) { + SRBTreeNode *pNode = pTree->root; + + while (pNode) { + int32_t c = pTree->cmprFn(pKey, pNode->payload); + + if (c < 0) { + pNode = pNode->left; + } else if (c > 0) { + pNode = pNode->right; + } else { + break; + } + } + + if (pNode) { + // TODO + } + + return pNode; +} + +SRBTreeNode *tRBTreeGet(SRBTree *pTree, void *pKey) { + SRBTreeNode *pNode = pTree->root; + + while (pNode) { + int32_t c = pTree->cmprFn(pKey, pNode->payload); + + if (c < 0) { + pNode = pNode->left; + } else if (c > 0) { + pNode = pNode->right; + } else { + break; + } + } + + return pNode; +} diff --git a/tests/script/tsim/alter/cached_schema_after_alter.sim b/tests/script/tsim/alter/cached_schema_after_alter.sim index bd2b1d272ce83525fc645451ea5a48bbaa2611be..30b879b6129c773373bea87b523fc876f838a194 100644 --- a/tests/script/tsim/alter/cached_schema_after_alter.sim +++ b/tests/script/tsim/alter/cached_schema_after_alter.sim @@ -14,7 +14,7 @@ print ========== cached_schema_after_alter.sim sql drop database $db -x step1 step1: -sql create database $db +sql create database $db print ====== create tables sql use $db @@ -32,10 +32,10 @@ if $rows != 1 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi sql select * from $tb2 if $rows != 1 then @@ -43,10 +43,10 @@ if $rows != 1 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi print ================== restart server to commit data into disk system sh/exec.sh -n dnode1 -s stop -x SIGINT @@ -61,10 +61,10 @@ if $rows != 1 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi sql select * from $tb2 print select * from $tb2 ==> $data00 $data01 $data02 @@ -73,10 +73,10 @@ if $rows != 1 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi $ts = $ts0 + $delta sql insert into $tb2 values ( $ts , 2, 2) @@ -86,16 +86,16 @@ if $rows != 2 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi if $data11 != 2 then return -1 -endi +endi if $data12 != 2 then return -1 -endi +endi sql select * from $tb2 order by ts asc if $rows != 2 then @@ -103,15 +103,15 @@ if $rows != 2 then endi if $data01 != 1 then return -1 -endi +endi if $data02 != 1 then return -1 -endi +endi if $data11 != 2 then return -1 -endi +endi if $data12 != 2 then return -1 -endi +endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/alter/dnode.sim b/tests/script/tsim/alter/dnode.sim index d773c1f8a936e24d3cf19b9300f580eb2970f318..be3c385d45a58a538d786a34fe59f74e5dc38678 100644 --- a/tests/script/tsim/alter/dnode.sim +++ b/tests/script/tsim/alter/dnode.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -print ======== step1 +print ======== step1 sql alter dnode 1 'resetlog' sql alter dnode 1 'monitor' '1' sql alter dnode 1 'monitor' '0' @@ -65,4 +65,4 @@ sql alter dnode 1 balance "vnode:2-dnode:1" -x step4 step4: print ======= over -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/alter/table.sim b/tests/script/tsim/alter/table.sim index 48ab7ddab050a41e7176df86ba882589715bad44..dccfc7f5d6f9bf2afb71cd0c44dc85e722758538 100644 --- a/tests/script/tsim/alter/table.sim +++ b/tests/script/tsim/alter/table.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -print ======== step1 +print ======== step1 sql create database d1 sql use d1 sql create table tb (ts timestamp, a int) diff --git a/tests/script/tsim/bnode/basic1.sim b/tests/script/tsim/bnode/basic1.sim index 003d0ceb3d118cdc727dbb0b436cee1a4ad3ddb1..0a200016368efffbdd360c891cc6f15d3a284b47 100644 --- a/tests/script/tsim/bnode/basic1.sim +++ b/tests/script/tsim/bnode/basic1.sim @@ -7,24 +7,24 @@ sql connect print =============== select * from information_schema.ins_dnodes sql select * from information_schema.ins_dnodes; -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 1 then +if $data00 != 1 then return -1 endi sql select * from information_schema.ins_mnodes; -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 1 then +if $data00 != 1 then return -1 endi -if $data02 != leader then +if $data02 != leader then return -1 endi @@ -33,62 +33,62 @@ sql create dnode $hostname port 7200 sleep 2000 sql select * from information_schema.ins_dnodes; -if $rows != 2 then +if $rows != 2 then return -1 endi -if $data00 != 1 then +if $data00 != 1 then return -1 endi -if $data10 != 2 then +if $data10 != 2 then return -1 endi print $data02 -if $data02 != 0 then +if $data02 != 0 then return -1 endi -if $data12 != 0 then +if $data12 != 0 then return -1 endi -if $data04 != ready then +if $data04 != ready then return -1 endi -if $data14 != ready then +if $data14 != ready then return -1 endi sql select * from information_schema.ins_mnodes; -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 1 then +if $data00 != 1 then return -1 endi -if $data02 != leader then +if $data02 != leader then return -1 endi #print =============== create drop bnode 1 #sql create bnode on dnode 1 #sql show bnodes -#if $rows != 1 then +#if $rows != 1 then # return -1 #endi -#if $data00 != 1 then +#if $data00 != 1 then # return -1 #endi #sql_error create bnode on dnode 1 # #sql drop bnode on dnode 1 #sql show bnodes -#if $rows != 0 then +#if $rows != 0 then # return -1 #endi #sql_error drop bnode on dnode 1 @@ -96,17 +96,17 @@ endi #print =============== create drop bnode 2 #sql create bnode on dnode 2 #sql show bnodes -#if $rows != 1 then +#if $rows != 1 then # return -1 #endi -#if $data00 != 2 then +#if $data00 != 2 then # return -1 #endi #sql_error create bnode on dnode 2 # #sql drop bnode on dnode 2 #sql show bnodes -#if $rows != 0 then +#if $rows != 0 then # return -1 #endi #sql_error drop bnode on dnode 2 @@ -115,7 +115,7 @@ endi #sql create bnode on dnode 1 #sql create bnode on dnode 2 #sql show bnodes -#if $rows != 2 then +#if $rows != 2 then # return -1 #endi @@ -127,7 +127,7 @@ endi # #sleep 2000 #sql show bnodes -#if $rows != 2 then +#if $rows != 2 then # return -1 #endi diff --git a/tests/script/tsim/compress/commitlog.sim b/tests/script/tsim/compress/commitlog.sim index bc9c231a9ec311fc500a4fb59bb46173ba19cf0e..38899b95ba9fbd7d156991c98279db878c8881d0 100644 --- a/tests/script/tsim/compress/commitlog.sim +++ b/tests/script/tsim/compress/commitlog.sim @@ -25,7 +25,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -46,7 +46,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -67,7 +67,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -83,7 +83,7 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi @@ -93,18 +93,18 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi - + $i = 2 $db = $dbPrefix . $i $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 -endi - -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +endi + +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/compress/compress.sim b/tests/script/tsim/compress/compress.sim index 766f97450c3b9469b2788fb12cb5099d91cc4163..4752f1ba50460901f98ec08bd97baba7b216b187 100644 --- a/tests/script/tsim/compress/compress.sim +++ b/tests/script/tsim/compress/compress.sim @@ -25,7 +25,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -47,7 +47,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -69,7 +69,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -85,7 +85,7 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi @@ -95,18 +95,18 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi - + $i = 2 $db = $dbPrefix . $i $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 -endi - -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +endi + +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/compress/compress2.sim b/tests/script/tsim/compress/compress2.sim index 87e50cce5bd273abcf82bf960adee78b96bac774..c55b74f2466c8d24e87c693b23321df5a372704c 100644 --- a/tests/script/tsim/compress/compress2.sim +++ b/tests/script/tsim/compress/compress2.sim @@ -26,7 +26,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -48,7 +48,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -70,7 +70,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -86,7 +86,7 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi @@ -96,18 +96,18 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi - + $i = 2 $db = $dbPrefix . $i $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 -endi - -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +endi + +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/compress/uncompress.sim b/tests/script/tsim/compress/uncompress.sim index ccd5db4b0cf473d763a761587830212729cf9810..f48fc6da2321f51742dd971c5e9ae05fb26cb827 100644 --- a/tests/script/tsim/compress/uncompress.sim +++ b/tests/script/tsim/compress/uncompress.sim @@ -26,7 +26,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -48,7 +48,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -70,7 +70,7 @@ while $count < $N endw sql select * from $tb -if $rows != $N then +if $rows != $N then return -1 endi @@ -85,7 +85,7 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi @@ -95,18 +95,18 @@ $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 endi - + $i = 2 $db = $dbPrefix . $i $tb = $tbPrefix . $i sql use $db sql select * from $tb print select * from $tb ==> $rows points -if $rows != $N then +if $rows != $N then return -1 -endi - -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +endi + +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/alter_option.sim b/tests/script/tsim/db/alter_option.sim index d81671eebdd6e64305988fb4ec058b6dd12fc147..3d260235f26d1bccd9094d118deebeceb35e82e5 100644 --- a/tests/script/tsim/db/alter_option.sim +++ b/tests/script/tsim/db/alter_option.sim @@ -186,13 +186,13 @@ sql_error alter database db replica 0 #sql alter database db replica 1 #sql select * from information_schema.ins_databases #print replica: $data4_db -#if $data4_db != 1 then +#if $data4_db != 1 then # return -1 #endi #sql alter database db replica 3 #sql select * from information_schema.ins_databases #print replica: $data4_db -#if $data4_db != 3 then +#if $data4_db != 3 then # return -1 #endi @@ -200,13 +200,13 @@ sql_error alter database db replica 0 #sql alter database db quorum 2 #sql select * from information_schema.ins_databases #print quorum $data5_db -#if $data5_db != 2 then +#if $data5_db != 2 then # return -1 #endi #sql alter database db quorum 1 #sql select * from information_schema.ins_databases #print quorum $data5_db -#if $data5_db != 1 then +#if $data5_db != 1 then # return -1 #endi @@ -233,7 +233,7 @@ endi #sql alter database db keep 1000,2000 #sql select * from information_schema.ins_databases #print keep $data7_db -#if $data7_db != 500,500,500 then +#if $data7_db != 500,500,500 then # return -1 #endi @@ -263,13 +263,13 @@ sql_error alter database db keep -1 #sql alter database db blocks 3 #sql select * from information_schema.ins_databases #print blocks $data9_db -#if $data9_db != 3 then +#if $data9_db != 3 then # return -1 #endi #sql alter database db blocks 11 #sql select * from information_schema.ins_databases #print blocks $data9_db -#if $data9_db != 11 then +#if $data9_db != 11 then # return -1 #endi @@ -300,13 +300,13 @@ print ============== step wal_level sql alter database db wal_level 1 sql select * from information_schema.ins_databases print wal_level $data20_db -if $data20_db != 1 then +if $data20_db != 1 then return -1 endi sql alter database db wal_level 2 sql select * from information_schema.ins_databases print wal_level $data20_db -if $data20_db != 2 then +if $data20_db != 2 then return -1 endi @@ -319,19 +319,19 @@ print ============== modify wal_fsync_period sql alter database db wal_fsync_period 2000 sql select * from information_schema.ins_databases print wal_fsync_period $data21_db -if $data21_db != 2000 then +if $data21_db != 2000 then return -1 endi sql alter database db wal_fsync_period 500 sql select * from information_schema.ins_databases print wal_fsync_period $data21_db -if $data21_db != 500 then +if $data21_db != 500 then return -1 endi sql alter database db wal_fsync_period 0 sql select * from information_schema.ins_databases print wal_fsync_period $data21_db -if $data21_db != 0 then +if $data21_db != 0 then return -1 endi sql_error alter database db wal_fsync_period 180001 @@ -351,31 +351,31 @@ print ============== modify cachelast [0, 1, 2, 3] sql alter database db cachemodel 'last_value' sql select * from information_schema.ins_databases print cachelast $data18_db -if $data18_db != last_value then +if $data18_db != last_value then return -1 endi sql alter database db cachemodel 'last_row' sql select * from information_schema.ins_databases print cachelast $data18_db -if $data18_db != last_row then +if $data18_db != last_row then return -1 endi sql alter database db cachemodel 'none' sql select * from information_schema.ins_databases print cachelast $data18_db -if $data18_db != none then +if $data18_db != none then return -1 endi sql alter database db cachemodel 'last_value' sql select * from information_schema.ins_databases print cachelast $data18_db -if $data18_db != last_value then +if $data18_db != last_value then return -1 endi sql alter database db cachemodel 'both' sql select * from information_schema.ins_databases print cachelast $data18_db -if $data18_db != both then +if $data18_db != both then return -1 endi diff --git a/tests/script/tsim/db/alter_replica_13.sim b/tests/script/tsim/db/alter_replica_13.sim index d232c9bcd3eee5422a4ae74ac3b401d0ba3420a6..1d06d3abb96b101d6fdaede7a321515a9999d8c5 100644 --- a/tests/script/tsim/db/alter_replica_13.sim +++ b/tests/script/tsim/db/alter_replica_13.sim @@ -36,10 +36,10 @@ endi print =============== step2: create database sql create database db vgroups 1 sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi -if $data(db)[4] != 1 then +if $data(db)[4] != 1 then return -1 endi @@ -82,7 +82,7 @@ step3: return -1 endi sql select * from information_schema.ins_dnodes -print ===> rows: $rows +print ===> rows: $rows print ===> $data00 $data01 $data02 $data03 $data04 $data05 print ===> $data10 $data11 $data12 $data13 $data14 $data15 print ===> $data20 $data21 $data22 $data23 $data24 $data25 @@ -115,7 +115,7 @@ step4: return -1 endi sql show db.vgroups -print ===> rows: $rows +print ===> rows: $rows print ===> $data00 $data01 $data02 $data03 $data04 $data05 if $data[0][4] != leader then goto step4 @@ -137,4 +137,4 @@ endi system sh/exec.sh -n dnode1 -s stop -x SIGINT system sh/exec.sh -n dnode2 -s stop -x SIGINT system sh/exec.sh -n dnode3 -s stop -x SIGINT -system sh/exec.sh -n dnode4 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode4 -s stop -x SIGINT diff --git a/tests/script/tsim/db/alter_replica_31.sim b/tests/script/tsim/db/alter_replica_31.sim index 17ab04052043be36f22e10d56ae06ddf717b758a..4ab6783d0702197fe32f34830a67c3cc64a7b4a7 100644 --- a/tests/script/tsim/db/alter_replica_31.sim +++ b/tests/script/tsim/db/alter_replica_31.sim @@ -23,7 +23,7 @@ step1: return -1 endi sql select * from information_schema.ins_dnodes -print ===> rows: $rows +print ===> rows: $rows print ===> $data00 $data01 $data02 $data03 $data04 $data05 print ===> $data10 $data11 $data12 $data13 $data14 $data15 print ===> $data20 $data21 $data22 $data23 $data24 $data25 @@ -47,10 +47,10 @@ endi print =============== step2: create database sql create database db vgroups 1 replica 3 sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi -if $data(db)[4] != 3 then +if $data(db)[4] != 3 then return -1 endi @@ -139,7 +139,7 @@ step3: return -1 endi sql show db.vgroups -print ===> rows: $rows +print ===> rows: $rows if $rows != 1 then goto step3 endi @@ -165,4 +165,4 @@ endi system sh/exec.sh -n dnode1 -s stop -x SIGINT system sh/exec.sh -n dnode2 -s stop -x SIGINT system sh/exec.sh -n dnode3 -s stop -x SIGINT -system sh/exec.sh -n dnode4 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode4 -s stop -x SIGINT diff --git a/tests/script/tsim/db/back_insert.sim b/tests/script/tsim/db/back_insert.sim index e2bdb3a64bb06f56f4fdaf1fba58913ac0b94808..b3f820729386df8ec4bfea69feaabf33f137b3ef 100644 --- a/tests/script/tsim/db/back_insert.sim +++ b/tests/script/tsim/db/back_insert.sim @@ -2,8 +2,8 @@ sql connect $x = 1 begin: sql reset query cache - sleep 100 - sql insert into db.tb values(now, $x ) -x begin - #print ===> insert successed $x - $x = $x + 1 -goto begin \ No newline at end of file + sleep 100 + sql insert into db.tb values(now, $x ) -x begin + #print ===> insert successed $x + $x = $x + 1 +goto begin diff --git a/tests/script/tsim/db/basic1.sim b/tests/script/tsim/db/basic1.sim index 679440590fe99648ee2907adbc027c3daa7aa526..69eeb9347b22b154c7609b2ff89e36aa43f63a82 100644 --- a/tests/script/tsim/db/basic1.sim +++ b/tests/script/tsim/db/basic1.sim @@ -25,15 +25,15 @@ endi print =============== show vgroups1 sql use d1 sql show vgroups -if $rows != 2 then +if $rows != 2 then return -1 endi -if $data00 != 2 then +if $data00 != 2 then return -1 endi -if $data10 != 3 then +if $data10 != 3 then return -1 endi @@ -59,11 +59,11 @@ if $rows != 2 then return -1 endi -if $data00 != 4 then +if $data00 != 4 then return -1 endi -if $data10 != 5 then +if $data10 != 5 then return -1 endi @@ -73,15 +73,15 @@ if $rows != 3 then return -1 endi -if $data00 != 6 then +if $data00 != 6 then return -1 endi -if $data10 != 7 then +if $data10 != 7 then return -1 endi -if $data20 != 8 then +if $data20 != 8 then return -1 endi @@ -91,19 +91,19 @@ if $rows != 4 then return -1 endi -if $data00 != 9 then +if $data00 != 9 then return -1 endi -if $data10 != 10 then +if $data10 != 10 then return -1 endi -if $data20 != 11 then +if $data20 != 11 then return -1 endi -if $data30 != 12 then +if $data30 != 12 then return -1 endi diff --git a/tests/script/tsim/db/basic2.sim b/tests/script/tsim/db/basic2.sim index 2ba7d806af2fe6cea4e9dea3c0ad31f86d5e728f..b7ac0b5edd8663f653cc9216bceb1eee6054331e 100644 --- a/tests/script/tsim/db/basic2.sim +++ b/tests/script/tsim/db/basic2.sim @@ -27,7 +27,7 @@ sql create table t3 (ts timestamp, i int); sql create table t4 (ts timestamp, i int); sql select * from information_schema.ins_databases -print rows: $rows +print rows: $rows print $data00 $data01 $data02 $data03 print $data10 $data11 $data12 $data13 if $rows != 3 then @@ -47,7 +47,7 @@ endi #endi sql show tables -if $rows != 4 then +if $rows != 4 then return -1 endi @@ -64,8 +64,8 @@ if $rows != 4 then endi sql show tables -if $rows != 3 then +if $rows != 3 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/basic3.sim b/tests/script/tsim/db/basic3.sim index 30faec0494fa05661a7434c14c894efe8d0f4915..db355213db84fcb8e22ced9dbef37486ef243982 100644 --- a/tests/script/tsim/db/basic3.sim +++ b/tests/script/tsim/db/basic3.sim @@ -23,12 +23,12 @@ if $data22 != 2 then return -1 endi -#if $data03 != 4 then +#if $data03 != 4 then # return -1 #endi sql show d1.tables -if $rows != 4 then +if $rows != 4 then return -1 endi @@ -44,8 +44,8 @@ if $rows != 4 then endi sql show d2.tables -if $rows != 3 then +if $rows != 3 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/basic4.sim b/tests/script/tsim/db/basic4.sim index f407c6352d38d858f8e918142ca820a76e668878..7a5e0ec76478d0a401a803b95b344172dfcef1c4 100644 --- a/tests/script/tsim/db/basic4.sim +++ b/tests/script/tsim/db/basic4.sim @@ -11,109 +11,109 @@ sql create table d1.t3 (ts timestamp, i int); sql create table d1.t4 (ts timestamp, i int); sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi -if $data20 != d1 then +if $data20 != d1 then return -1 endi -if $data22 != 1 then +if $data22 != 1 then return -1 endi -if $data24 != 1 then +if $data24 != 1 then return -1 endi sql show d1.tables -if $rows != 4 then +if $rows != 4 then return -1 endi sql show d1.vgroups -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 2 then +if $data00 != 2 then return -1 endi -if $data01 != d1 then +if $data01 != d1 then return -1 endi -print =============== drop table +print =============== drop table sql drop table d1.t1 sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi -if $data20 != d1 then +if $data20 != d1 then return -1 endi -if $data22 != 1 then +if $data22 != 1 then return -1 endi -if $data24 != 1 then +if $data24 != 1 then return -1 endi sql show d1.tables -if $rows != 3 then +if $rows != 3 then return -1 endi sql show d1.vgroups -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 2 then +if $data00 != 2 then return -1 endi -if $data01 != d1 then +if $data01 != d1 then return -1 endi -print =============== drop all table +print =============== drop all table sql drop table d1.t2 sql drop table d1.t3 sql drop table d1.t4 sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi -if $data20 != d1 then +if $data20 != d1 then return -1 endi -if $data22 != 1 then +if $data22 != 1 then return -1 endi -if $data24 != 1 then +if $data24 != 1 then return -1 endi sql show d1.tables -if $rows != 0 then +if $rows != 0 then return -1 endi sql show d1.vgroups -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != 2 then +if $data00 != 2 then return -1 endi -if $data01 != d1 then +if $data01 != d1 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/basic5.sim b/tests/script/tsim/db/basic5.sim index 9b809c35f09c28dc81cefedb24e3bd507d792f34..933fb8cf4b49a1d737b58864c6f9cd63f00b9a1e 100644 --- a/tests/script/tsim/db/basic5.sim +++ b/tests/script/tsim/db/basic5.sim @@ -13,13 +13,13 @@ sql create table tb1 using st1 tags(1); sql insert into tb1 values (now, 1); sql show stables -if $rows != 1 then +if $rows != 1 then print $rows return -1 endi sql show tables -if $rows != 1 then +if $rows != 1 then return -1 endi @@ -35,12 +35,12 @@ sql use db1; sql create stable st1 (ts timestamp, f1 int) tags(t1 int) sql show stables -if $rows != 1 then +if $rows != 1 then return -1 endi sql show tables -if $rows != 0 then +if $rows != 0 then return -1 endi diff --git a/tests/script/tsim/db/basic6.sim b/tests/script/tsim/db/basic6.sim index 2377a65ac0983f5a60d4bc5c881d2377d2c92a43..50435747874475c58b1c493abf935f15ca1f61d9 100644 --- a/tests/script/tsim/db/basic6.sim +++ b/tests/script/tsim/db/basic6.sim @@ -14,7 +14,7 @@ $st = $stPrefix . $i $tb = $tbPrefix . $i print =============== step1 -# quorum presicion +# quorum presicion sql create database $db vgroups 8 replica 1 duration 2 keep 10 minrows 80 maxrows 10000 wal_level 2 wal_fsync_period 1000 comp 0 cachemodel 'last_value' precision 'us' sql select * from information_schema.ins_databases print $data00 $data01 $data02 $data03 $data04 $data05 $data06 $data07 $data08 $data09 @@ -46,7 +46,7 @@ endi #if $data29 != 12 then # return -1 #endi - + print =============== step2 sql_error create database $db sql create database if not exists $db @@ -60,7 +60,7 @@ sql drop database $db sql select * from information_schema.ins_databases if $rows != 2 then return -1 -endi +endi print =============== step4 sql_error drop database $db @@ -102,22 +102,22 @@ while $i < 5 sql create table $tb using $st tags(1) sql show stables - if $rows != 1 then + if $rows != 1 then return -1 endi print $data00 $data01 $data02 $data03 - if $data00 != $st then + if $data00 != $st then return -1 endi sql show tables - if $rows != 1 then + if $rows != 1 then return -1 endi print $data00 $data01 $data02 $data03 - if $data00 != $tb then + if $data00 != $tb then return -1 endi @@ -127,8 +127,8 @@ endw print =============== step7 $i = 0 while $i < 5 - $db = $dbPrefix . $i - sql drop database $db + $db = $dbPrefix . $i + sql drop database $db $i = $i + 1 endw @@ -143,20 +143,20 @@ sql create table $st (ts timestamp, i int) tags (j int) sql create table $tb using $st tags(1) sql show stables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $st then +if $data00 != $st then return -1 endi sql show tables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $tb then +if $data00 != $tb then return -1 endi @@ -168,12 +168,12 @@ sql create database $db sql use $db sql show stables -if $rows != 0 then +if $rows != 0 then return -1 endi sql show tables -if $rows != 0 then +if $rows != 0 then return -1 endi @@ -182,20 +182,20 @@ sql create table $st (ts timestamp, i int) tags (j int) sql create table $tb using $st tags(1) sql show stables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $st then +if $data00 != $st then return -1 endi sql show tables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $tb then +if $data00 != $tb then return -1 endi @@ -207,12 +207,12 @@ sql create database $db sql use $db sql show stables -if $rows != 0 then +if $rows != 0 then return -1 endi sql show tables -if $rows != 0 then +if $rows != 0 then return -1 endi @@ -221,20 +221,20 @@ sql create table $st (ts timestamp, i int) tags (j int) sql create table $tb using $st tags(1) sql show stables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $st then +if $data00 != $st then return -1 endi sql show tables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $tb then +if $data00 != $tb then return -1 endi @@ -245,12 +245,12 @@ sql insert into $tb values (now+4a, 3) sql insert into $tb values (now+5a, 4) sql select * from $tb -if $rows != 5 then +if $rows != 5 then return -1 endi sql select * from $st -if $rows != 5 then +if $rows != 5 then return -1 endi @@ -262,12 +262,12 @@ sql create database $db sql use $db sql show stables -if $rows != 0 then +if $rows != 0 then return -1 endi sql show tables -if $rows != 0 then +if $rows != 0 then return -1 endi @@ -276,20 +276,20 @@ sql create table $st (ts timestamp, i int) tags (j int) sql create table $tb using $st tags(1) sql show stables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $st then +if $data00 != $st then return -1 endi sql show tables -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != $tb then +if $data00 != $tb then return -1 endi @@ -300,12 +300,12 @@ sql insert into $tb values (now+4a, 3) sql insert into $tb values (now+5a, 4) sql select * from $tb -if $rows != 5 then +if $rows != 5 then return -1 endi sql select * from $st -if $rows != 5 then +if $rows != 5 then return -1 endi diff --git a/tests/script/tsim/db/commit.sim b/tests/script/tsim/db/commit.sim index 731f2aa256f219f37aed66c8a08098f9b322706c..223324503411000433f33b890e095d1534027572 100644 --- a/tests/script/tsim/db/commit.sim +++ b/tests/script/tsim/db/commit.sim @@ -39,9 +39,9 @@ sql create table tb (ts timestamp, i int) $x = 1 while $x < 41 $time = $x . m - sql insert into tb values (now + $time , $x ) + sql insert into tb values (now + $time , $x ) $x = $x + 1 -endw +endw sql select * from tb order by ts desc print ===> rows $rows @@ -71,7 +71,7 @@ if $data01 != 40 then return -1 endi -$oldnum = $rows +$oldnum = $rows $num = $rows + 2 print ======== step3 import old data @@ -120,4 +120,4 @@ if $data01 != 40 then endi system sh/exec.sh -n dnode1 -s stop -x SIGINT -system sh/exec.sh -n dnode2 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode2 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_reuse1.sim b/tests/script/tsim/db/delete_reuse1.sim index 680fe6b2eddac8d61f4a8f1da8c03f7fc554e408..9dcb3c6ac1a3198698b1ba88192707262aa6a7ae 100644 --- a/tests/script/tsim/db/delete_reuse1.sim +++ b/tests/script/tsim/db/delete_reuse1.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -print ======== step1 +print ======== step1 sql create database d1 replica 1 vgroups 1 sql create database d2 replica 1 vgroups 1 sql create database d3 replica 1 vgroups 1 @@ -47,7 +47,7 @@ step2: print ========= step3 sql reset query cache -sleep 50 +sleep 50 sql create database d1 replica 1 sql create table d1.t1 (ts timestamp, i int) @@ -65,20 +65,20 @@ while $x < 20 sql insert into d1.t1 values(now, -1) -x step4 return -1 step4: - + sql create database d1 replica 1 sql reset query cache - sleep 50 + sleep 50 sql create table d1.t1 (ts timestamp, i int) sql insert into d1.t1 values(now, $x ) sql select * from d1.t1 if $rows != 1 then return -1 endi - + $x = $x + 1 - + print ===> loop times: $x -endw +endw -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_reuse2.sim b/tests/script/tsim/db/delete_reuse2.sim index d181b6b7806a30a70e2ba8ea05d2ab6022f037c8..4480b60b1b7a3eecfa605cfc4e287b4928bd971a 100644 --- a/tests/script/tsim/db/delete_reuse2.sim +++ b/tests/script/tsim/db/delete_reuse2.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -print ======== step1 +print ======== step1 sql create database d1 replica 1 sql create database d2 replica 1 sql create database d3 replica 1 @@ -48,7 +48,7 @@ step2: print ========= step3 sql create database db1 replica 1 sql reset query cache -sleep 50 +sleep 50 sql create table db1.tb1 (ts timestamp, i int) sql insert into db1.tb1 values(now, 2) @@ -61,7 +61,7 @@ print ========= step4 $x = 1 while $x < 20 - $db = db . $x + $db = db . $x $tb = tb . $x sql use $db sql drop database $db @@ -69,14 +69,14 @@ while $x < 20 sql insert into $tb values(now, -1) -x step4 return -1 step4: - - $x = $x + 1 - $db = db . $x + + $x = $x + 1 + $db = db . $x $tb = tb . $x - + sql reset query cache - sleep 50 - + sleep 50 + sql create database $db replica 1 sql use $db sql create table $tb (ts timestamp, i int) @@ -85,8 +85,8 @@ while $x < 20 if $rows != 1 then return -1 endi - + print ===> loop times: $x -endw +endw -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_reusevnode.sim b/tests/script/tsim/db/delete_reusevnode.sim index d194f82d084b6855841970f8b7ffb822557a6100..7af5c9d39d2c76cf6f2dcf245e219c7a81577405 100644 --- a/tests/script/tsim/db/delete_reusevnode.sim +++ b/tests/script/tsim/db/delete_reusevnode.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -print ======== step1 +print ======== step1 $tbPrefix = t $i = 0 @@ -21,13 +21,13 @@ while $i < 30 print times $i $i = $i + 1 -endw +endw print ======== step2 sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 -endi +endi system sh/stop_dnodes.sh @@ -94,4 +94,4 @@ if $rows != 2 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_reusevnode2.sim b/tests/script/tsim/db/delete_reusevnode2.sim index 754a6d695b6db790c85657129a751d873a796301..91473e5ee16d61f80eedf2bca85031426ebab92a 100644 --- a/tests/script/tsim/db/delete_reusevnode2.sim +++ b/tests/script/tsim/db/delete_reusevnode2.sim @@ -62,4 +62,4 @@ if $rows != 2 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_writing1.sim b/tests/script/tsim/db/delete_writing1.sim index 279f8dece872b7c7aea0c1f46dee600894dd22a3..6fec09989d83160a64b09e9781aa1422274baa61 100644 --- a/tests/script/tsim/db/delete_writing1.sim +++ b/tests/script/tsim/db/delete_writing1.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -sql create database db +sql create database db sql create table db.tb (ts timestamp, i int) sql insert into db.tb values(now, 1) @@ -11,18 +11,18 @@ print ======== start back run_back tsim/db/back_insert.sim sleep 1000 -print ======== step1 -$x = 1 +print ======== step1 +$x = 1 while $x < 10 print drop database times $x sql drop database if exists db - sql create database db + sql create database db sql create table db.tb (ts timestamp, i int) sleep 1000 - + $x = $x + 1 endw -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/delete_writing2.sim b/tests/script/tsim/db/delete_writing2.sim index 8eab126ae89472efac9e6f90c5b5e8a8ea033436..ad156f30eb0be57bff2ba3e696e02c8ce2c378bb 100644 --- a/tests/script/tsim/db/delete_writing2.sim +++ b/tests/script/tsim/db/delete_writing2.sim @@ -3,7 +3,7 @@ system sh/deploy.sh -n dnode1 -i 1 system sh/exec.sh -n dnode1 -s start sql connect -sql create database db +sql create database db sql create table db.tb (ts timestamp, i int) sql insert into db.tb values(now, 1) @@ -11,11 +11,11 @@ sql create database db2 sql create table db2.tb2 (ts timestamp, i int) sql insert into db2.tb2 values(now, 1) -sql create database db3 +sql create database db3 sql create table db3.tb3 (ts timestamp, i int) sql insert into db3.tb3 values(now, 1) -sql create database db4 +sql create database db4 sql create table db4.tb4 (ts timestamp, i int) sql insert into db4.tb4 values(now, 1) @@ -23,19 +23,19 @@ print ======== start back run_back tsim/db/back_insert.sim sleep 1000 -print ======== step1 -$x = 1 +print ======== step1 +$x = 1 while $x < 10 print drop database times $x sql drop database if exists db - sql create database db + sql create database db sql create table db.tb (ts timestamp, i int) sleep 1000 - + $x = $x + 1 endw -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/dropdnodes.sim b/tests/script/tsim/db/dropdnodes.sim index 8a46d5f9ce405cb84957900c37986cbede044464..20b4a136df102d4ec419b4a109192feb3768beef 100644 --- a/tests/script/tsim/db/dropdnodes.sim +++ b/tests/script/tsim/db/dropdnodes.sim @@ -12,7 +12,7 @@ system sh/exec.sh -n dnode1 -s start system sh/exec.sh -n dnode2 -s start sleep 2000 -sql connect +sql connect sql create dnode $hostname2 sleep 2000 @@ -61,13 +61,13 @@ sql show tables print $rows if $rows != 16 then return -1 -endi +endi sql select * from mt print $rows if $rows != 16 then return -1 -endi +endi print ========== step3 @@ -82,26 +82,26 @@ sql show tables print $rows if $rows != 8 then return -1 -endi +endi sql select * from mt print $rows if $rows != 8 then return -1 -endi +endi sql select * from db.t5 if $rows != 1 then return -1 -endi +endi sql select * from db.t13 if $rows != 1 then return -1 -endi +endi sql_error select * from db.t1 sql_error select * from db.t9 system sh/exec.sh -n dnode1 -s stop -x SIGINT -system sh/exec.sh -n dnode2 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode2 -s stop -x SIGINT diff --git a/tests/script/tsim/db/keep.sim b/tests/script/tsim/db/keep.sim index e146a666d083af5c0431ad1cd639b9321096c5f7..f0653c4801bd219023ce43904b98b9a92cf40616 100644 --- a/tests/script/tsim/db/keep.sim +++ b/tests/script/tsim/db/keep.sim @@ -14,7 +14,7 @@ while $x < 41 sql insert into tb values (now - $time , $x ) -x step2 step2: $x = $x + 1 -endw +endw sql select * from tb print ===> rows $rows last $data01 @@ -42,10 +42,10 @@ sql alter database keepdb keep 60 sql flush database keepdb sql select * from information_schema.ins_databases print $data00 $data01 $data02 $data03 $data04 $data05 $data06 $data07 -if $data22 != 2 then +if $data22 != 2 then return -1 endi -if $data27 != 86400m,86400m,86400m then +if $data27 != 86400m,86400m,86400m then return -1 endi @@ -56,7 +56,7 @@ while $x < 81 sql insert into tb values (now - $time , $x ) -x step4 step4: $x = $x + 1 -endw +endw sql select * from tb print ===> rows $rows last $data01 @@ -83,10 +83,10 @@ endi print ======== step6 alter db sql alter database keepdb keep 30 sql select * from information_schema.ins_databases -if $data22 != 2 then +if $data22 != 2 then return -1 endi -if $data27 != 43200m,43200m,43200m then +if $data27 != 43200m,43200m,43200m then return -1 endi @@ -110,7 +110,7 @@ while $x < 121 sql insert into tb values (now - $time , $x ) -x step8 step8: $x = $x + 1 -endw +endw sql select * from tb print ===> rows $rows last $data01 @@ -137,4 +137,4 @@ error3: print ======= test success system sh/exec.sh -n dnode1 -s stop -x SIGINT -system sh/exec.sh -n dnode2 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode2 -s stop -x SIGINT diff --git a/tests/script/tsim/db/len.sim b/tests/script/tsim/db/len.sim index ae475ddf47c8d327d317148fe27b22c52b336eaa..047dafd5f8c13ce8533babd1a0c90d9a93d6982c 100644 --- a/tests/script/tsim/db/len.sim +++ b/tests/script/tsim/db/len.sim @@ -11,33 +11,33 @@ sql create database -x step1 step1: sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi print =============== step2 sql create database a sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi sql drop database a sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi print =============== step3 sql create database a12345678 sql select * from information_schema.ins_databases -if $rows != 3 then +if $rows != 3 then return -1 endi sql drop database a12345678 sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi @@ -46,15 +46,15 @@ sql create database a012345678901201234567890120123456789012a0123456789012012345 return -1 step4: sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi print =============== step5 -sql create database a;1 +sql create database a;1 sql drop database a sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi @@ -64,7 +64,7 @@ sql create database a'1 -x step6 step6: sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi @@ -73,7 +73,7 @@ sql create database (a) -x step7 return -1 step7: sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi @@ -82,8 +82,8 @@ sql create database a.1 -x step8 return -1 step8: sql select * from information_schema.ins_databases -if $rows != 2 then +if $rows != 2 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/repeat.sim b/tests/script/tsim/db/repeat.sim index 98d66244f519769f7436cbcf52ec8c97b7cba15e..b0627659d082be380f768aa378f4202d86717958 100644 --- a/tests/script/tsim/db/repeat.sim +++ b/tests/script/tsim/db/repeat.sim @@ -56,4 +56,4 @@ sql drop database d10 sql drop database d11 sql drop database d12 -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/db/show_create_db.sim b/tests/script/tsim/db/show_create_db.sim index 45007d01d699a1cb3a04b8e08b174523540e78e3..3a51fedbff18e00cf087146c403aa027f857d3d3 100644 --- a/tests/script/tsim/db/show_create_db.sim +++ b/tests/script/tsim/db/show_create_db.sim @@ -7,7 +7,7 @@ print =============== step2 sql create database db sql show create database db -if $rows != 1 then +if $rows != 1 then return -1 endi @@ -15,13 +15,13 @@ print =============== step3 sql use db sql show create database db -if $rows != 1 then +if $rows != 1 then return -1 endi -if $data00 != db then +if $data00 != db then return -1 -endi +endi sql drop database db diff --git a/tests/script/tsim/db/show_create_table.sim b/tests/script/tsim/db/show_create_table.sim index 44fa09577e0edcc1d9da8d4563a8b6184965882f..0aeee42d21936736b97fd943aca448c6e9ce7046 100644 --- a/tests/script/tsim/db/show_create_table.sim +++ b/tests/script/tsim/db/show_create_table.sim @@ -11,14 +11,14 @@ sql create table t0 using meters tags(1,'ch') sql create table normalTbl(ts timestamp, zone binary(8)) sql use db -sql show create table meters -if $rows != 1 then +sql show create table meters +if $rows != 1 then return -1 endi print ===============check sub table sql show create table t0 -if $rows != 1 then +if $rows != 1 then return -1 endi if $data00 == 't0' then @@ -27,8 +27,8 @@ endi print ===============check normal table -sql show create table normalTbl -if $rows != 1 then +sql show create table normalTbl +if $rows != 1 then return -1 endi @@ -37,8 +37,8 @@ if $data00 == 'normalTbl' then endi print ===============check super table -sql show create table meters -if $rows != 1 then +sql show create table meters +if $rows != 1 then return -1 endi @@ -49,7 +49,7 @@ endi print ===============check sub table with prefix sql show create table db.t0 -if $rows != 1 then +if $rows != 1 then return -1 endi @@ -58,8 +58,8 @@ if $data00 == 't0' then endi print ===============check normal table with prefix -sql show create table db.normalTbl -if $rows != 1 then +sql show create table db.normalTbl +if $rows != 1 then return -1 endi @@ -69,8 +69,8 @@ endi print ===============check super table with prefix -sql show create table db.meters -if $rows != 1 then +sql show create table db.meters +if $rows != 1 then return -1 endi diff --git a/tests/script/tsim/db/tables.sim b/tests/script/tsim/db/tables.sim index cdee504753535c042f77a399911b1f11662f1e3b..273a1fd45d7b18457662dfecd6788e5e33389bec 100644 --- a/tests/script/tsim/db/tables.sim +++ b/tests/script/tsim/db/tables.sim @@ -8,7 +8,7 @@ sql create database db sql select * from information_schema.ins_databases print $rows $data07 -if $rows != 3 then +if $rows != 3 then return -1 endi @@ -125,4 +125,4 @@ if $data01 != 4 then return -1 endi -system sh/exec.sh -n dnode1 -s stop -x SIGINT \ No newline at end of file +system sh/exec.sh -n dnode1 -s stop -x SIGINT diff --git a/tests/script/tsim/parser/fill_stb.sim b/tests/script/tsim/parser/fill_stb.sim index 51ae6f4b4153f59356a73edbcf7002c774fd8c05..656b1ac94e8e0954e98b1d10692afc5d696bfd64 100644 --- a/tests/script/tsim/parser/fill_stb.sim +++ b/tests/script/tsim/parser/fill_stb.sim @@ -136,7 +136,8 @@ if $data74 != -4.00000 then endi ## fill(value) + group by -sql select max(c1), max(c2), max(c3), max(c4), max(c5) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, -1, -2, -3, -4, -5, -6, -7, -8) group by t1 +print select _wstart, max(c1), max(c2), max(c3), max(c4), max(c5) from $stb where ts >= $ts0 and ts <= $tsu partition by t1 interval(5m) fill(value, -1, -2, -3, -4, -5, -6, -7, -8) +sql select _wstart, max(c1), max(c2), max(c3), max(c4), max(c5) from $stb where ts >= $ts0 and ts <= $tsu partition by t1 interval(5m) fill(value, -1, -2, -3, -4, -5, -6, -7, -8) $val = $rowNum * 2 print $rowNum, $val @@ -148,18 +149,13 @@ if $rows != 190 then print expect 190, actual:$rows return -1 endi -if $data06 != 0 then - return -1 -endi if $data11 != -1 then return -1 endi -#if $data16 != 0 then -# return -1 -#endi # number of fill values is smaller than number of selected columns -sql select max(c1), max(c2), max(c3) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 6, 6) +print select _wstart, max(c1), max(c2), max(c3) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 6, 6) +sql select _wstart, max(c1), max(c2), max(c3) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 6, 6) if $data11 != 6 then return -1 endi @@ -174,11 +170,11 @@ endi sql_error select max(c1), max(c2), max(c3), max(c4), max(c5) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill (6, 6, 6, 6, 6) # fill_char_values_to_arithmetic_fields -sql_error select sum(c1), avg(c2), max(c3), min(c4), avg(c4), count(c6), last(c7), last(c8) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 'c', 'c', 'c', 'c', 'c', 'c', 'c', 'c') +sql select sum(c1), avg(c2), max(c3), min(c4), avg(c4), count(c6), last(c7), last(c8) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 'c', 'c', 'c', 'c', 'c', 'c', 'c', 'c') # fill_multiple_columns sql_error select sum(c1), avg(c2), min(c3), max(c4), count(c6), first(c7), last(c8) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 99, 99, 99, 99, 99, abc, abc) -sql select sum(c1), avg(c2), min(c3), max(c4) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 99, 99, 99, 99) +sql select _wstart, sum(c1), avg(c2), min(c3), max(c4) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 99, 99, 99, 99) $val = $rowNum * 2 $val = $val - 1 if $rows != $val then @@ -196,11 +192,14 @@ sql select * from $stb if $data09 != nchar0 then return -1 endi -sql select max(c4) from $stb where t1 > 4 and ts >= $ts0 and ts <= $tsu interval(5m) fill(value, -1) group by t1 -if $rows != 0 then - return -1 -endi -sql select min(c1), max(c4) from $stb where t1 > 4 and ts >= $ts0 and ts <= $tsu interval(5m) fill(value, -1) + +print select max(c4) from $stb where t1 > 4 and ts >= $ts0 and ts <= $tsu partition by t1 interval(5m) fill(value, -1) +sql select max(c4) from $stb where t1 > 4 and ts >= $ts0 and ts <= $tsu partition by t1 interval(5m) fill(value, -1) +#if $rows != 0 then +# return -1 +#endi + +sql select _wstart, min(c1), max(c4) from $stb where t1 > 4 and ts >= $ts0 and ts <= $tsu interval(5m) fill(value, -1) $val = $rowNum * 2 $val = $val - 1 if $rows != $val then @@ -223,11 +222,12 @@ if $data12 != -1.000000000 then endi # fill_into_nonarithmetic_fieds -sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 20000000, 20000000, 20000000) +print select _wstart, first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 20000000, 20000000, 20000000) +sql select _wstart, first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 20000000, 20000000, 20000000) #if $data11 != 20000000 then -if $data11 != 1 then - return -1 -endi +#if $data11 != 1 then +# return -1 +#endi sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 1, 1, 1) sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 1.1, 1.1, 1.1) @@ -235,16 +235,15 @@ sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= sql select first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '1e', '1e1') # fill quoted values into bool column will throw error unless the value is 'true' or 'false' Note:2018-10-24 # fill values into binary or nchar columns will be set to NULL automatically Note:2018-10-24 -sql_error select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '1e', '1e1','1e1') +sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '1e', '1e1','1e1') sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, true, true, true) sql select first(c7), first(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 'true', 'true','true') - # fill nonarithmetic values into arithmetic fields sql_error select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, abc); -sql_error select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 'true'); +sql select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 'true'); -sql select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '2e1'); +sql select _wstart, count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '2e1'); $val = $rowNum * 2 $val = $val - 1 if $rows != $val then @@ -253,11 +252,11 @@ endi if $data01 != $rowNum then return -1 endi -if $data11 != 20 then - return -1 -endi +#if $data11 != 20 then +# return -1 +#endi -sql select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 2e1); +sql select _wstart, count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, 2e1); if $rows != $val then return -1 endi @@ -268,43 +267,44 @@ if $data11 != 20 then return -1 endi -sql select count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '20'); +sql select _wstart, count(*) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(value, '20'); if $rows != $val then return -1 endi if $data01 != $rowNum then return -1 endi -if $data11 != 20 then - return -1 -endi +#if $data11 != 20 then +# return -1 +#endi ## linear fill -sql select max(c1), min(c2), avg(c3), sum(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(linear) group by t1 +sql select _wstart, max(c1), min(c2), avg(c3), sum(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu partition by t1 interval(5m) fill(linear) $val = $rowNum * 2 $val = $val - 1 $val = $val * $tbNum if $rows != $val then return -1 endi -if $data08 != 0 then - return -1 -endi -if $data15 != NULL then - return -1 -endi -if $data16 != NULL then - return -1 -endi -if $data17 != NULL then - return -1 -endi -if $data18 != 0 then - return -1 -endi +#if $data08 != 0 then +# return -1 +#endi +#if $data15 != NULL then +# return -1 +#endi +#if $data16 != NULL then +# return -1 +#endi +#if $data17 != NULL then +# return -1 +#endi +#if $data18 != 0 then +# return -1 +#endi ## [TBASE-365] -sql select max(c1), min(c2), avg(c3), sum(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 interval(5m) fill(linear) group by t1 +sql select _wstart, max(c1), min(c2), avg(c3), sum(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(linear) +print select _wstart, max(c1), min(c2), avg(c3), sum(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(linear) if $rows != 95 then return -1 endi @@ -332,14 +332,8 @@ endi if $data17 != NULL then return -1 endi -if $data08 != 5 then - return -1 -endi -if $data18 != 5 then - return -1 -endi -sql select max(c1), min(c2), sum(c3), avg(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(linear) +sql select _wstart, max(c1), min(c2), sum(c3), avg(c4), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu interval(5m) fill(linear) $val = $rowNum * 2 $val = $val - 1 if $rows != $val then @@ -359,7 +353,8 @@ endi ## previous fill print fill(prev) -sql select max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 interval(5m) fill(prev) group by t1 limit 5 +print select _wstart, max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(prev) limit 5 +sql select _wstart, max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(prev) limit 5 if $rows != 25 then return -1 endi @@ -372,69 +367,43 @@ endi if $data04 != NULL then return -1 endi -if $data09 != 5 then - return -1 -endi if $data12 != NULL then return -1 endi -if $data19 != 5 then - return -1 -endi if $data18 != nchar0 then return -1 endi -if $data59 != 6 then - return -1 -endi -if $data69 != 6 then - return -1 -endi ## NULL fill print fill(NULL) -sql select max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 interval(5m) fill(value, NULL) group by t1 limit 5 +print select _wstart, max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(value, NULL) limit 5 +sql select _wstart, max(c1), min(c2), avg(c3), sum(c4), count(c5), first(c7), last(c8), first(c9) from $stb where ts >= $ts0 and ts <= $tsu and t1 > 4 partition by t1 interval(5m) fill(value, NULL) limit 5 if $rows != 25 then return -1 endi if $data01 != 0 then return -1 endi -if $data02 != NULL then - return -1 -endi -if $data04 != NULL then +if $data02 != 0 then return -1 endi if $data06 != 1 then return -1 endi -if $data09 != 5 then +if $data11 != 0 then return -1 endi -if $data11 != NULL then - return -1 -endi -if $data12 != NULL then - return -1 -endi -if $data19 != 5 then +if $data12 != 0 then return -1 endi if $data18 != NULL then return -1 endi -if $data59 != 6 then - return -1 -endi -if $data69 != 6 then - return -1 -endi print =============== clear sql drop database $db sql select * from information_schema.ins_databases -if $rows != 0 then +if $rows != 2 then return -1 endi diff --git a/tests/script/tsim/parser/function.sim b/tests/script/tsim/parser/function.sim index 0219a84c646eea8aecc5dee6223bb29f4df5b44f..0002a5d09555083569fc7a6ff32a376e56746dfe 100644 --- a/tests/script/tsim/parser/function.sim +++ b/tests/script/tsim/parser/function.sim @@ -70,6 +70,7 @@ if $data00 != @15-08-18 00:00:00.000@ then return -1 endi if $data01 != 2.068333156 then + print expect 2.068333156, actual: $data01 return -1 endi if $data02 != 2.063999891 then @@ -128,6 +129,7 @@ if $data03 != 2 then return -1 endi if $data11 != 2.077099980 then + print expect 2.077099980, actual: $data11 return -1 endi if $data12 != 2.077000022 then diff --git a/tests/script/tsim/sma/rsmaPersistenceRecovery.sim b/tests/script/tsim/sma/rsmaPersistenceRecovery.sim index f53cd45d484cccf0b7ab564c80abb700e3ef9f19..faff48b61c1216e744d349a301f39fabd6dc578a 100644 --- a/tests/script/tsim/sma/rsmaPersistenceRecovery.sim +++ b/tests/script/tsim/sma/rsmaPersistenceRecovery.sim @@ -35,6 +35,7 @@ sleep 7000 print =============== select * from retention level 2 from memory sql select * from ct1; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 2 file rows $rows > 2 return -1 @@ -51,6 +52,7 @@ endi print =============== select * from retention level 1 from memory sql select * from ct1 where ts > now-8d; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 1 file rows $rows > 2 return -1 @@ -89,6 +91,7 @@ system sh/exec.sh -n dnode1 -s start print =============== select * from retention level 2 from file sql select * from ct1; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 2 file rows $rows > 2 return -1 @@ -104,6 +107,7 @@ endi print =============== select * from retention level 1 from file sql select * from ct1 where ts > now-8d; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 1 file rows $rows > 2 return -1 @@ -141,6 +145,7 @@ sleep 7000 print =============== select * from retention level 2 from file and memory after rsma qtaskinfo recovery sql select * from ct1; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 2 file/mem rows $rows > 2 return -1 @@ -163,6 +168,7 @@ endi print =============== select * from retention level 1 from file and memory after rsma qtaskinfo recovery sql select * from ct1 where ts > now-8d; print $data00 $data01 $data02 +print $data10 $data11 $data12 if $rows > 2 then print retention level 1 file/mem rows $rows > 2 return -1 diff --git a/tests/script/tsim/valgrind/checkError6.sim b/tests/script/tsim/valgrind/checkError6.sim index 00de00f71d06810e9d2a72f2b8d06bad5aa42266..fcc5b04c907852f87c469a3dc9d32c5ba1295327 100644 --- a/tests/script/tsim/valgrind/checkError6.sim +++ b/tests/script/tsim/valgrind/checkError6.sim @@ -114,8 +114,8 @@ sql select tbcol5 - tbcol3 from stb sql select spread( tbcol2 )/44, spread(tbcol2), 0.204545455 * 44 from stb; sql select min(tbcol) * max(tbcol) /4, sum(tbcol2) * apercentile(tbcol2, 20), apercentile(tbcol2, 33) + 52/9 from stb; sql select distinct(tbname), tgcol from stb; -#sql select sum(tbcol) from stb partition by tbname interval(1s) slimit 1 soffset 1; -#sql select sum(tbcol) from stb partition by tbname interval(1s) slimit 2 soffset 4 limit 10 offset 1; +sql select sum(tbcol) from stb partition by tbname interval(1s) slimit 1 soffset 1; +sql select sum(tbcol) from stb partition by tbname interval(1s) slimit 2 soffset 4 limit 10 offset 1; print =============== step5: explain sql explain analyze select ts from stb where -2; diff --git a/tests/script/tsim/valgrind/checkError7.sim b/tests/script/tsim/valgrind/checkError7.sim index a66ddb30df063416e0f04da0dd58de9bebed186d..af42d1e76b50bc07e9c7a484bd24c9544517f980 100644 --- a/tests/script/tsim/valgrind/checkError7.sim +++ b/tests/script/tsim/valgrind/checkError7.sim @@ -66,7 +66,7 @@ $null= system_content sh/checkValgrind.sh -n dnode1 print cmd return result ----> [ $system_content ] -if $system_content > 2 then +if $system_content > 0 then return -1 endi diff --git a/tests/script/tsim/valgrind/checkError8.sim b/tests/script/tsim/valgrind/checkError8.sim index 7ca01bc3d04a489e389939221b88b9aa9432e939..2f204768eb1fc8922a07853155edfc29e97c3975 100644 --- a/tests/script/tsim/valgrind/checkError8.sim +++ b/tests/script/tsim/valgrind/checkError8.sim @@ -143,7 +143,7 @@ $null= system_content sh/checkValgrind.sh -n dnode1 print cmd return result ----> [ $system_content ] -if $system_content > 2 then +if $system_content > 0 then return -1 endi diff --git a/tests/system-test/1-insert/time_range_wise.py b/tests/system-test/1-insert/time_range_wise.py index 8f61da221d5484765c6b03f1431bc7ce264d4d4a..c31d8d25475e24107c7319b86ad01112fbe93272 100644 --- a/tests/system-test/1-insert/time_range_wise.py +++ b/tests/system-test/1-insert/time_range_wise.py @@ -293,7 +293,7 @@ class TDTestCase: dbname = tdSql.getData(0,0) tdSql.query("select * from information_schema.ins_databases") for index , value in enumerate(tdSql.cursor.description): - if value[0] == "retention": + if value[0] == "retentions": r_index = index break for row in tdSql.queryResult: diff --git a/tests/system-test/2-query/interp.py b/tests/system-test/2-query/interp.py index 9348a8ca8f520e3ee8ea7f1c56f211dd53cd7dd4..934ba9e161c8787dc36cfdafc15044eb9e0ec425 100644 --- a/tests/system-test/2-query/interp.py +++ b/tests/system-test/2-query/interp.py @@ -551,6 +551,29 @@ class TDTestCase: tdSql.checkData(0, 0, 15) tdSql.checkData(1, 0, 15) + tdLog.printNoPrefix("==========step9:test error cases") + + tdSql.error(f"select interp(c0) from {dbname}.{tbname}") + tdSql.error(f"select interp(c0) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05')") + tdSql.error(f"select interp(c0) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d)") + tdSql.error(f"select interp(c0) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') fill(null)") + tdSql.error(f"select interp(c0) from {dbname}.{tbname} every(1s) fill(null)") + tdSql.error(f"select interp(c0) from {dbname}.{tbname} where ts >= '2020-02-10 00:00:05' and ts <= '2020-02-15 00:00:05' every(1s) fill(null)") + + # input can only be numerical types + tdSql.error(f"select interp(ts) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(c6) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(c7) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(c8) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + + # input can only be columns + tdSql.error(f"select interp(1) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(1.5) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(true) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp(false) from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp('abcd') from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + tdSql.error(f"select interp('中文字符') from {dbname}.{tbname} range('2020-02-10 00:00:05', '2020-02-15 00:00:05') every(1d) fill(null)") + def stop(self): tdSql.close() tdLog.success(f"{__file__} successfully executed") diff --git a/tools/CMakeLists.txt b/tools/CMakeLists.txt index 25d6e3317598da57afa836ca45476dd752d17f3b..5eec1746185cc31ebcf7a19f18bfcc4fb198f9d8 100644 --- a/tools/CMakeLists.txt +++ b/tools/CMakeLists.txt @@ -128,6 +128,7 @@ ELSE () COMMAND cmake -E copy ./taosadapter.service ${CMAKE_BINARY_DIR}/test/cfg/ COMMAND cmake -E copy taosadapter-debug ${CMAKE_BINARY_DIR}/build/bin ) + unset(_upx_prefix) ELSEIF (TD_DARWIN) include(ExternalProject) ExternalProject_Add(taosadapter @@ -149,8 +150,42 @@ ELSE () COMMAND cmake -E copy ./taosadapter.service ${CMAKE_BINARY_DIR}/test/cfg/ COMMAND cmake -E copy taosadapter-debug ${CMAKE_BINARY_DIR}/build/bin ) +# unset(_upx_prefix) + ELSEIF (TD_WINDOWS) + include(ExternalProject) + set(_upx_prefix "${CMAKE_BINARY_DIR}/.taos/externals/upx") + ExternalProject_Add(upx + PREFIX "${_upx_prefix}" + URL https://github.com/upx/upx/releases/download/v3.96/upx-3.96-win32.zip + CONFIGURE_COMMAND cmake -E true + BUILD_COMMAND cmake -E true + INSTALL_COMMAND cmake -E true + ) + + ExternalProject_Add(taosadapter + PREFIX "taosadapter" + SOURCE_DIR ${CMAKE_CURRENT_SOURCE_DIR}/taosadapter + BUILD_ALWAYS off + DEPENDS taos + BUILD_IN_SOURCE 1 + CONFIGURE_COMMAND cmake -E echo "taosadapter no need cmake to config" + PATCH_COMMAND + COMMAND git clean -f -d + BUILD_COMMAND + COMMAND set CGO_CFLAGS=-I${CMAKE_CURRENT_SOURCE_DIR}/../include/client + COMMAND set CGO_LDFLAGS=-L${CMAKE_BINARY_DIR}/build/lib + COMMAND go build -a -o taosadapter.exe -ldflags "-s -w -X github.com/taosdata/taosadapter/version.Version=${taos_version} -X github.com/taosdata/taosadapter/version.CommitID=${taosadapter_commit_sha1}" + COMMAND go build -a -o taosadapter-debug.exe -ldflags "-X github.com/taosdata/taosadapter/version.Version=${taos_version} -X github.com/taosdata/taosadapter/version.CommitID=${taosadapter_commit_sha1}" + INSTALL_COMMAND + COMMAND ${_upx_prefix}/src/upx/upx taosadapter.exe + COMMAND cmake -E copy taosadapter.exe ${CMAKE_BINARY_DIR}/build/bin + COMMAND cmake -E make_directory ${CMAKE_BINARY_DIR}/test/cfg/ + COMMAND cmake -E copy ./example/config/taosadapter.toml ${CMAKE_BINARY_DIR}/test/cfg/ + COMMAND cmake -E copy ./taosadapter.service ${CMAKE_BINARY_DIR}/test/cfg/ + COMMAND cmake -E copy taosadapter-debug.exe ${CMAKE_BINARY_DIR}/build/bin + ) unset(_upx_prefix) ELSE () - MESSAGE("${Yellow} Windows system still use original embedded httpd ${ColourReset}") + MESSAGE("${Yellow} taosAdapter Not supported yet ${ColourReset}") ENDIF () ENDIF ()