diff --git a/docs-cn/12-taos-sql/07-function.md b/docs-cn/12-taos-sql/07-function.md index 78389609db6b471ed2fb7fc0628bb7dbe72f61b9..2349e6aa3c02eb62fba1fc7e4eef15e08e3924d1 100644 --- a/docs-cn/12-taos-sql/07-function.md +++ b/docs-cn/12-taos-sql/07-function.md @@ -321,6 +321,32 @@ taos> SELECT HISTOGRAM(voltage, 'log_bin', '{"start": 1, "factor": 3, "count": 3 {"lower_bin":27, "upper_bin":inf, "count":1} | ``` +### ELAPSED + +```mysql +SELECT ELAPSED(field_name[, time_unit]) FROM { tb_name | stb_name } [WHERE clause] [INTERVAL(interval [, offset]) [SLIDING sliding]]; +``` + +**功能说明**:elapsed函数表达了统计周期内连续的时间长度,和twa函数配合使用可以计算统计曲线下的面积。在通过INTERVAL子句指定窗口的情况下,统计在给定时间范围内的每个窗口内有数据覆盖的时间范围;如果没有INTERVAL子句,则返回整个给定时间范围内的有数据覆盖的时间范围。注意,ELAPSED返回的并不是时间范围的绝对值,而是绝对值除以time_unit所得到的单位个数。 + +**返回结果类型**:Double + +**应用字段**:Timestamp类型 + +**支持的版本**:2.6.0.0 及以后的版本。 + +**适用于**: 表,超级表,嵌套查询的外层查询 + +**说明**: +- field_name参数只能是表的第一列,即timestamp主键列。 +- 按time_unit参数指定的时间单位返回,最小是数据库的时间分辨率。time_unit参数未指定时,以数据库的时间分辨率为时间单位。 +- 可以和interval组合使用,返回每个时间窗口的时间戳差值。需要特别注意的是,除第一个时间窗口和最后一个时间窗口外,中间窗口的时间戳差值均为窗口长度。 +- order by asc/desc不影响差值的计算结果。 +- 对于超级表,需要和group by tbname子句组合使用,不可以直接使用。 +- 对于普通表,不支持和group by子句组合使用。 +- 对于嵌套查询,仅当内层查询会输出隐式时间戳列时有效。例如select elapsed(ts) from (select diff(value) from sub1)语句,diff函数会让内层查询输出隐式时间戳列,此为主键列,可以用于elapsed函数的第一个参数。相反,例如select elapsed(ts) from (select * from sub1) 语句,ts列输出到外层时已经没有了主键列的含义,无法使用elapsed函数。此外,elapsed函数作为一个与时间线强依赖的函数,形如select elapsed(ts) from (select diff(value) from st group by tbname)尽管会返回一条计算结果,但并无实际意义,这种用法后续也将被限制。 +- 不支持与leastsquares、diff、derivative、top、bottom、last_row、interp等函数混合使用。 + ## 选择函数 在使用所有的选择函数的时候,可以同时指定输出 ts 列或标签列(包括 tbname),这样就可以方便地知道被选出的值是源于哪个数据行的。 diff --git a/docs-en/12-taos-sql/07-function.md b/docs-en/12-taos-sql/07-function.md index 1c2961d7b59ffbfb6df7d8488953b2d4a17201e9..3589efe9cdd618110203a8439ba03eaaf315f48c 100644 --- a/docs-en/12-taos-sql/07-function.md +++ b/docs-en/12-taos-sql/07-function.md @@ -327,6 +327,32 @@ taos> SELECT HISTOGRAM(voltage, 'log_bin', '{"start": 1, "factor": 3, "count": 3 {"lower_bin":27, "upper_bin":inf, "count":1} | ``` +### ELAPSED + +```mysql +SELECT ELAPSED(field_name[, 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 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 + +**Applicable Column type**:Timestamp + +**Applicable versions**:Sicne version 2.6.0.0 + +**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. +- 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 can't be used with `leastsquares`, `diff`, `derivative`, `top`, `bottom`, `last_row`, `interp`. + ## Selection Functions When any select function is used, timestamp column or tag columns including `tbname` can be specified to show that the selected value are from which rows.