- 05 1月, 2021 3 次提交
-
-
由 Timo Walther 提交于
This adds a name-based field mode to the Row class. A row can operate in 3 different modes: name-based, position-based, or a hybrid of both when leaving the Flink runtime. It simplifies the handling of large rows (possibly with hundreds of fields) and will make it easier to switch between DataStream API and Table API. See the documentation of the Row class for more information. This closes #14420.
-
由 Rui Li 提交于
This fixes the ClassNotFound problem when create a generic but not loaded function in Hive catalog This closes #14534
-
由 Till Rohrmann 提交于
In order to maintain the invariant that a failed createPartitionRequestClient won't leave a failed connection future in the clients field, this commit removes the client future if its creation fails. This closes #14528.
-
- 04 1月, 2021 5 次提交
-
-
由 Matthias Pohl 提交于
-
由 wangyemao-github 提交于
-
由 Matthias Pohl 提交于
-
由 paul8263 提交于
-
由 Yuan Mei 提交于
-
- 02 1月, 2021 1 次提交
-
-
由 HuangXingBo 提交于
-
- 01 1月, 2021 1 次提交
-
-
由 Shuo Cheng 提交于
Co-authored-by: Nshuo.cs <shuo.cs@alibaba-inc.com>
-
- 31 12月, 2020 5 次提交
-
-
由 zhuxiaoshang 提交于
Co-authored-by: Nzhuxiaoshang <zhushang@qutoutiao.net>
-
由 huangxingbo 提交于
This closes #14524.
-
由 Mulan 提交于
This closes #14527
-
由 SteNicholas 提交于
This closes #14464
-
由 Danny Chan 提交于
Some of the rel data types for legacy planner: - GenericRelDataType - ArrayRelDataType - MultisetRelDataType Do not implement the digest correctly, especially for GenericRelDataType , the RelDataTypeFactory caches the type instances based on its digest, a wrong digest impl would mess up the type instance creation, e.g. without this patch, all the GenericRelDataType instances have same digest `ANY`. The MapRelDataType is not fixed because we ignore the key/value nullability when creating the type, which conflicts with the digest. There is already issue (FLINK-20785) to track this. This closes #14451
-
- 30 12月, 2020 17 次提交
-
-
由 Chesnay Schepler 提交于
-
由 Chesnay Schepler 提交于
-
由 Chesnay Schepler 提交于
-
由 Matthias Pohl 提交于
This closes #14521.
-
由 Matthias Pohl 提交于
[FLINK-20594][runtime] Replaced DefaultExecutionSlotAllocator and its Factory class by SlotSharingExecutionSlotAllocator and its Factory. TestingPhysicalSlotProvider was moved into its own class file and generalized to match other testcases besides SlotSharingExecutionSlotAllocatorTest. This closes #14465.
-
由 Chesnay Schepler 提交于
-
由 xiaoHoly 提交于
[FLINK-18654][docs][jdbc] Correct missleading documentation in "Partitioned Scan" section of JDBC connector This closes #14523
-
由 Chesnay Schepler 提交于
-
由 AkisAya 提交于
[FLINK-20606][hive][table] Fix CREATE FUNCTION is failed under Hive catalog in SQL CLI using -j option This closes #14392
-
由 Roman Khachatryan 提交于
This reduces the number of failures in UnalignedCheckpointITCase.union to ~2/100.
-
由 Roman Khachatryan 提交于
In scenarios with multiple inputs (e.g. co-group; not union) one input may receive a checkpoint barrier while the second input is still restoring state. This (previous) state is currently not included into the snapshot, which therefore will be incomplete.
-
由 Roman Khachatryan 提交于
In StreamTaskNetworkInput.prepareSnapshot, internal channelIndex is inconsistent with CheckpointedInputGate channelIndex. This change fixes data loss in UnalignedCheckpointITCase.cogroup.
-
由 Roman Khachatryan 提交于
Most of the bugs in UC are revealed with higher back-pressure which is not created with alignment timeout. This change disables it by default and adds a new test (p=20) with the timeout enabled.
-
由 yangsanity 提交于
This closes #14437
-
由 huangxingbo 提交于
[FLINK-20693][table-planner-blink][python] Port BatchExecPythonCorrelate and StreamExecPythonCorrelate to Java This closes #14503.
-
由 Till Rohrmann 提交于
The DeclarativeSlotPool.registerNewSlotsListener can be used to register a DeclarativeSlotPool.NewSlotsListener which is notified whenever the DeclarativeSlotPool receives new slots or reserved slots are freed. This closes #14482.
-
由 Till Rohrmann 提交于
This closes #14518.
-
- 29 12月, 2020 5 次提交
-
-
由 Chesnay Schepler 提交于
-
由 Chesnay Schepler 提交于
It was not possible to just run 'mvn spotless:check', but instead one needed to use the fully qualified name of the plugin, i.e., 'mvn com.diffplug.spotless:spotless-maven-plugin:check'. This is due to 'force-shading'; this module is declared as a child of the flink-parent pom (i.e., the root pom of the project), but is not actually declaring that very module as it's parent. This seems to throw of maven, which fails to retrieve the plugin information from the parent. Note that this is just a band-aid; a proper fix would declare flink-parent as the parent of force-shading, but this is the easier fix.
-
由 Yangze Guo 提交于
[FLINK-17331][runtime] Explicitly get the ByteBuf length of all classes which is written to NettyMessage Currently, the length of some header fields in NettyMessage is hardcoded: InputChannelID, ExecutionAttemptID, e.t.c. So if we make some changes for such field, then we are not ware that it also needs to change the respective length for related netty messages component. This PR explicitly get the ByteBuf length of all classes which is written to NettyMessage to avoid such problems.
-
由 Rainie Li 提交于
Display more informative message stating that Watermarks are only available when using event time if there are no watermarks to display in the web UI. This closes #14483.
-
由 huangxingbo 提交于
This closes #14514.
-
- 28 12月, 2020 3 次提交
-
-
由 Piotr Nowojski 提交于
-
由 Aljoscha Krettek 提交于
-
由 Aljoscha Krettek 提交于
-