From 0a6ec7c0d763e288118e6e88352c7d9115244ac0 Mon Sep 17 00:00:00 2001 From: Ganlin Zhao Date: Wed, 14 Sep 2022 11:17:12 +0800 Subject: [PATCH 1/3] fix(query): restrict max/min function input type not including timestamp type --- source/libs/function/src/builtins.c | 20 ++------------------ source/libs/function/src/builtinsimpl.c | 6 +++--- 2 files changed, 5 insertions(+), 21 deletions(-) diff --git a/source/libs/function/src/builtins.c b/source/libs/function/src/builtins.c index 5844784ea4..f3d3393ac3 100644 --- a/source/libs/function/src/builtins.c +++ b/source/libs/function/src/builtins.c @@ -311,22 +311,6 @@ static int32_t translateInOutStr(SFunctionNode* pFunc, char* pErrBuf, int32_t le return TSDB_CODE_SUCCESS; } -static int32_t translateMinMax(SFunctionNode* pFunc, char* pErrBuf, int32_t len) { - if (1 != LIST_LENGTH(pFunc->pParameterList)) { - return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName); - } - - uint8_t paraType = ((SExprNode*)nodesListGetNode(pFunc->pParameterList, 0))->resType.type; - if (!IS_TIMESTAMP_TYPE(paraType) && !IS_NUMERIC_TYPE(paraType) && !IS_NULL_TYPE(paraType)) { - return invaildFuncParaTypeErrMsg(pErrBuf, len, pFunc->functionName); - } else if (IS_NULL_TYPE(paraType)) { - paraType = TSDB_DATA_TYPE_BIGINT; - } - - pFunc->node.resType = (SDataType){.bytes = tDataTypes[paraType].bytes, .type = paraType}; - return TSDB_CODE_SUCCESS; -} - static int32_t translateTrimStr(SFunctionNode* pFunc, char* pErrBuf, int32_t len, bool isLtrim) { if (1 != LIST_LENGTH(pFunc->pParameterList)) { return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName); @@ -2076,7 +2060,7 @@ const SBuiltinFuncDefinition funcMgtBuiltins[] = { .name = "min", .type = FUNCTION_TYPE_MIN, .classification = FUNC_MGT_AGG_FUNC | FUNC_MGT_SPECIAL_DATA_REQUIRED | FUNC_MGT_SELECT_FUNC, - .translateFunc = translateMinMax, + .translateFunc = translateInOutNum, .dataRequiredFunc = statisDataRequired, .getEnvFunc = getMinmaxFuncEnv, .initFunc = minmaxFunctionSetup, @@ -2091,7 +2075,7 @@ const SBuiltinFuncDefinition funcMgtBuiltins[] = { .name = "max", .type = FUNCTION_TYPE_MAX, .classification = FUNC_MGT_AGG_FUNC | FUNC_MGT_SPECIAL_DATA_REQUIRED | FUNC_MGT_SELECT_FUNC, - .translateFunc = translateMinMax, + .translateFunc = translateInOutNum, .dataRequiredFunc = statisDataRequired, .getEnvFunc = getMinmaxFuncEnv, .initFunc = minmaxFunctionSetup, diff --git a/source/libs/function/src/builtinsimpl.c b/source/libs/function/src/builtinsimpl.c index 0d7fd1a6da..c9e345f49f 100644 --- a/source/libs/function/src/builtinsimpl.c +++ b/source/libs/function/src/builtinsimpl.c @@ -1204,7 +1204,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { pBuf->tuplePos = saveTupleData(pCtx, index, pCtx->pSrcBlock); } } else { - if (IS_SIGNED_NUMERIC_TYPE(type) || IS_TIMESTAMP_TYPE(type)) { + if (IS_SIGNED_NUMERIC_TYPE(type)) { int64_t prev = 0; GET_TYPED_DATA(prev, int64_t, type, &pBuf->v); @@ -1263,7 +1263,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { int32_t start = pInput->startRowIndex; int32_t numOfRows = pInput->numOfRows; - if (IS_SIGNED_NUMERIC_TYPE(type) || IS_TIMESTAMP_TYPE(type) || type == TSDB_DATA_TYPE_BOOL) { + if (IS_SIGNED_NUMERIC_TYPE(type) || type == TSDB_DATA_TYPE_BOOL) { if (type == TSDB_DATA_TYPE_TINYINT || type == TSDB_DATA_TYPE_BOOL) { int8_t* pData = (int8_t*)pCol->pData; int8_t* val = (int8_t*)&pBuf->v; @@ -1357,7 +1357,7 @@ int32_t doMinMaxHelper(SqlFunctionCtx* pCtx, int32_t isMinFunc) { numOfElems += 1; } - } else if (type == TSDB_DATA_TYPE_BIGINT || type == TSDB_DATA_TYPE_TIMESTAMP) { + } else if (type == TSDB_DATA_TYPE_BIGINT) { int64_t* pData = (int64_t*)pCol->pData; int64_t* val = (int64_t*)&pBuf->v; From 8905d37b7065ffc0832ae96d9141e55e55ff76b8 Mon Sep 17 00:00:00 2001 From: Ganlin Zhao Date: Wed, 14 Sep 2022 13:35:46 +0800 Subject: [PATCH 2/3] fix test cases --- tests/system-test/2-query/max.py | 21 ++------------- tests/system-test/2-query/min.py | 45 +++++--------------------------- 2 files changed, 8 insertions(+), 58 deletions(-) diff --git a/tests/system-test/2-query/max.py b/tests/system-test/2-query/max.py index 169b1c2c38..5cc9a2d9e8 100644 --- a/tests/system-test/2-query/max.py +++ b/tests/system-test/2-query/max.py @@ -38,18 +38,7 @@ class TDTestCase: elif i>=9: tdSql.checkData(0, 0, np.max(floatData)) - tdSql.query(f"select max(now()) from {dbname}.stb_1") - tdSql.checkRows(1) - - tdSql.query(f"select last(ts) from {dbname}.stb_1") - lastTs = tdSql.getData(0, 0) - tdSql.query(f"select max(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, lastTs) - - tdSql.query(f"select last(ts) from {dbname}.stb") - lastTs = tdSql.getData(0, 0) - tdSql.query(f"select max(ts) from {dbname}.stb") - tdSql.checkData(0, 0, lastTs) + tdSql.error(f"select max(now()) from {dbname}.stb_1") tdSql.query(f"select max(col1) from {dbname}.stb_1 where col2<=5") tdSql.checkData(0,0,5) @@ -78,13 +67,7 @@ class TDTestCase: elif i>=9: tdSql.checkData(0, 0, np.max(floatData)) - tdSql.query(f"select max(now()) from {dbname}.ntb") - tdSql.checkRows(1) - - tdSql.query(f"select last(ts) from {dbname}.ntb") - lastTs = tdSql.getData(0, 0) - tdSql.query(f"select max(ts) from {dbname}.ntb") - tdSql.checkData(0, 0, lastTs) + tdSql.error(f"select max(now()) from {dbname}.ntb") tdSql.query(f"select max(col1) from {dbname}.ntb where col2<=5") tdSql.checkData(0,0,5) diff --git a/tests/system-test/2-query/min.py b/tests/system-test/2-query/min.py index 3d46b7b222..d97c4340f4 100644 --- a/tests/system-test/2-query/min.py +++ b/tests/system-test/2-query/min.py @@ -37,6 +37,8 @@ class TDTestCase: floatData.append(i + 0.1) # max verifacation + tdSql.error(f"select min(now()) from {dbname}.stb_1") + tdSql.error(f"select min(ts) from {dbname}.stb_1") tdSql.error(f"select min(col7) from {dbname}.stb_1") tdSql.error(f"select min(col8) from {dbname}.stb_1") tdSql.error(f"select min(col9) from {dbname}.stb_1") @@ -67,20 +69,9 @@ class TDTestCase: tdSql.query(f"select min(col1) from {dbname}.stb_1 where col2>=5") tdSql.checkData(0,0,5) - tdSql.query(f"select min(now()) from {dbname}.stb_1") - tdSql.checkRows(1) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - + tdSql.error(f"select min(now()) from {dbname}.stb_1") + tdSql.error(f"select min(ts) from {dbname}.stb_1") tdSql.error(f"select min(col7) from {dbname}.stb_1") tdSql.error(f"select min(col8) from {dbname}.stb_1") tdSql.error(f"select min(col9) from {dbname}.stb_1") @@ -111,19 +102,8 @@ class TDTestCase: tdSql.query(f"select min(col1) from {dbname}.stb where col2>=5") tdSql.checkData(0,0,5) - tdSql.query(f"select min(now()) from {dbname}.stb_1") - tdSql.checkRows(1) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - + tdSql.error(f"select min(now()) from {dbname}.stb_1") + tdSql.error(f"select min(ts) from {dbname}.stb_1") tdSql.error(f"select min(col7) from {dbname}.ntb") tdSql.error(f"select min(col8) from {dbname}.ntb") tdSql.error(f"select min(col9) from {dbname}.ntb") @@ -154,19 +134,6 @@ class TDTestCase: tdSql.query(f"select min(col1) from {dbname}.ntb where col2>=5") tdSql.checkData(0,0,5) - tdSql.query(f"select min(now()) from {dbname}.stb_1") - tdSql.checkRows(1) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - - tdSql.query(f"select first(ts) from {dbname}.stb_1") - firstTs = tdSql.getData(0, 0) - tdSql.query(f"select min(ts) from {dbname}.stb_1") - tdSql.checkData(0, 0, firstTs) - def stop(self): tdSql.close() From d2f6a7928686556ea44d9c52f2b59aaeeac25f59 Mon Sep 17 00:00:00 2001 From: Ganlin Zhao Date: Wed, 14 Sep 2022 13:38:39 +0800 Subject: [PATCH 3/3] fix docs --- docs/en/12-taos-sql/10-function.md | 44 +++++++++++++++--------------- docs/zh/12-taos-sql/10-function.md | 38 +++++++++++++------------- 2 files changed, 41 insertions(+), 41 deletions(-) diff --git a/docs/en/12-taos-sql/10-function.md b/docs/en/12-taos-sql/10-function.md index f74d0dbe5c..ab1d2f900b 100644 --- a/docs/en/12-taos-sql/10-function.md +++ b/docs/en/12-taos-sql/10-function.md @@ -126,7 +126,7 @@ SELECT COS(field_name) FROM { tb_name | stb_name } [WHERE clause] SELECT FLOOR(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**Description**: The rounded down value of a specific field +**Description**: The rounded down value of a specific field **More explanations**: The restrictions are same as those of the `CEIL` function. #### LOG @@ -173,7 +173,7 @@ SELECT POW(field_name, power) FROM { tb_name | stb_name } [WHERE clause] 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. @@ -434,7 +434,7 @@ SELECT TO_ISO8601(ts[, timezone]) FROM { tb_name | stb_name } [WHERE clause]; **More explanations**: - 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 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 @@ -769,14 +769,14 @@ SELECT HISTOGRAM(field_name，bin_type, bin_description, normalized) FROM tb_nam **Explanations**： - bin_type: parameter to indicate the bucket type, valid inputs are: "user_input", "linear_bin", "log_bin"。 -- 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]": +- 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]. @@ -862,9 +862,9 @@ SELECT INTERP(field_name) FROM { tb_name | stb_name } [WHERE where_condition] RA - `INTERP` is used to get the value that matches the specified time slice from a column. If no such value exists an interpolation value will be returned based on `FILL` parameter. - The input data of `INTERP` is the value of the specified column and a `where` clause can be used to filter the original data. If no `where` condition is specified then all original data is the input. -- The output time range of `INTERP` is specified by `RANGE(timestamp1,timestamp2)` parameter, with timestamp1<=timestamp2. timestamp1 is the starting point of the output time range and must be specified. timestamp2 is the ending point of the output time range and must be specified. -- The number of rows in the result set of `INTERP` is determined by the parameter `EVERY`. Starting from timestamp1, one interpolation is performed for every time interval specified `EVERY` parameter. -- Interpolation is performed based on `FILL` parameter. +- The output time range of `INTERP` is specified by `RANGE(timestamp1,timestamp2)` parameter, with timestamp1<=timestamp2. timestamp1 is the starting point of the output time range and must be specified. timestamp2 is the ending point of the output time range and must be specified. +- The number of rows in the result set of `INTERP` is determined by the parameter `EVERY`. Starting from timestamp1, one interpolation is performed for every time interval specified `EVERY` parameter. +- Interpolation is performed based on `FILL` parameter. - `INTERP` can only be used to interpolate in single timeline. So it must be used with `partition by tbname` when it's used on a STable. ### LAST @@ -917,7 +917,7 @@ SELECT MAX(field_name) FROM { tb_name | stb_name } [WHERE clause]; **Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: Numeric, Timestamp +**Applicable data types**: Numeric **Applicable table types**: standard tables and supertables @@ -932,7 +932,7 @@ SELECT MIN(field_name) FROM {tb_name | stb_name} [WHERE clause]; **Return value type**:Same as the data type of the column being operated upon -**Applicable data types**: Numeric, Timestamp +**Applicable data types**: Numeric **Applicable table types**: standard tables and supertables @@ -968,7 +968,7 @@ SELECT SAMPLE(field_name, K) FROM { tb_name | stb_name } [WHERE clause] **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 @@ -1046,10 +1046,10 @@ SELECT CSUM(field_name) FROM { tb_name | stb_name } [WHERE clause] **Applicable table types**: standard tables and supertables -**More explanations**: - +**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. +- 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 @@ -1067,8 +1067,8 @@ SELECT DERIVATIVE(field_name, time_interval, ignore_negative) FROM tb_name [WHER **Applicable table types**: standard tables and supertables -**More explanation**: - +**More explanation**: + - It can be used together with `PARTITION BY tbname` against a STable. - It can be used together with a selected column. For example: select \_rowts, DERIVATIVE() from。 @@ -1086,7 +1086,7 @@ SELECT {DIFF(field_name, ignore_negative) | DIFF(field_name)} FROM tb_name [WHER **Applicable table types**: standard tables and supertables -**More explanation**: +**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。 @@ -1123,9 +1123,9 @@ SELECT MAVG(field_name, K) FROM { tb_name | stb_name } [WHERE clause] **Applicable table types**: standard tables and supertables -**More explanations**: - -- Arithmetic operation can't be performed on the result of `MAVG`. +**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 diff --git a/docs/zh/12-taos-sql/10-function.md b/docs/zh/12-taos-sql/10-function.md index 9c5b7f771e..86e9aaa80f 100644 --- a/docs/zh/12-taos-sql/10-function.md +++ b/docs/zh/12-taos-sql/10-function.md @@ -127,7 +127,7 @@ SELECT COS(field_name) FROM { tb_name | stb_name } [WHERE clause] SELECT FLOOR(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**功能说明**：获得指定字段的向下取整数的结果。 +**功能说明**：获得指定字段的向下取整数的结果。 其他使用说明参见 CEIL 函数描述。 #### LOG @@ -174,7 +174,7 @@ SELECT POW(field_name, power) FROM { tb_name | stb_name } [WHERE clause] SELECT ROUND(field_name) FROM { tb_name | stb_name } [WHERE clause]; ``` -**功能说明**：获得指定字段的四舍五入的结果。 +**功能说明**：获得指定字段的四舍五入的结果。 其他使用说明参见 CEIL 函数描述。 @@ -435,7 +435,7 @@ SELECT TO_ISO8601(ts[, timezone]) FROM { tb_name | stb_name } [WHERE clause]; **使用说明**： - timezone 参数允许输入的时区格式为: [z/Z, +/-hhmm, +/-hh, +/-hh:mm]。例如，TO_ISO8601(1, "+00:00")。 -- 如果输入是表示 UNIX 时间戳的整形，返回格式精度由时间戳的位数决定; +- 如果输入是表示 UNIX 时间戳的整形，返回格式精度由时间戳的位数决定; - 如果输入是 TIMESTAMP 类型的列，返回格式的时间戳精度与当前 DATABASE 设置的时间精度一致。 @@ -770,14 +770,14 @@ SELECT HISTOGRAM(field_name，bin_type, bin_description, normalized) FROM tb_nam **详细说明**： - bin_type 用户指定的分桶类型, 有效输入类型为"user_input“, ”linear_bin", "log_bin"。 -- bin_description 描述如何生成分桶区间，针对三种桶类型，分别为以下描述格式(均为 JSON 格式字符串)： - - "user_input": "[1, 3, 5, 7]" +- bin_description 描述如何生成分桶区间，针对三种桶类型，分别为以下描述格式(均为 JSON 格式字符串)： + - "user_input": "[1, 3, 5, 7]" 用户指定 bin 的具体数值。 - + - "linear_bin": "{"start": 0.0, "width": 5.0, "count": 5, "infinity": true}" "start" 表示数据起始点，"width" 表示每次 bin 偏移量, "count" 为 bin 的总数，"infinity" 表示是否添加（-inf, inf）作为区间起点和终点， 生成区间为[-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" 表示数据起始点，"factor" 表示按指数递增的因子，"count" 为 bin 的总数，"infinity" 表示是否添加（-inf, inf）作为区间起点和终点， 生成区间为[-inf, 1.0, 2.0, 4.0, 8.0, 16.0, +inf]。 @@ -918,7 +918,7 @@ SELECT MAX(field_name) FROM { tb_name | stb_name } [WHERE clause]; **返回数据类型**：同应用的字段。 -**适用数据类型**：数值类型，时间戳类型。 +**适用数据类型**：数值类型。 **适用于**：表和超级表。 @@ -933,7 +933,7 @@ SELECT MIN(field_name) FROM {tb_name | stb_name} [WHERE clause]; **返回数据类型**：同应用的字段。 -**适用数据类型**：数值类型，时间戳类型。 +**适用数据类型**：数值类型。 **适用于**：表和超级表。 @@ -969,7 +969,7 @@ SELECT SAMPLE(field_name, K) FROM { tb_name | stb_name } [WHERE clause] **适用于**：表和超级表。 -**使用说明**： +**使用说明**： - 不能参与表达式计算；该函数可以应用在普通表和超级表上； - 使用在超级表上的时候，需要搭配 PARTITION by tbname 使用，将结果强制规约到单个时间线。 @@ -1047,10 +1047,10 @@ SELECT CSUM(field_name) FROM { tb_name | stb_name } [WHERE clause] **适用于**：表和超级表。 -**使用说明**： - +**使用说明**： + - 不支持 +、-、*、/ 运算，如 csum(col1) + csum(col2)。 -- 只能与聚合（Aggregation）函数一起使用。 该函数可以应用在普通表和超级表上。 +- 只能与聚合（Aggregation）函数一起使用。 该函数可以应用在普通表和超级表上。 - 使用在超级表上的时候，需要搭配 PARTITION BY tbname使用，将结果强制规约到单个时间线。 @@ -1068,8 +1068,8 @@ SELECT DERIVATIVE(field_name, time_interval, ignore_negative) FROM tb_name [WHER **适用于**：表和超级表。 -**使用说明**: - +**使用说明**: + - DERIVATIVE 函数可以在由 PARTITION BY 划分出单独时间线的情况下用于超级表（也即 PARTITION BY tbname）。 - 可以与选择相关联的列一起使用。 例如: select \_rowts, DERIVATIVE() from。 @@ -1087,7 +1087,7 @@ SELECT {DIFF(field_name, ignore_negative) | DIFF(field_name)} FROM tb_name [WHER **适用于**：表和超级表。 -**使用说明**: +**使用说明**: - 输出结果行数是范围内总行数减一，第一行没有结果输出。 - 可以与选择相关联的列一起使用。 例如: select \_rowts, DIFF() from。 @@ -1124,9 +1124,9 @@ SELECT MAVG(field_name, K) FROM { tb_name | stb_name } [WHERE clause] **适用于**：表和超级表。 -**使用说明**： - -- 不支持 +、-、*、/ 运算，如 mavg(col1, k1) + mavg(col2, k1); +**使用说明**： + +- 不支持 +、-、*、/ 运算，如 mavg(col1, k1) + mavg(col2, k1); - 只能与普通列，选择（Selection）、投影（Projection）函数一起使用，不能与聚合（Aggregation）函数一起使用； - 使用在超级表上的时候，需要搭配 PARTITION BY tbname使用，将结果强制规约到单个时间线。