From 15146174aa5a24b0425a2335667397b4b73538c1 Mon Sep 17 00:00:00 2001 From: factosea <285808407@qq.com> Date: Thu, 13 Mar 2025 18:43:31 +0800 Subject: [PATCH 1/4] docs: interp --- docs/en/14-reference/03-taos-sql/06-select.md | 9 +++ .../14-reference/03-taos-sql/10-function.md | 69 +++++++++---------- docs/zh/14-reference/03-taos-sql/06-select.md | 9 +++ .../14-reference/03-taos-sql/10-function.md | 66 ++++++++---------- 4 files changed, 79 insertions(+), 74 deletions(-) diff --git a/docs/en/14-reference/03-taos-sql/06-select.md b/docs/en/14-reference/03-taos-sql/06-select.md index 3b89c4fe5a..98559104c0 100644 --- a/docs/en/14-reference/03-taos-sql/06-select.md +++ b/docs/en/14-reference/03-taos-sql/06-select.md @@ -277,6 +277,15 @@ TDengine supports INNER JOIN based on the timestamp primary key, with the follow 5. Both sides of JOIN support subqueries. 6. Does not support mixing with the FILL clause. +## INTERP +The INTERP clause is used in conjunction with the [INTERP](../function/#interp) function to determine the rules for capturing time slices and interpolation when the INTERP function is executed. +- The output time range for INTERP is specified by the RANGE(timestamp1, timestamp2) field, which must satisfy timestamp1 \<= timestamp2. Here, timestamp1 is the start value of the output time range, i.e., if the conditions for interpolation are met at timestamp1, then timestamp1 is the first record output, and timestamp2 is the end value of the output time range, i.e., the timestamp of the last record output cannot be greater than timestamp2. +- INTERP determines the number of results within the output time range based on the EVERY(time_unit) field, starting from timestamp1 and interpolating at fixed intervals of time (time_unit value), where time_unit can be time units: 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), 1w (weeks). For example, EVERY(500a) will interpolate the specified data every 500 milliseconds. +- INTERP determines how to interpolate at each time point that meets the output conditions based on the FILL field. For how to use the FILL clause, refer to [FILL Clause](../time-series-extensions/) +- INTERP can interpolate at a single time point specified in the RANGE field, in which case the EVERY field can be omitted. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00') FILL(linear). +- INTERP query supports NEAR FILL mode, i.e., when FILL is needed, it uses the data closest to the current time point for interpolation. When the timestamps before and after are equally close to the current time slice, FILL the previous row's value. This mode is not supported in stream computing and window queries. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', '2023-01-01 00:10:00') FILL(NEAR).(Supported from version 3.3.4.9). +- INTERP `RANGE` clause supports the expansion of the time range (supported from version 3.3.4.9), such as `RANGE('2023-01-01 00:00:00', 10s)` means to find data 10s before and after the time point '2023-01-01 00:00:00' for interpolation, FILL PREV/NEXT/NEAR respectively means to look for data forward/backward/around the time point, if there is no data around the time point, then use the value specified by FILL for interpolation, therefore the FILL clause must specify a value at this time. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', 10s) FILL(PREV, 1). Currently, only the combination of time point and time range is supported, not the combination of time interval and time range, i.e., RANGE('2023-01-01 00:00:00', '2023-02-01 00:00:00', 1h) is not supported. The specified time range rules are similar to EVERY, the unit cannot be year or month, the value cannot be 0, and cannot have quotes. When using this extension, other FILL modes except FILL PREV/NEXT/NEAR are not supported, and the EVERY clause cannot be specified. + ## GROUP BY If a GROUP BY clause is specified in the statement, the SELECT list can only contain the following expressions: diff --git a/docs/en/14-reference/03-taos-sql/10-function.md b/docs/en/14-reference/03-taos-sql/10-function.md index 8397c59177..27b5d2a577 100644 --- a/docs/en/14-reference/03-taos-sql/10-function.md +++ b/docs/en/14-reference/03-taos-sql/10-function.md @@ -1932,43 +1932,6 @@ FIRST(expr) - If all values in a column in the result set are NULL, the return for that column is also NULL; - If all columns in the result set are NULL, no results are returned. - For tables with composite primary keys, if there are multiple entries with the smallest timestamp, only the data with the smallest composite primary key is returned. - -### INTERP - -```sql -INTERP(expr [, ignore_null_values]) - -ignore_null_values: { - 0 - | 1 -} -``` - -**Function Description**: Returns the record value or interpolated value of a specified column at a specified time slice. The ignore_null_values parameter can be 0 or 1, where 1 means to ignore NULL values, default is 0. - -**Return Data Type**: Same as the field type. - -**Applicable Data Types**: Numeric types. - -**Applicable to**: Tables and supertables. - -**Usage Instructions** - -- INTERP is used to obtain the record value of a specified column at a specified time slice. If there is no row data that meets the conditions at that time slice, interpolation will be performed according to the settings of the FILL parameter. -- The input data for INTERP is the data of the specified column, which can be filtered through conditional statements (where clause). If no filtering condition is specified, the input is all data. -- INTERP SQL queries need to be used together with the RANGE, EVERY, and FILL keywords; stream computing cannot use RANGE, needs EVERY and FILL keywords together. -- The output time range for INTERP is specified by the RANGE(timestamp1, timestamp2) field, which must satisfy timestamp1 \<= timestamp2. Here, timestamp1 is the start value of the output time range, i.e., if the conditions for interpolation are met at timestamp1, then timestamp1 is the first record output, and timestamp2 is the end value of the output time range, i.e., the timestamp of the last record output cannot be greater than timestamp2. -- INTERP determines the number of results within the output time range based on the EVERY(time_unit) field, starting from timestamp1 and interpolating at fixed intervals of time (time_unit value), where time_unit can be time units: 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), 1w (weeks). For example, EVERY(500a) will interpolate the specified data every 500 milliseconds. -- INTERP determines how to interpolate at each time point that meets the output conditions based on the FILL field. For how to use the FILL clause, refer to [FILL Clause](../time-series-extensions/) -- INTERP can interpolate at a single time point specified in the RANGE field, in which case the EVERY field can be omitted. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00') FILL(linear). -- When INTERP is applied to a supertable, it will sort all the subtable data under that supertable by primary key column and perform interpolation calculations, and can also be used with PARTITION BY tbname to force the results to a single timeline. -- INTERP can be used with the pseudocolumn _irowts to return the timestamp corresponding to the interpolation point (supported from version 3.0.2.0). -- INTERP can be used with the pseudocolumn _isfilled to display whether the return result is from the original record or generated by the interpolation algorithm (supported from version 3.0.3.0). -- For queries on tables with composite primary keys, if there are data with the same timestamp, only the data with the smallest composite primary key participates in the calculation. -- INTERP query supports NEAR FILL mode, i.e., when FILL is needed, it uses the data closest to the current time point for interpolation. When the timestamps before and after are equally close to the current time slice, FILL the previous row's value. This mode is not supported in stream computing and window queries. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', '2023-01-01 00:10:00') FILL(NEAR).(Supported from version 3.3.4.9). -- INTERP can only use the pseudocolumn `_irowts_origin` when using FILL PREV/NEXT/NEAR modes. `_irowts_origin` is supported from version 3.3.4.9. -- INTERP `RANGE` clause supports the expansion of the time range (supported from version 3.3.4.9), such as `RANGE('2023-01-01 00:00:00', 10s)` means to find data 10s before and after the time point '2023-01-01 00:00:00' for interpolation, FILL PREV/NEXT/NEAR respectively means to look for data forward/backward/around the time point, if there is no data around the time point, then use the value specified by FILL for interpolation, therefore the FILL clause must specify a value at this time. For example: SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', 10s) FILL(PREV, 1). Currently, only the combination of time point and time range is supported, not the combination of time interval and time range, i.e., RANGE('2023-01-01 00:00:00', '2023-02-01 00:00:00', 1h) is not supported. The specified time range rules are similar to EVERY, the unit cannot be year or month, the value cannot be 0, and cannot have quotes. When using this extension, other FILL modes except FILL PREV/NEXT/NEAR are not supported, and the EVERY clause cannot be specified. - ### LAST ```sql @@ -2212,6 +2175,38 @@ ignore_option: { - When there is no composite primary key, if different subtables have data with the same timestamp, a "Duplicate timestamps not allowed" message will be displayed - When using composite primary keys, the timestamp and primary key combinations of different subtables may be the same, which row is used depends on which one is found first, meaning that the results of running diff() multiple times in this situation may vary. +### INTERP + +```sql +INTERP(expr [, ignore_null_values]) + +ignore_null_values: { + 0 + | 1 +} +``` + +**Function Description**: Returns the record value or interpolated value of a specified column at a specified time slice. The ignore_null_values parameter can be 0 or 1, where 1 means to ignore NULL values, default is 0. + +**Return Data Type**: Same as the field type. + +**Applicable Data Types**: Numeric types. + +**Applicable to**: Tables and supertables. + +**Usage Instructions** + +- INTERP is used to obtain the record value of a specified column at a specified time slice. If there is no row data that meets the conditions at that time slice, interpolation will be performed according to the settings of the [FILL](../distinguished/#FILL-Clause) parameter. +- The input data for INTERP is the data of the specified column, which can be filtered through conditional statements (where clause). If no filtering condition is specified, the input is all data. +- INTERP SQL queries need to be used together with the RANGE, EVERY, and FILL keywords; stream computing cannot use RANGE, needs EVERY and FILL keywords together. + +- When INTERP is applied to a supertable, it will sort all the subtable data under that supertable by primary key column and perform interpolation calculations, and can also be used with PARTITION BY tbname to force the results to a single timeline. +- INTERP can be used with the pseudocolumn _irowts to return the timestamp corresponding to the interpolation point (supported from version 3.0.2.0). +- INTERP can be used with the pseudocolumn _isfilled to display whether the return result is from the original record or generated by the interpolation algorithm (supported from version 3.0.3.0). +- For queries on tables with composite primary keys, if there are data with the same timestamp, only the data with the smallest composite primary key participates in the calculation. + +- INTERP can only use the pseudocolumn `_irowts_origin` when using FILL PREV/NEXT/NEAR modes. `_irowts_origin` is supported from version 3.3.4.9. + ### IRATE ```sql diff --git a/docs/zh/14-reference/03-taos-sql/06-select.md b/docs/zh/14-reference/03-taos-sql/06-select.md index 696fe03d41..32a89c8cdb 100644 --- a/docs/zh/14-reference/03-taos-sql/06-select.md +++ b/docs/zh/14-reference/03-taos-sql/06-select.md @@ -277,6 +277,15 @@ TDengine 支持基于时间戳主键的 INNER JOIN，规则如下： 5. JOIN 两侧均支持子查询。 6. 不支持与 FILL 子句混合使用。 +## INTERP +interp 子句配合 [INTERP](../function/#interp) 函数使用, 用于确定 INTERP 函数执行时，时间断面的截取规则和插值规则。 +- INTERP 的输出时间范围根据 RANGE(timestamp1, timestamp2) 字段来指定，需满足 timestamp1 \<= timestamp2。其中 timestamp1 为输出时间范围的起始值，即如果 timestamp1 时刻符合插值条件则 timestamp1 为输出的第一条记录，timestamp2 为输出时间范围的结束值，即输出的最后一条记录的 timestamp 不能大于 timestamp2。 +- INTERP 根据 EVERY(time_unit) 字段来确定输出时间范围内的结果条数，即从 timestamp1 开始每隔固定长度的时间（time_unit 值）进行插值，time_unit 可取值时间单位：1a(毫秒)、1s(秒)、1m(分)、1h(小时)、1d(天)、1w(周)。例如 EVERY(500a) 将对于指定数据每500毫秒间隔进行一次插值。 +- INTERP 根据 FILL 字段来决定在每个符合输出条件的时刻如何进行插值。关于 FILL 子句如何使用请参考 [FILL 子句](./distinguished#fill-子句) +- INTERP 可以在 RANGE 字段中只指定唯一的时间戳对单个时间点进行插值，在这种情况下，EVERY 字段可以省略。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00') FILL(linear)`。 +- INTERP 查询支持 NEAR FILL 模式，即当需要 FILL 时，使用距离当前时间点最近的数据进行插值，当前后时间戳与当前时间断面一样近时，FILL 前一行的值. 此模式在流计算中和窗口查询中不支持。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', '2023-01-01 00:10:00') FILL(NEAR)` (v3.3.4.9 及以后支持)。 +- INTERP `RANGE`子句支持时间范围的扩展(v3.3.4.9 及以后支持)，如`RANGE('2023-01-01 00:00:00', 10s)`表示在时间点 '2023-01-01 00:00:00' 查找前后 10s 的数据进行插值，FILL PREV/NEXT/NEAR 分别表示从时间点向前/向后/前后查找数据，若时间点周围没有数据，则使用 FILL 指定的值进行插值，因此此时 FILL 子句必须指定值。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', 10s) FILL(PREV, 1)`。目前仅支持时间点和时间范围的组合，不支持时间区间和时间范围的组合，即不支持 `RANGE('2023-01-01 00:00:00', '2023-02-01 00:00:00', 1h)`。所指定的时间范围规则与 EVERY 类似，单位不能是年或月，值不能为 0，不能带引号。使用该扩展时，不支持除 `FILL PREV/NEXT/NEAR` 外的其他 FILL 模式，且不能指定 EVERY 子句。 + ## GROUP BY 如果在语句中同时指定了 GROUP BY 子句，那么 SELECT 列表只能包含如下表达式： diff --git a/docs/zh/14-reference/03-taos-sql/10-function.md b/docs/zh/14-reference/03-taos-sql/10-function.md index 882d567c48..9602e7fa5c 100644 --- a/docs/zh/14-reference/03-taos-sql/10-function.md +++ b/docs/zh/14-reference/03-taos-sql/10-function.md @@ -1861,43 +1861,6 @@ FIRST(expr) - 如果结果集中的某列全部为 NULL 值，则该列的返回结果也是 NULL； - 如果结果集中所有列全部为 NULL 值，则不返回结果。 - 对于存在复合主键的表的查询，若最小时间戳的数据有多条，则只有对应的复合主键最小的数据被返回。 - -### INTERP - -```sql -INTERP(expr [, ignore_null_values]) - -ignore_null_values: { - 0 - | 1 -} -``` - -**功能说明**：返回指定时间截面指定列的记录值或插值。ignore_null_values 参数的值可以是 0 或 1，为 1 时表示忽略 NULL 值，缺省值为 0。 - -**返回数据类型**：同字段类型。 - -**适用数据类型**：数值类型。 - -**适用于**：表和超级表。 - -**使用说明** - -- INTERP 用于在指定时间断面获取指定列的记录值，如果该时间断面不存在符合条件的行数据，那么会根据 FILL 参数的设定进行插值。 -- INTERP 的输入数据为指定列的数据，可以通过条件语句（where 子句）来对原始列数据进行过滤，如果没有指定过滤条件则输入为全部数据。 -- INTERP SQL 查询需要同时与 RANGE、EVERY 和 FILL 关键字一起使用；流计算不能使用 RANGE，需要 EVERY 和 FILL 关键字一起使用。 -- INTERP 的输出时间范围根据 RANGE(timestamp1, timestamp2) 字段来指定，需满足 timestamp1 \<= timestamp2。其中 timestamp1 为输出时间范围的起始值，即如果 timestamp1 时刻符合插值条件则 timestamp1 为输出的第一条记录，timestamp2 为输出时间范围的结束值，即输出的最后一条记录的 timestamp 不能大于 timestamp2。 -- INTERP 根据 EVERY(time_unit) 字段来确定输出时间范围内的结果条数，即从 timestamp1 开始每隔固定长度的时间（time_unit 值）进行插值，time_unit 可取值时间单位：1a(毫秒)、1s(秒)、1m(分)、1h(小时)、1d(天)、1w(周)。例如 EVERY(500a) 将对于指定数据每500毫秒间隔进行一次插值。 -- INTERP 根据 FILL 字段来决定在每个符合输出条件的时刻如何进行插值。关于 FILL 子句如何使用请参考 [FILL 子句](../distinguished/#fill-子句) -- INTERP 可以在 RANGE 字段中只指定唯一的时间戳对单个时间点进行插值，在这种情况下，EVERY 字段可以省略。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00') FILL(linear)`。 -- INTERP 作用于超级表时，会将该超级表下的所有子表数据按照主键列排序后进行插值计算，也可以搭配 PARTITION BY tbname 使用，将结果强制规约到单个时间线。 -- INTERP 可以与伪列 _irowts 一起使用，返回插值点所对应的时间戳(v3.0.2.0 以后支持)。 -- INTERP 可以与伪列 _isfilled 一起使用，显示返回结果是否为原始记录或插值算法产生的数据(v3.0.3.0 以后支持)。 -- INTERP 对于带复合主键的表的查询，若存在相同时间戳的数据，则只有对应的复合主键最小的数据参与运算。 -- INTERP 查询支持 NEAR FILL 模式，即当需要 FILL 时，使用距离当前时间点最近的数据进行插值，当前后时间戳与当前时间断面一样近时，FILL 前一行的值. 此模式在流计算中和窗口查询中不支持。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', '2023-01-01 00:10:00') FILL(NEAR)` (v3.3.4.9 及以后支持)。 -- INTERP 只有在使用 FILL PREV/NEXT/NEAR 模式时才可以使用伪列 `_irowts_origin`。`_irowts_origin`在 v3.3.4.9 及以后支持。 -- INTERP `RANGE`子句支持时间范围的扩展(v3.3.4.9 及以后支持)，如`RANGE('2023-01-01 00:00:00', 10s)`表示在时间点 '2023-01-01 00:00:00' 查找前后 10s 的数据进行插值，FILL PREV/NEXT/NEAR 分别表示从时间点向前/向后/前后查找数据，若时间点周围没有数据，则使用 FILL 指定的值进行插值，因此此时 FILL 子句必须指定值。例如 `SELECT INTERP(col) FROM tb RANGE('2023-01-01 00:00:00', 10s) FILL(PREV, 1)`。目前仅支持时间点和时间范围的组合，不支持时间区间和时间范围的组合，即不支持 `RANGE('2023-01-01 00:00:00', '2023-02-01 00:00:00', 1h)`。所指定的时间范围规则与 EVERY 类似，单位不能是年或月，值不能为 0，不能带引号。使用该扩展时，不支持除 `FILL PREV/NEXT/NEAR` 外的其他 FILL 模式，且不能指定 EVERY 子句。 - ### LAST ```sql @@ -2139,6 +2102,35 @@ ignore_option: { - 当没有复合主键时，如果不同的子表有相同时间戳的数据，会提示 "Duplicate timestamps not allowed" - 当使用复合主键时，不同子表的时间戳和主键组合可能相同，使用哪一行取决于先找到哪一行，这意味着在这种情况下多次运行 diff() 的结果可能会不同。 +### INTERP + +```sql +INTERP(expr [, ignore_null_values]) + +ignore_null_values: { + 0 + | 1 +} +``` + +**功能说明**：返回指定时间截面指定列的记录值或插值。ignore_null_values 参数的值可以是 0 或 1，为 1 时表示忽略 NULL 值，缺省值为 0。 + +**返回数据类型**：同字段类型。 + +**适用数据类型**：数值类型。 + +**适用于**：表和超级表。 + +**使用说明** +- INTERP 用于在指定时间断面获取指定列的记录值，如果该时间断面不存在符合条件的行数据，那么会根据 [FILL](../distinguished/#fill-子句) 参数的设定进行插值。 +- INTERP 的输入数据为指定列的数据，可以通过条件语句（where 子句）来对原始列数据进行过滤，如果没有指定过滤条件则输入为全部数据。 +- INTERP SQL 查询需要同时与 RANGE、EVERY 和 FILL 关键字一起使用；流计算不能使用 RANGE，需要 EVERY 和 FILL 关键字一起使用。使用规则见 [INTERP](../select/#interp) 子句。 +- INTERP 作用于超级表时，会将该超级表下的所有子表数据按照主键列排序后进行插值计算，也可以搭配 PARTITION BY tbname 使用，将结果强制规约到单个时间线。 +- INTERP 可以与伪列 `_irowts` 一起使用，返回插值点所对应的时间戳(v3.0.2.0 以后支持)。 +- INTERP 可以与伪列 `_isfilled` 一起使用，显示返回结果是否为原始记录或插值算法产生的数据(v3.0.3.0 以后支持)。 +- INTERP 只有在使用 [FILL](../distinguished/#fill-子句) PREV/NEXT/NEAR 模式时才可以使用伪列 `_irowts_origin`, 用于返回 `interp` 函数所使用的原始数据的时间戳列。若范围内无值, 则返回 NULL。`_irowts_origin` 在 v3.3.4.9 以后支持。 +- INTERP 对于带复合主键的表的查询，若存在相同时间戳的数据，则只有对应的复合主键最小的数据参与运算。 + ### IRATE ```sql From 08a85d05f0e6f772cd549169da6c1e47f69aef07 Mon Sep 17 00:00:00 2001 From: factosea <285808407@qq.com> Date: Fri, 14 Mar 2025 09:40:11 +0800 Subject: [PATCH 2/4] docs: interp ins --- docs/en/14-reference/03-taos-sql/06-select.md | 2 +- docs/en/14-reference/03-taos-sql/10-function.md | 8 +++----- docs/zh/14-reference/03-taos-sql/06-select.md | 3 ++- docs/zh/14-reference/03-taos-sql/10-function.md | 8 ++++---- 4 files changed, 10 insertions(+), 11 deletions(-) diff --git a/docs/en/14-reference/03-taos-sql/06-select.md b/docs/en/14-reference/03-taos-sql/06-select.md index 98559104c0..2c2b0de8fb 100644 --- a/docs/en/14-reference/03-taos-sql/06-select.md +++ b/docs/en/14-reference/03-taos-sql/06-select.md @@ -278,7 +278,7 @@ TDengine supports INNER JOIN based on the timestamp primary key, with the follow 6. Does not support mixing with the FILL clause. ## INTERP -The INTERP clause is used in conjunction with the [INTERP](../function/#interp) function to determine the rules for capturing time slices and interpolation when the INTERP function is executed. +The INTERP clause is a dedicated syntax for the INTERP function (../function/#interp). When an SQL statement contains an INTERP clause, it can only query the INTERP function and cannot be used with other functions. Additionally, the INTERP clause cannot be used simultaneously with window clauses (window_clause) or group by clauses (group_by_clause). The INTERP function must be used with the RANGE, EVERY, and FILL clauses; stream computing does not support the use of RANGE but requires the use of the EVERY and FILL keywords. - The output time range for INTERP is specified by the RANGE(timestamp1, timestamp2) field, which must satisfy timestamp1 \<= timestamp2. Here, timestamp1 is the start value of the output time range, i.e., if the conditions for interpolation are met at timestamp1, then timestamp1 is the first record output, and timestamp2 is the end value of the output time range, i.e., the timestamp of the last record output cannot be greater than timestamp2. - INTERP determines the number of results within the output time range based on the EVERY(time_unit) field, starting from timestamp1 and interpolating at fixed intervals of time (time_unit value), where time_unit can be time units: 1a (milliseconds), 1s (seconds), 1m (minutes), 1h (hours), 1d (days), 1w (weeks). For example, EVERY(500a) will interpolate the specified data every 500 milliseconds. - INTERP determines how to interpolate at each time point that meets the output conditions based on the FILL field. For how to use the FILL clause, refer to [FILL Clause](../time-series-extensions/) diff --git a/docs/en/14-reference/03-taos-sql/10-function.md b/docs/en/14-reference/03-taos-sql/10-function.md index 27b5d2a577..cf143ce759 100644 --- a/docs/en/14-reference/03-taos-sql/10-function.md +++ b/docs/en/14-reference/03-taos-sql/10-function.md @@ -2196,16 +2196,14 @@ ignore_null_values: { **Usage Instructions** -- INTERP is used to obtain the record value of a specified column at a specified time slice. If there is no row data that meets the conditions at that time slice, interpolation will be performed according to the settings of the [FILL](../distinguished/#FILL-Clause) parameter. +- INTERP is used to obtain the record value of a specified column at the specified time slice. It has a dedicated syntax (interp_clause) when used. For syntax introduction, see [reference link](../select/#interp). +- When there is no row data that meets the conditions at the specified time slice, the INTERP function will interpolate according to the settings of the [FILL](../distinguished/#FILL-Clause) parameter. - The input data for INTERP is the data of the specified column, which can be filtered through conditional statements (where clause). If no filtering condition is specified, the input is all data. -- INTERP SQL queries need to be used together with the RANGE, EVERY, and FILL keywords; stream computing cannot use RANGE, needs EVERY and FILL keywords together. - - When INTERP is applied to a supertable, it will sort all the subtable data under that supertable by primary key column and perform interpolation calculations, and can also be used with PARTITION BY tbname to force the results to a single timeline. - INTERP can be used with the pseudocolumn _irowts to return the timestamp corresponding to the interpolation point (supported from version 3.0.2.0). - INTERP can be used with the pseudocolumn _isfilled to display whether the return result is from the original record or generated by the interpolation algorithm (supported from version 3.0.3.0). -- For queries on tables with composite primary keys, if there are data with the same timestamp, only the data with the smallest composite primary key participates in the calculation. - - INTERP can only use the pseudocolumn `_irowts_origin` when using FILL PREV/NEXT/NEAR modes. `_irowts_origin` is supported from version 3.3.4.9. +- For queries on tables with composite primary keys, if there are data with the same timestamp, only the data with the smallest composite primary key participates in the calculation. ### IRATE diff --git a/docs/zh/14-reference/03-taos-sql/06-select.md b/docs/zh/14-reference/03-taos-sql/06-select.md index 32a89c8cdb..189faf8884 100644 --- a/docs/zh/14-reference/03-taos-sql/06-select.md +++ b/docs/zh/14-reference/03-taos-sql/06-select.md @@ -278,7 +278,8 @@ TDengine 支持基于时间戳主键的 INNER JOIN，规则如下： 6. 不支持与 FILL 子句混合使用。 ## INTERP -interp 子句配合 [INTERP](../function/#interp) 函数使用, 用于确定 INTERP 函数执行时，时间断面的截取规则和插值规则。 + +interp 子句是 INTERP 函数(../function/#interp)的专用语法，当 SQL 语句中存在 interp 子句时，只能查询 INTERP 函数而不能与其他函数一起查询，同时 interp 子句与窗口子句(window_clause)、分组子句(group_by_clause)也不能同时使用。INTERP 函数在使用时需要与 RANGE、EVERY 和 FILL 子句一起使用；流计算不支持使用 RANGE，但需要与 EVERY 和 FILL 关键字一起使用。 - INTERP 的输出时间范围根据 RANGE(timestamp1, timestamp2) 字段来指定，需满足 timestamp1 \<= timestamp2。其中 timestamp1 为输出时间范围的起始值，即如果 timestamp1 时刻符合插值条件则 timestamp1 为输出的第一条记录，timestamp2 为输出时间范围的结束值，即输出的最后一条记录的 timestamp 不能大于 timestamp2。 - INTERP 根据 EVERY(time_unit) 字段来确定输出时间范围内的结果条数，即从 timestamp1 开始每隔固定长度的时间（time_unit 值）进行插值，time_unit 可取值时间单位：1a(毫秒)、1s(秒)、1m(分)、1h(小时)、1d(天)、1w(周)。例如 EVERY(500a) 将对于指定数据每500毫秒间隔进行一次插值。 - INTERP 根据 FILL 字段来决定在每个符合输出条件的时刻如何进行插值。关于 FILL 子句如何使用请参考 [FILL 子句](./distinguished#fill-子句) diff --git a/docs/zh/14-reference/03-taos-sql/10-function.md b/docs/zh/14-reference/03-taos-sql/10-function.md index 9602e7fa5c..8271a91721 100644 --- a/docs/zh/14-reference/03-taos-sql/10-function.md +++ b/docs/zh/14-reference/03-taos-sql/10-function.md @@ -2122,14 +2122,14 @@ ignore_null_values: { **适用于**：表和超级表。 **使用说明** -- INTERP 用于在指定时间断面获取指定列的记录值，如果该时间断面不存在符合条件的行数据，那么会根据 [FILL](../distinguished/#fill-子句) 参数的设定进行插值。 +- INTERP 用于在指定时间断面获取指定列的记录值，使用时有专用语法(interp_clause)，语法介绍[参考链接](../select/#interp) 。 +- 当指定时间断面不存在符合条件的行数据时，INTERP 函数会根据 [FILL](../distinguished/#fill-子句) 参数的设定进行插值。 - INTERP 的输入数据为指定列的数据，可以通过条件语句（where 子句）来对原始列数据进行过滤，如果没有指定过滤条件则输入为全部数据。 -- INTERP SQL 查询需要同时与 RANGE、EVERY 和 FILL 关键字一起使用；流计算不能使用 RANGE，需要 EVERY 和 FILL 关键字一起使用。使用规则见 [INTERP](../select/#interp) 子句。 - INTERP 作用于超级表时，会将该超级表下的所有子表数据按照主键列排序后进行插值计算，也可以搭配 PARTITION BY tbname 使用，将结果强制规约到单个时间线。 - INTERP 可以与伪列 `_irowts` 一起使用，返回插值点所对应的时间戳(v3.0.2.0 以后支持)。 - INTERP 可以与伪列 `_isfilled` 一起使用，显示返回结果是否为原始记录或插值算法产生的数据(v3.0.3.0 以后支持)。 -- INTERP 只有在使用 [FILL](../distinguished/#fill-子句) PREV/NEXT/NEAR 模式时才可以使用伪列 `_irowts_origin`, 用于返回 `interp` 函数所使用的原始数据的时间戳列。若范围内无值, 则返回 NULL。`_irowts_origin` 在 v3.3.4.9 以后支持。 -- INTERP 对于带复合主键的表的查询，若存在相同时间戳的数据，则只有对应的复合主键最小的数据参与运算。 +- 只有在使用 FILL PREV/NEXT/NEAR 模式时才可以使用伪列 `_irowts_origin`, 用于返回 `interp` 函数所使用的原始数据的时间戳列。若范围内无值, 则返回 NULL。`_irowts_origin` 在 v3.3.4.9 以后支持。 +- 对于带复合主键的表的查询，若存在相同时间戳的数据，则只有对应的复合主键最小的数据参与运算。 ### IRATE From a19c4993dd070409a8f5a4d51930f6e413efa6f4 Mon Sep 17 00:00:00 2001 From: Pan Wei <72057773+dapan1121@users.noreply.github.com> Date: Fri, 14 Mar 2025 11:12:24 +0800 Subject: [PATCH 3/4] Update 10-function.md --- docs/zh/14-reference/03-taos-sql/10-function.md | 1 - 1 file changed, 1 deletion(-) diff --git a/docs/zh/14-reference/03-taos-sql/10-function.md b/docs/zh/14-reference/03-taos-sql/10-function.md index 8271a91721..2ac299b0b1 100644 --- a/docs/zh/14-reference/03-taos-sql/10-function.md +++ b/docs/zh/14-reference/03-taos-sql/10-function.md @@ -2124,7 +2124,6 @@ ignore_null_values: { **使用说明** - INTERP 用于在指定时间断面获取指定列的记录值，使用时有专用语法(interp_clause)，语法介绍[参考链接](../select/#interp) 。 - 当指定时间断面不存在符合条件的行数据时，INTERP 函数会根据 [FILL](../distinguished/#fill-子句) 参数的设定进行插值。 -- INTERP 的输入数据为指定列的数据，可以通过条件语句（where 子句）来对原始列数据进行过滤，如果没有指定过滤条件则输入为全部数据。 - INTERP 作用于超级表时，会将该超级表下的所有子表数据按照主键列排序后进行插值计算，也可以搭配 PARTITION BY tbname 使用，将结果强制规约到单个时间线。 - INTERP 可以与伪列 `_irowts` 一起使用，返回插值点所对应的时间戳(v3.0.2.0 以后支持)。 - INTERP 可以与伪列 `_isfilled` 一起使用，显示返回结果是否为原始记录或插值算法产生的数据(v3.0.3.0 以后支持)。 From b4a64236dbd4c929eb0b2e4c0081d566bc9a6869 Mon Sep 17 00:00:00 2001 From: Pan Wei <72057773+dapan1121@users.noreply.github.com> Date: Fri, 14 Mar 2025 11:14:45 +0800 Subject: [PATCH 4/4] Update 10-function.md --- docs/en/14-reference/03-taos-sql/10-function.md | 1 - 1 file changed, 1 deletion(-) diff --git a/docs/en/14-reference/03-taos-sql/10-function.md b/docs/en/14-reference/03-taos-sql/10-function.md index cf143ce759..4e60f3284f 100644 --- a/docs/en/14-reference/03-taos-sql/10-function.md +++ b/docs/en/14-reference/03-taos-sql/10-function.md @@ -2198,7 +2198,6 @@ ignore_null_values: { - INTERP is used to obtain the record value of a specified column at the specified time slice. It has a dedicated syntax (interp_clause) when used. For syntax introduction, see [reference link](../select/#interp). - When there is no row data that meets the conditions at the specified time slice, the INTERP function will interpolate according to the settings of the [FILL](../distinguished/#FILL-Clause) parameter. -- The input data for INTERP is the data of the specified column, which can be filtered through conditional statements (where clause). If no filtering condition is specified, the input is all data. - When INTERP is applied to a supertable, it will sort all the subtable data under that supertable by primary key column and perform interpolation calculations, and can also be used with PARTITION BY tbname to force the results to a single timeline. - INTERP can be used with the pseudocolumn _irowts to return the timestamp corresponding to the interpolation point (supported from version 3.0.2.0). - INTERP can be used with the pseudocolumn _isfilled to display whether the return result is from the original record or generated by the interpolation algorithm (supported from version 3.0.3.0).