TDengine/source/libs/executor/src/projectoperator.c

1206 lines
40 KiB
C
Raw Normal View History

2022-07-25 06:15:49 +00:00
/*
* Copyright (c) 2019 TAOS Data, Inc. <jhtao@taosdata.com>
*
* 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 <http://www.gnu.org/licenses/>.
*/
2023-04-28 03:42:34 +00:00
#include "executorInt.h"
2022-12-07 09:54:09 +00:00
#include "filter.h"
2022-07-25 06:15:49 +00:00
#include "functionMgt.h"
2023-04-27 16:23:38 +00:00
#include "operator.h"
#include "querytask.h"
#include "tdatablock.h"
2022-07-25 06:15:49 +00:00
2022-11-22 14:42:44 +00:00
typedef struct SProjectOperatorInfo {
SOptrBasicInfo binfo;
SAggSupporter aggSup;
SArray* pPseudoColInfo;
SLimitInfo limitInfo;
bool mergeDataBlocks;
SSDataBlock* pFinalRes;
2024-01-16 00:13:47 +00:00
bool inputIgnoreGroup;
2024-03-12 03:45:54 +00:00
bool outputIgnoreGroup;
2022-11-22 14:42:44 +00:00
} SProjectOperatorInfo;
typedef struct SIndefOperatorInfo {
SOptrBasicInfo binfo;
SAggSupporter aggSup;
SArray* pPseudoColInfo;
SExprSupp scalarSup;
uint64_t groupId;
SSDataBlock* pNextGroupRes;
} SIndefOperatorInfo;
static int32_t doGenerateSourceData(SOperatorInfo* pOperator);
2024-07-24 09:08:08 +00:00
static SSDataBlock* doProjectOperation1(SOperatorInfo* pOperator);
static int32_t doProjectOperation(SOperatorInfo* pOperator, SSDataBlock** pResBlock);
static SSDataBlock* doApplyIndefinitFunction1(SOperatorInfo* pOperator);
static int32_t doApplyIndefinitFunction(SOperatorInfo* pOperator, SSDataBlock** pResBlock);
static int32_t setRowTsColumnOutputInfo(SqlFunctionCtx* pCtx, int32_t numOfCols, SArray** pResList);
static int32_t setFunctionResultOutput(SOperatorInfo* pOperator, SOptrBasicInfo* pInfo, SAggSupporter* pSup,
int32_t stage, int32_t numOfExprs);
2022-07-25 06:15:49 +00:00
static void destroyProjectOperatorInfo(void* param) {
2022-07-25 06:15:49 +00:00
if (NULL == param) {
return;
}
SProjectOperatorInfo* pInfo = (SProjectOperatorInfo*)param;
cleanupBasicInfo(&pInfo->binfo);
cleanupAggSup(&pInfo->aggSup);
taosArrayDestroy(pInfo->pPseudoColInfo);
blockDataDestroy(pInfo->pFinalRes);
taosMemoryFreeClear(param);
}
static void destroyIndefinitOperatorInfo(void* param) {
2022-07-25 06:15:49 +00:00
SIndefOperatorInfo* pInfo = (SIndefOperatorInfo*)param;
if (pInfo == NULL) {
return;
}
2022-07-25 06:15:49 +00:00
cleanupBasicInfo(&pInfo->binfo);
2022-07-25 06:15:49 +00:00
taosArrayDestroy(pInfo->pPseudoColInfo);
cleanupAggSup(&pInfo->aggSup);
cleanupExprSupp(&pInfo->scalarSup);
taosMemoryFreeClear(param);
}
2023-06-15 06:13:17 +00:00
void streamOperatorReleaseState(SOperatorInfo* pOperator) {
SOperatorInfo* downstream = pOperator->pDownstream[0];
if (downstream->fpSet.releaseStreamStateFn) {
downstream->fpSet.releaseStreamStateFn(downstream);
}
}
void streamOperatorReloadState(SOperatorInfo* pOperator) {
SOperatorInfo* downstream = pOperator->pDownstream[0];
if (downstream->fpSet.reloadStreamStateFn) {
downstream->fpSet.reloadStreamStateFn(downstream);
}
}
2024-07-24 09:08:08 +00:00
int32_t createProjectOperatorInfo(SOperatorInfo* downstream, SProjectPhysiNode* pProjPhyNode, SExecTaskInfo* pTaskInfo,
SOperatorInfo** pOptrInfo) {
QRY_OPTR_CHECK(pOptrInfo);
int32_t code = TSDB_CODE_SUCCESS;
2022-07-25 06:15:49 +00:00
SProjectOperatorInfo* pInfo = taosMemoryCalloc(1, sizeof(SProjectOperatorInfo));
SOperatorInfo* pOperator = taosMemoryCalloc(1, sizeof(SOperatorInfo));
if (pInfo == NULL || pOperator == NULL) {
code = TSDB_CODE_OUT_OF_MEMORY;
2022-07-25 06:15:49 +00:00
goto _error;
}
pOperator->exprSupp.hasWindowOrGroup = false;
2022-08-03 12:45:21 +00:00
pOperator->pTaskInfo = pTaskInfo;
2024-07-24 09:08:08 +00:00
int32_t lino = 0;
2022-07-25 06:15:49 +00:00
2022-11-27 16:27:49 +00:00
SSDataBlock* pResBlock = createDataBlockFromDescNode(pProjPhyNode->node.pOutputDataBlockDesc);
2024-08-05 03:57:18 +00:00
TSDB_CHECK_NULL(pResBlock, code, lino, _error, terrno);
2022-07-25 06:15:49 +00:00
initLimitInfo(pProjPhyNode->node.pLimit, pProjPhyNode->node.pSlimit, &pInfo->limitInfo);
pInfo->binfo.pRes = pResBlock;
2024-07-27 10:55:34 +00:00
pInfo->pFinalRes = NULL;
code = createOneDataBlock(pResBlock, false, &pInfo->pFinalRes);
TSDB_CHECK_CODE(code, lino, _error);
pInfo->binfo.inputTsOrder = pProjPhyNode->node.inputTsOrder;
pInfo->binfo.outputTsOrder = pProjPhyNode->node.outputTsOrder;
2024-01-16 00:13:47 +00:00
pInfo->inputIgnoreGroup = pProjPhyNode->inputIgnoreGroup;
2024-03-12 03:45:54 +00:00
pInfo->outputIgnoreGroup = pProjPhyNode->ignoreGroupId;
2024-07-24 09:08:08 +00:00
2023-10-08 11:05:59 +00:00
if (pTaskInfo->execModel == OPTR_EXEC_MODEL_STREAM || pTaskInfo->execModel == OPTR_EXEC_MODEL_QUEUE) {
pInfo->mergeDataBlocks = false;
} else {
if (!pProjPhyNode->ignoreGroupId) {
pInfo->mergeDataBlocks = false;
} else {
pInfo->mergeDataBlocks = pProjPhyNode->mergeDataBlock;
}
}
2022-07-25 07:23:13 +00:00
2022-07-25 06:15:49 +00:00
int32_t numOfRows = 4096;
size_t keyBufSize = sizeof(int64_t) + sizeof(int64_t) + POINTER_BYTES;
// Make sure the size of SSDataBlock will never exceed the size of 2MB.
int32_t TWOMB = 2 * 1024 * 1024;
if (numOfRows * pResBlock->info.rowSize > TWOMB) {
numOfRows = TWOMB / pResBlock->info.rowSize;
}
2022-07-25 06:15:49 +00:00
initResultSizeInfo(&pOperator->resultInfo, numOfRows);
2024-08-15 09:03:32 +00:00
int32_t numOfCols = 0;
SExprInfo* pExprInfo = NULL;
code = createExprInfo(pProjPhyNode->pProjections, NULL, &pExprInfo, &numOfCols);
TSDB_CHECK_CODE(code, lino, _error);
2022-12-29 03:35:46 +00:00
code = initAggSup(&pOperator->exprSupp, &pInfo->aggSup, pExprInfo, numOfCols, keyBufSize, pTaskInfo->id.str,
pTaskInfo->streamInfo.pState, &pTaskInfo->storageAPI.functionStore);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
initBasicInfo(&pInfo->binfo, pResBlock);
2024-07-24 09:08:08 +00:00
code = setFunctionResultOutput(pOperator, &pInfo->binfo, &pInfo->aggSup, MAIN_SCAN, numOfCols);
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
2022-11-04 14:13:40 +00:00
code = filterInitFromNode((SNode*)pProjPhyNode->node.pConditions, &pOperator->exprSupp.pFilterInfo, 0);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _error);
2022-11-04 14:13:40 +00:00
2024-07-24 09:08:08 +00:00
code = setRowTsColumnOutputInfo(pOperator->exprSupp.pCtx, numOfCols, &pInfo->pPseudoColInfo);
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
2022-12-07 09:54:09 +00:00
setOperatorInfo(pOperator, "ProjectOperator", QUERY_NODE_PHYSICAL_PLAN_PROJECT, false, OP_NOT_OPENED, pInfo,
pTaskInfo);
2024-07-24 09:08:08 +00:00
pOperator->fpSet = createOperatorFpSet(optrDummyOpenFn, doProjectOperation1, NULL, destroyProjectOperatorInfo,
2023-07-05 03:07:08 +00:00
optrDefaultBufFn, NULL, optrDefaultGetNextExtFn, NULL);
2024-07-24 09:08:08 +00:00
setOperatorStreamStateFn(pOperator, streamOperatorReleaseState, streamOperatorReloadState);
2022-07-25 06:15:49 +00:00
2023-10-25 02:47:01 +00:00
if (NULL != downstream) {
code = appendDownstream(pOperator, &downstream, 1);
if (code != TSDB_CODE_SUCCESS) {
goto _error;
}
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
*pOptrInfo = pOperator;
return code;
2022-07-25 06:15:49 +00:00
2022-08-03 12:45:21 +00:00
_error:
2024-08-08 08:24:18 +00:00
if (pInfo != NULL) destroyProjectOperatorInfo(pInfo);
2024-08-07 07:53:23 +00:00
if (pOperator != NULL) {
pOperator->info = NULL;
2024-08-15 09:03:32 +00:00
if (pOperator->pDownstream == NULL && downstream != NULL) {
destroyOperator(downstream);
}
2024-08-07 07:53:23 +00:00
destroyOperator(pOperator);
}
pTaskInfo->code = code;
2024-07-24 09:08:08 +00:00
return code;
2022-07-25 06:15:49 +00:00
}
static int32_t discardGroupDataBlock(SSDataBlock* pBlock, SLimitInfo* pLimitInfo) {
if (pLimitInfo->remainGroupOffset > 0) {
// it is the first group
2022-11-28 04:32:40 +00:00
if (pLimitInfo->currentGroupId == 0 || pLimitInfo->currentGroupId == pBlock->info.id.groupId) {
pLimitInfo->currentGroupId = pBlock->info.id.groupId;
2022-07-25 06:15:49 +00:00
return PROJECT_RETRIEVE_CONTINUE;
2022-11-28 04:32:40 +00:00
} else if (pLimitInfo->currentGroupId != pBlock->info.id.groupId) {
2022-07-25 06:15:49 +00:00
// now it is the data from a new group
pLimitInfo->remainGroupOffset -= 1;
2022-11-28 04:32:40 +00:00
pLimitInfo->currentGroupId = pBlock->info.id.groupId;
2022-07-25 06:15:49 +00:00
// ignore data block in current group
if (pLimitInfo->remainGroupOffset > 0) {
return PROJECT_RETRIEVE_CONTINUE;
}
2023-03-22 07:18:54 +00:00
pLimitInfo->currentGroupId = 0;
}
2022-07-25 06:15:49 +00:00
}
return PROJECT_RETRIEVE_DONE;
}
static int32_t setInfoForNewGroup(SSDataBlock* pBlock, SLimitInfo* pLimitInfo, SOperatorInfo* pOperator) {
// remainGroupOffset == 0
// here check for a new group data, we need to handle the data of the previous group.
if (!(pLimitInfo->remainGroupOffset == 0 || pLimitInfo->remainGroupOffset == -1)) {
qError("project failed at: %s:%d", __func__, __LINE__);
return TSDB_CODE_INVALID_PARA;
}
2022-07-25 06:15:49 +00:00
bool newGroup = false;
if (0 == pBlock->info.id.groupId) {
pLimitInfo->numOfOutputGroups = 1;
} else if (pLimitInfo->currentGroupId != pBlock->info.id.groupId) {
2023-03-22 07:18:54 +00:00
pLimitInfo->currentGroupId = pBlock->info.id.groupId;
2022-07-25 06:15:49 +00:00
pLimitInfo->numOfOutputGroups += 1;
newGroup = true;
2023-03-22 07:18:54 +00:00
} else {
return PROJECT_RETRIEVE_CONTINUE;
}
2022-07-25 06:15:49 +00:00
2023-03-22 07:18:54 +00:00
if ((pLimitInfo->slimit.limit >= 0) && (pLimitInfo->slimit.limit < pLimitInfo->numOfOutputGroups)) {
setOperatorCompleted(pOperator);
return PROJECT_RETRIEVE_DONE;
2022-07-25 06:15:49 +00:00
}
2023-03-22 07:18:54 +00:00
// reset the value for a new group data
// existing rows that belongs to previous group.
if (newGroup) {
resetLimitInfoForNextGroup(pLimitInfo);
}
2023-03-22 07:18:54 +00:00
return PROJECT_RETRIEVE_CONTINUE;
2022-07-25 06:15:49 +00:00
}
// todo refactor
2022-08-03 12:45:21 +00:00
static int32_t doIngroupLimitOffset(SLimitInfo* pLimitInfo, uint64_t groupId, SSDataBlock* pBlock,
SOperatorInfo* pOperator) {
2022-07-25 06:15:49 +00:00
// set current group id
pLimitInfo->currentGroupId = groupId;
2023-01-20 10:08:34 +00:00
bool limitReached = applyLimitOffset(pLimitInfo, pBlock, pOperator->pTaskInfo);
2024-04-16 10:47:03 +00:00
if (pBlock->info.rows == 0 && 0 != pLimitInfo->limit.limit) {
2022-07-25 06:15:49 +00:00
return PROJECT_RETRIEVE_CONTINUE;
2023-01-20 10:08:34 +00:00
} else {
2023-03-22 07:18:54 +00:00
if (limitReached && (pLimitInfo->slimit.limit >= 0 && pLimitInfo->slimit.limit <= pLimitInfo->numOfOutputGroups)) {
2022-11-09 11:14:27 +00:00
setOperatorCompleted(pOperator);
} else if (limitReached && groupId == 0) {
setOperatorCompleted(pOperator);
2022-07-25 06:15:49 +00:00
}
}
return PROJECT_RETRIEVE_DONE;
}
2024-07-24 09:08:08 +00:00
int32_t doProjectOperation(SOperatorInfo* pOperator, SSDataBlock** pResBlock) {
QRY_OPTR_CHECK(pResBlock);
2022-07-25 06:15:49 +00:00
SProjectOperatorInfo* pProjectInfo = pOperator->info;
SOptrBasicInfo* pInfo = &pProjectInfo->binfo;
2024-08-05 03:57:18 +00:00
SExprSupp* pSup = &pOperator->exprSupp;
SSDataBlock* pRes = pInfo->pRes;
SSDataBlock* pFinalRes = pProjectInfo->pFinalRes;
int32_t code = 0;
int64_t st = 0;
int32_t order = pInfo->inputTsOrder;
int32_t scanFlag = 0;
2022-07-25 06:15:49 +00:00
blockDataCleanup(pFinalRes);
SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo;
2022-09-28 06:17:26 +00:00
2022-07-25 06:15:49 +00:00
if (pOperator->status == OP_EXEC_DONE) {
2024-07-24 09:08:08 +00:00
return code;
2022-07-25 06:15:49 +00:00
}
if (pOperator->cost.openCost == 0) {
st = taosGetTimestampUs();
}
2023-10-25 02:47:01 +00:00
SOperatorInfo* downstream = pOperator->numOfDownstream > 0 ? pOperator->pDownstream[0] : NULL;
2022-08-03 12:45:21 +00:00
SLimitInfo* pLimitInfo = &pProjectInfo->limitInfo;
2022-07-25 06:15:49 +00:00
if (downstream == NULL) {
code = doGenerateSourceData(pOperator);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
}
2024-03-12 03:45:54 +00:00
if (pProjectInfo->outputIgnoreGroup) {
pRes->info.id.groupId = 0;
}
2024-07-24 09:08:08 +00:00
*pResBlock = (pRes->info.rows > 0)? pRes:NULL;
return code;
2022-07-25 06:15:49 +00:00
}
while (1) {
while (1) {
blockDataCleanup(pRes);
// The downstream exec may change the value of the newgroup, so use a local variable instead.
2023-07-05 03:07:08 +00:00
SSDataBlock* pBlock = getNextBlockFromDownstream(pOperator, 0);
if (pBlock == NULL) {
2023-03-31 08:20:21 +00:00
qDebug("set op close, exec %d, status %d rows %" PRId64 , pTaskInfo->execModel, pOperator->status, pFinalRes->info.rows);
2022-11-09 11:14:27 +00:00
setOperatorCompleted(pOperator);
2022-07-25 06:15:49 +00:00
break;
}
// if (pTaskInfo->execModel == OPTR_EXEC_MODEL_QUEUE) {
// qDebug("set status recv");
// pOperator->status = OP_EXEC_RECV;
// }
2022-07-25 06:15:49 +00:00
// for stream interval
2022-08-12 11:28:29 +00:00
if (pBlock->info.type == STREAM_RETRIEVE || pBlock->info.type == STREAM_DELETE_RESULT ||
2023-09-15 03:41:33 +00:00
pBlock->info.type == STREAM_DELETE_DATA || pBlock->info.type == STREAM_CREATE_CHILD_TABLE ||
pBlock->info.type == STREAM_CHECKPOINT) {
2024-07-24 09:08:08 +00:00
*pResBlock = pBlock;
return code;
2022-07-25 06:15:49 +00:00
}
2024-01-16 00:13:47 +00:00
if (pProjectInfo->inputIgnoreGroup) {
2024-01-15 03:58:29 +00:00
pBlock->info.id.groupId = 0;
}
2022-07-25 06:15:49 +00:00
int32_t status = discardGroupDataBlock(pBlock, pLimitInfo);
if (status == PROJECT_RETRIEVE_CONTINUE) {
continue;
}
2024-07-24 10:43:35 +00:00
(void) setInfoForNewGroup(pBlock, pLimitInfo, pOperator);
2022-07-25 06:15:49 +00:00
if (pOperator->status == OP_EXEC_DONE) {
break;
}
if (pProjectInfo->mergeDataBlocks) {
pFinalRes->info.scanFlag = scanFlag = pBlock->info.scanFlag;
} else {
pRes->info.scanFlag = scanFlag = pBlock->info.scanFlag;
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
code = setInputDataBlock(pSup, pBlock, order, scanFlag, false);
if (code) {
T_LONG_JMP(pTaskInfo->env, code);
}
code = blockDataEnsureCapacity(pInfo->pRes, pInfo->pRes->info.rows + pBlock->info.rows);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
}
2022-07-25 06:15:49 +00:00
code = projectApplyFunctions(pSup->pExprInfo, pInfo->pRes, pBlock, pSup->pCtx, pSup->numOfExprs,
pProjectInfo->pPseudoColInfo);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
2022-07-25 06:15:49 +00:00
}
2022-11-28 04:32:40 +00:00
status = doIngroupLimitOffset(pLimitInfo, pBlock->info.id.groupId, pInfo->pRes, pOperator);
2022-07-25 06:15:49 +00:00
if (status == PROJECT_RETRIEVE_CONTINUE) {
continue;
}
break;
}
if (pProjectInfo->mergeDataBlocks) {
if (pRes->info.rows > 0) {
2022-12-29 03:35:46 +00:00
pFinalRes->info.id.groupId = 0; // clear groupId
2022-07-25 06:15:49 +00:00
pFinalRes->info.version = pRes->info.version;
// continue merge data, ignore the group id
2024-07-28 06:29:56 +00:00
int32_t ret = blockDataMerge(pFinalRes, pRes);
if (ret < 0) {
2024-07-24 09:08:08 +00:00
pTaskInfo->code = code;
return code;
}
2024-04-03 01:54:49 +00:00
if (pFinalRes->info.rows + pRes->info.rows <= pOperator->resultInfo.threshold && (pOperator->status != OP_EXEC_DONE)) {
2022-07-25 06:15:49 +00:00
continue;
}
}
// do apply filter
2024-07-24 09:08:08 +00:00
code = doFilter(pFinalRes, pOperator->exprSupp.pFilterInfo, NULL);
if (code) {
pTaskInfo->code = code;
return code;
}
2022-07-26 11:07:43 +00:00
// when apply the limit/offset for each group, pRes->info.rows may be 0, due to limit constraint.
if (pFinalRes->info.rows > 0 || (pOperator->status == OP_EXEC_DONE)) {
2023-03-29 02:46:56 +00:00
qDebug("project return %" PRId64 " rows, status %d", pFinalRes->info.rows, pOperator->status);
2022-07-25 06:15:49 +00:00
break;
}
} else {
// do apply filter
if (pRes->info.rows > 0) {
2024-07-24 09:08:08 +00:00
code = doFilter(pRes, pOperator->exprSupp.pFilterInfo, NULL);
if (code) {
pTaskInfo->code = code;
return code;
}
2022-07-25 06:15:49 +00:00
if (pRes->info.rows == 0) {
continue;
}
}
// no results generated
break;
}
}
SSDataBlock* p = pProjectInfo->mergeDataBlocks ? pFinalRes : pRes;
pOperator->resultInfo.totalRows += p->info.rows;
2022-12-22 12:02:43 +00:00
p->info.dataLoad = 1;
2022-07-25 06:15:49 +00:00
if (pOperator->cost.openCost == 0) {
pOperator->cost.openCost = (taosGetTimestampUs() - st) / 1000.0;
}
2022-08-03 12:45:21 +00:00
2024-03-12 03:45:54 +00:00
if (pProjectInfo->outputIgnoreGroup) {
p->info.id.groupId = 0;
}
2024-04-01 08:43:36 +00:00
if (pTaskInfo->execModel == OPTR_EXEC_MODEL_STREAM) {
printDataBlock(p, getStreamOpName(pOperator->operatorType), GET_TASKID(pTaskInfo));
}
2024-07-24 09:08:08 +00:00
*pResBlock = (p->info.rows > 0)? p:NULL;
return code;
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
SSDataBlock* doProjectOperation1(SOperatorInfo* pOperator) {
SSDataBlock* pRes = NULL;
2024-07-27 03:02:35 +00:00
int32_t code = doProjectOperation(pOperator, &pRes);
if (code && pOperator->pTaskInfo->code == 0) {
pOperator->pTaskInfo->code = code;
}
2024-07-24 09:08:08 +00:00
return pRes;
}
int32_t createIndefinitOutputOperatorInfo(SOperatorInfo* downstream, SPhysiNode* pNode,
SExecTaskInfo* pTaskInfo, SOperatorInfo** pOptrInfo) {
QRY_OPTR_CHECK(pOptrInfo);
int32_t code = 0;
int32_t lino = 0;
int32_t numOfRows = 4096;
size_t keyBufSize = sizeof(int64_t) + sizeof(int64_t) + POINTER_BYTES;
2022-07-25 06:15:49 +00:00
SIndefOperatorInfo* pInfo = taosMemoryCalloc(1, sizeof(SIndefOperatorInfo));
SOperatorInfo* pOperator = taosMemoryCalloc(1, sizeof(SOperatorInfo));
if (pInfo == NULL || pOperator == NULL) {
2024-07-24 09:08:08 +00:00
code = TSDB_CODE_OUT_OF_MEMORY;
2022-07-25 06:15:49 +00:00
goto _error;
}
2022-08-03 12:45:21 +00:00
pOperator->pTaskInfo = pTaskInfo;
2022-07-25 06:15:49 +00:00
SExprSupp* pSup = &pOperator->exprSupp;
pSup->hasWindowOrGroup = false;
2022-07-25 06:15:49 +00:00
SIndefRowsFuncPhysiNode* pPhyNode = (SIndefRowsFuncPhysiNode*)pNode;
if (pPhyNode->pExprs != NULL) {
int32_t num = 0;
2024-08-05 03:57:18 +00:00
SExprInfo* pSExpr = NULL;
code = createExprInfo(pPhyNode->pExprs, NULL, &pSExpr, &num);
QUERY_CHECK_CODE(code, lino, _error);
2024-07-24 09:08:08 +00:00
code = initExprSupp(&pInfo->scalarSup, pSExpr, num, &pTaskInfo->storageAPI.functionStore);
2022-07-25 06:15:49 +00:00
if (code != TSDB_CODE_SUCCESS) {
goto _error;
}
}
2022-11-27 16:27:49 +00:00
SSDataBlock* pResBlock = createDataBlockFromDescNode(pPhyNode->node.pOutputDataBlockDesc);
2024-08-05 09:08:21 +00:00
TSDB_CHECK_NULL(pResBlock, code, lino, _error, terrno);
2022-07-25 06:15:49 +00:00
// Make sure the size of SSDataBlock will never exceed the size of 2MB.
int32_t TWOMB = 2 * 1024 * 1024;
if (numOfRows * pResBlock->info.rowSize > TWOMB) {
numOfRows = TWOMB / pResBlock->info.rowSize;
}
2022-10-18 03:43:58 +00:00
initBasicInfo(&pInfo->binfo, pResBlock);
2022-07-25 06:15:49 +00:00
initResultSizeInfo(&pOperator->resultInfo, numOfRows);
2024-07-24 09:08:08 +00:00
code = blockDataEnsureCapacity(pResBlock, numOfRows);
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
2024-08-15 09:38:29 +00:00
int32_t numOfExpr = 0;
SExprInfo* pExprInfo = NULL;
code = createExprInfo(pPhyNode->pFuncs, NULL, &pExprInfo, &numOfExpr);
TSDB_CHECK_CODE(code, lino, _error);
2024-07-24 09:08:08 +00:00
code = initAggSup(pSup, &pInfo->aggSup, pExprInfo, numOfExpr, keyBufSize, pTaskInfo->id.str,
pTaskInfo->streamInfo.pState, &pTaskInfo->storageAPI.functionStore);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _error);
code = setFunctionResultOutput(pOperator, &pInfo->binfo, &pInfo->aggSup, MAIN_SCAN, numOfExpr);
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
2022-11-04 14:13:40 +00:00
code = filterInitFromNode((SNode*)pPhyNode->node.pConditions, &pOperator->exprSupp.pFilterInfo, 0);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
pInfo->binfo.pRes = pResBlock;
pInfo->binfo.inputTsOrder = pNode->inputTsOrder;
pInfo->binfo.outputTsOrder = pNode->outputTsOrder;
2024-07-24 09:08:08 +00:00
code = setRowTsColumnOutputInfo(pSup->pCtx, numOfExpr, &pInfo->pPseudoColInfo);
TSDB_CHECK_CODE(code, lino, _error);
2022-07-25 06:15:49 +00:00
2022-12-07 09:54:09 +00:00
setOperatorInfo(pOperator, "IndefinitOperator", QUERY_NODE_PHYSICAL_PLAN_INDEF_ROWS_FUNC, false, OP_NOT_OPENED, pInfo,
pTaskInfo);
2024-07-24 09:08:08 +00:00
pOperator->fpSet = createOperatorFpSet(optrDummyOpenFn, doApplyIndefinitFunction1, NULL, destroyIndefinitOperatorInfo,
2023-07-05 03:07:08 +00:00
optrDefaultBufFn, NULL, optrDefaultGetNextExtFn, NULL);
2022-07-25 06:15:49 +00:00
code = appendDownstream(pOperator, &downstream, 1);
2022-07-25 06:15:49 +00:00
if (code != TSDB_CODE_SUCCESS) {
goto _error;
}
2024-07-24 09:08:08 +00:00
*pOptrInfo = pOperator;
return code;
2022-07-25 06:15:49 +00:00
2022-08-03 12:45:21 +00:00
_error:
2024-08-08 08:24:18 +00:00
if (pInfo != NULL) destroyIndefinitOperatorInfo(pInfo);
2024-08-07 07:53:23 +00:00
if (pOperator != NULL) {
pOperator->info = NULL;
2024-08-15 09:03:32 +00:00
if (pOperator->pDownstream == NULL && downstream != NULL) {
destroyOperator(downstream);
}
2024-08-07 07:53:23 +00:00
destroyOperator(pOperator);
}
pTaskInfo->code = code;
2024-07-24 09:08:08 +00:00
return code;
2022-07-25 06:15:49 +00:00
}
static void doHandleDataBlock(SOperatorInfo* pOperator, SSDataBlock* pBlock, SOperatorInfo* downstream,
SExecTaskInfo* pTaskInfo) {
SIndefOperatorInfo* pIndefInfo = pOperator->info;
SOptrBasicInfo* pInfo = &pIndefInfo->binfo;
SExprSupp* pSup = &pOperator->exprSupp;
int32_t order = pInfo->inputTsOrder;
int32_t scanFlag = pBlock->info.scanFlag;
int32_t code = TSDB_CODE_SUCCESS;
2022-07-25 06:15:49 +00:00
// there is an scalar expression that needs to be calculated before apply the group aggregation.
SExprSupp* pScalarSup = &pIndefInfo->scalarSup;
if (pScalarSup->pExprInfo != NULL) {
code = projectApplyFunctions(pScalarSup->pExprInfo, pBlock, pBlock, pScalarSup->pCtx, pScalarSup->numOfExprs,
pIndefInfo->pPseudoColInfo);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
2022-07-25 06:15:49 +00:00
}
}
2024-07-24 09:08:08 +00:00
code = setInputDataBlock(pSup, pBlock, order, scanFlag, false);
if (code) {
T_LONG_JMP(pTaskInfo->env, code);
}
code = blockDataEnsureCapacity(pInfo->pRes, pInfo->pRes->info.rows + pBlock->info.rows);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
}
2022-07-25 06:15:49 +00:00
code = projectApplyFunctions(pSup->pExprInfo, pInfo->pRes, pBlock, pSup->pCtx, pSup->numOfExprs,
pIndefInfo->pPseudoColInfo);
if (code != TSDB_CODE_SUCCESS) {
T_LONG_JMP(pTaskInfo->env, code);
2022-07-25 06:15:49 +00:00
}
}
2024-07-24 09:08:08 +00:00
SSDataBlock* doApplyIndefinitFunction1(SOperatorInfo* pOperator) {
SSDataBlock* pResBlock = NULL;
pOperator->pTaskInfo->code = doApplyIndefinitFunction(pOperator, &pResBlock);
return pResBlock;
}
int32_t doApplyIndefinitFunction(SOperatorInfo* pOperator, SSDataBlock** pResBlock) {
QRY_OPTR_CHECK(pResBlock);
2022-07-25 06:15:49 +00:00
SIndefOperatorInfo* pIndefInfo = pOperator->info;
SOptrBasicInfo* pInfo = &pIndefInfo->binfo;
SExprSupp* pSup = &pOperator->exprSupp;
2024-07-24 09:08:08 +00:00
int64_t st = 0;
int32_t code = 0;
SSDataBlock* pRes = pInfo->pRes;
2022-07-25 06:15:49 +00:00
blockDataCleanup(pRes);
SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo;
if (pOperator->status == OP_EXEC_DONE) {
2024-07-24 09:08:08 +00:00
return 0;
2022-07-25 06:15:49 +00:00
}
if (pOperator->cost.openCost == 0) {
st = taosGetTimestampUs();
}
SOperatorInfo* downstream = pOperator->pDownstream[0];
while (1) {
// here we need to handle the existsed group results
if (pIndefInfo->pNextGroupRes != NULL) { // todo extract method
for (int32_t k = 0; k < pSup->numOfExprs; ++k) {
SqlFunctionCtx* pCtx = &pSup->pCtx[k];
SResultRowEntryInfo* pResInfo = GET_RES_INFO(pCtx);
pResInfo->initialized = false;
pCtx->pOutput = NULL;
}
doHandleDataBlock(pOperator, pIndefInfo->pNextGroupRes, downstream, pTaskInfo);
pIndefInfo->pNextGroupRes = NULL;
}
if (pInfo->pRes->info.rows < pOperator->resultInfo.threshold) {
while (1) {
// The downstream exec may change the value of the newgroup, so use a local variable instead.
2023-07-05 03:07:08 +00:00
SSDataBlock* pBlock = getNextBlockFromDownstream(pOperator, 0);
2022-07-25 06:15:49 +00:00
if (pBlock == NULL) {
2022-11-09 11:14:27 +00:00
setOperatorCompleted(pOperator);
2022-07-25 06:15:49 +00:00
break;
}
pInfo->pRes->info.scanFlag = pBlock->info.scanFlag;
2022-07-25 06:15:49 +00:00
2022-11-28 04:32:40 +00:00
if (pIndefInfo->groupId == 0 && pBlock->info.id.groupId != 0) {
pIndefInfo->groupId = pBlock->info.id.groupId; // this is the initial group result
2022-07-25 06:15:49 +00:00
} else {
2022-11-28 04:32:40 +00:00
if (pIndefInfo->groupId != pBlock->info.id.groupId) { // reset output buffer and computing status
pIndefInfo->groupId = pBlock->info.id.groupId;
2022-07-25 06:15:49 +00:00
pIndefInfo->pNextGroupRes = pBlock;
break;
}
}
doHandleDataBlock(pOperator, pBlock, downstream, pTaskInfo);
if (pInfo->pRes->info.rows >= pOperator->resultInfo.threshold) {
break;
}
}
}
2024-07-24 09:08:08 +00:00
code = doFilter(pInfo->pRes, pOperator->exprSupp.pFilterInfo, NULL);
if (code) {
pTaskInfo->code = code;
return code;
}
2022-07-25 06:15:49 +00:00
size_t rows = pInfo->pRes->info.rows;
if (rows > 0 || pOperator->status == OP_EXEC_DONE) {
break;
} else {
blockDataCleanup(pInfo->pRes);
}
}
size_t rows = pInfo->pRes->info.rows;
pOperator->resultInfo.totalRows += rows;
if (pOperator->cost.openCost == 0) {
pOperator->cost.openCost = (taosGetTimestampUs() - st) / 1000.0;
}
2024-07-24 09:08:08 +00:00
*pResBlock = (rows > 0) ? pInfo->pRes : NULL;
return code;
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
int32_t initCtxOutputBuffer(SqlFunctionCtx* pCtx, int32_t size) {
int32_t code = TSDB_CODE_SUCCESS;
2022-07-25 06:15:49 +00:00
for (int32_t j = 0; j < size; ++j) {
struct SResultRowEntryInfo* pResInfo = GET_RES_INFO(&pCtx[j]);
if (isRowEntryInitialized(pResInfo) || fmIsPseudoColumnFunc(pCtx[j].functionId) || pCtx[j].functionId == -1 ||
fmIsScalarFunc(pCtx[j].functionId)) {
continue;
}
code = pCtx[j].fpSet.init(&pCtx[j], pCtx[j].resultInfo);
if (code) {
return code;
}
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
return 0;
2022-07-25 06:15:49 +00:00
}
/*
* The start of each column SResultRowEntryInfo is denote by RowCellInfoOffset.
* Note that in case of top/bottom query, the whole multiple rows of result is treated as only one row of results.
* +------------+-----------------result column 1------------+------------------result column 2-----------+
* | SResultRow | SResultRowEntryInfo | intermediate buffer1 | SResultRowEntryInfo | intermediate buffer 2|
* +------------+--------------------------------------------+--------------------------------------------+
* offset[0] offset[1] offset[2]
*/
// TODO refactor: some function move away
2024-07-24 09:08:08 +00:00
int32_t setFunctionResultOutput(SOperatorInfo* pOperator, SOptrBasicInfo* pInfo, SAggSupporter* pSup, int32_t stage,
2022-07-25 06:15:49 +00:00
int32_t numOfExprs) {
SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo;
SqlFunctionCtx* pCtx = pOperator->exprSupp.pCtx;
int32_t* rowEntryInfoOffset = pOperator->exprSupp.rowEntryInfoOffset;
SResultRowInfo* pResultRowInfo = &pInfo->resultRowInfo;
initResultRowInfo(pResultRowInfo);
int64_t tid = 0;
int64_t groupId = 0;
SResultRow* pRow = doSetResultOutBufByKey(pSup->pResultBuf, pResultRowInfo, (char*)&tid, sizeof(tid), true, groupId,
2023-03-12 03:03:41 +00:00
pTaskInfo, false, pSup, true);
2024-08-06 09:32:20 +00:00
if (pRow == NULL || pTaskInfo->code != 0) {
return pTaskInfo->code;
}
2022-07-25 06:15:49 +00:00
for (int32_t i = 0; i < numOfExprs; ++i) {
struct SResultRowEntryInfo* pEntry = getResultEntryInfo(pRow, i, rowEntryInfoOffset);
cleanupResultRowEntry(pEntry);
pCtx[i].resultInfo = pEntry;
pCtx[i].scanFlag = stage;
}
2024-07-24 09:08:08 +00:00
return initCtxOutputBuffer(pCtx, numOfExprs);
2022-07-25 06:15:49 +00:00
}
2024-07-24 09:08:08 +00:00
int32_t setRowTsColumnOutputInfo(SqlFunctionCtx* pCtx, int32_t numOfCols, SArray** pResList) {
QRY_OPTR_CHECK(pResList);
2022-07-25 06:15:49 +00:00
SArray* pList = taosArrayInit(4, sizeof(int32_t));
2024-07-24 09:08:08 +00:00
if (pList == NULL) {
return terrno;
}
2022-07-25 06:15:49 +00:00
for (int32_t i = 0; i < numOfCols; ++i) {
if (fmIsPseudoColumnFunc(pCtx[i].functionId)) {
2024-07-24 09:08:08 +00:00
void* px = taosArrayPush(pList, &i);
if (px == NULL) {
return terrno;
}
2022-07-25 06:15:49 +00:00
}
}
2024-07-24 09:08:08 +00:00
*pResList = pList;
return 0;
2022-07-25 06:15:49 +00:00
}
int32_t doGenerateSourceData(SOperatorInfo* pOperator) {
2022-07-25 06:15:49 +00:00
SProjectOperatorInfo* pProjectInfo = pOperator->info;
SExprSupp* pSup = &pOperator->exprSupp;
SSDataBlock* pRes = pProjectInfo->binfo.pRes;
2024-07-24 09:08:08 +00:00
SExprInfo* pExpr = pSup->pExprInfo;
int64_t st = taosGetTimestampUs();
SExecTaskInfo* pTaskInfo = pOperator->pTaskInfo;
2022-07-25 06:15:49 +00:00
2024-07-24 09:08:08 +00:00
int32_t code = blockDataEnsureCapacity(pRes, pOperator->resultInfo.capacity);
if (code) {
return code;
}
2022-07-25 06:15:49 +00:00
for (int32_t k = 0; k < pSup->numOfExprs; ++k) {
int32_t outputSlotId = pExpr[k].base.resSchema.slotId;
if (pExpr[k].pExpr->nodeType == QUERY_NODE_VALUE) {
SColumnInfoData* pColInfoData = taosArrayGet(pRes->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pColInfoData == NULL) {
return terrno;
}
2022-07-25 06:15:49 +00:00
int32_t type = pExpr[k].base.pParam[0].param.nType;
if (TSDB_DATA_TYPE_NULL == type) {
colDataSetNNULL(pColInfoData, 0, 1);
} else {
2024-07-24 09:08:08 +00:00
code = colDataSetVal(pColInfoData, 0, taosVariantGet(&pExpr[k].base.pParam[0].param, type), false);
if (code) {
return code;
}
}
} else if (pExpr[k].pExpr->nodeType == QUERY_NODE_FUNCTION) {
SqlFunctionCtx* pfCtx = &pSup->pCtx[k];
2023-07-06 08:11:41 +00:00
// UDF scalar functions will be calculated here, for example, select foo(n) from (select 1 n).
// UDF aggregate functions will be handled in agg operator.
if (fmIsScalarFunc(pfCtx->functionId)) {
SArray* pBlockList = taosArrayInit(4, POINTER_BYTES);
2024-07-24 09:08:08 +00:00
if (pBlockList == NULL) {
return terrno;
}
void* px = taosArrayPush(pBlockList, &pRes);
if (px == NULL) {
return terrno;
}
SColumnInfoData* pResColData = taosArrayGet(pRes->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pResColData == NULL) {
return terrno;
}
SColumnInfoData idata = {.info = pResColData->info, .hasNull = true};
SScalarParam dest = {.columnData = &idata};
2024-07-24 09:08:08 +00:00
code = scalarCalculate((SNode*)pExpr[k].pExpr->_function.pFunctNode, pBlockList, &dest);
if (code != TSDB_CODE_SUCCESS) {
taosArrayDestroy(pBlockList);
return code;
}
2022-07-25 06:15:49 +00:00
int32_t startOffset = pRes->info.rows;
if (pRes->info.capacity <= 0) {
qError("project failed at: %s:%d", __func__, __LINE__);
return TSDB_CODE_FAILED;
}
2024-07-24 09:08:08 +00:00
code = colDataAssign(pResColData, &idata, dest.numOfRows, &pRes->info);
if (code) {
return code;
}
2024-07-24 09:08:08 +00:00
colDataDestroy(&idata);
taosArrayDestroy(pBlockList);
2023-07-06 08:11:41 +00:00
} else {
return TSDB_CODE_OPS_NOT_SUPPORT;
}
2022-07-25 06:15:49 +00:00
} else {
return TSDB_CODE_OPS_NOT_SUPPORT;
2022-07-25 06:15:49 +00:00
}
}
pRes->info.rows = 1;
2024-07-24 09:08:08 +00:00
code = doFilter(pRes, pOperator->exprSupp.pFilterInfo, NULL);
if (code) {
pTaskInfo->code = code;
return code;
}
2022-07-25 06:15:49 +00:00
2024-07-24 09:08:08 +00:00
(void) doIngroupLimitOffset(&pProjectInfo->limitInfo, 0, pRes, pOperator);
2022-07-25 06:15:49 +00:00
pOperator->resultInfo.totalRows += pRes->info.rows;
2022-11-09 11:14:27 +00:00
setOperatorCompleted(pOperator);
2022-07-25 06:15:49 +00:00
if (pOperator->cost.openCost == 0) {
pOperator->cost.openCost = (taosGetTimestampUs() - st) / 1000.0;
}
2024-07-24 09:08:08 +00:00
return code;
2022-07-25 13:50:51 +00:00
}
2022-11-28 04:32:40 +00:00
static void setPseudoOutputColInfo(SSDataBlock* pResult, SqlFunctionCtx* pCtx, SArray* pPseudoList) {
size_t num = (pPseudoList != NULL) ? taosArrayGetSize(pPseudoList) : 0;
for (int32_t i = 0; i < num; ++i) {
pCtx[i].pOutput = taosArrayGet(pResult->pDataBlock, i);
2024-07-24 09:08:08 +00:00
if (pCtx[i].pOutput == NULL) {
qError("failed to get the output buf, ptr is null");
}
2022-11-28 04:32:40 +00:00
}
}
int32_t projectApplyFunctions(SExprInfo* pExpr, SSDataBlock* pResult, SSDataBlock* pSrcBlock, SqlFunctionCtx* pCtx,
int32_t numOfOutput, SArray* pPseudoList) {
2024-07-24 09:08:08 +00:00
int32_t lino = 0;
2024-07-04 03:04:59 +00:00
int32_t code = TSDB_CODE_SUCCESS;
2022-11-28 04:32:40 +00:00
setPseudoOutputColInfo(pResult, pCtx, pPseudoList);
2022-12-12 17:30:10 +00:00
pResult->info.dataLoad = 1;
2022-11-28 04:32:40 +00:00
2024-07-08 12:38:23 +00:00
SArray* processByRowFunctionCtx = NULL;
2022-11-28 04:32:40 +00:00
if (pSrcBlock == NULL) {
for (int32_t k = 0; k < numOfOutput; ++k) {
int32_t outputSlotId = pExpr[k].base.resSchema.slotId;
if (pExpr[k].pExpr->nodeType != QUERY_NODE_VALUE) {
qError("project failed at: %s:%d", __func__, __LINE__);
code = TSDB_CODE_INVALID_PARA;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
SColumnInfoData* pColInfoData = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pColInfoData == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
int32_t type = pExpr[k].base.pParam[0].param.nType;
if (TSDB_DATA_TYPE_NULL == type) {
2023-02-20 02:04:08 +00:00
colDataSetNNULL(pColInfoData, 0, 1);
2022-11-28 04:32:40 +00:00
} else {
2024-07-24 09:08:08 +00:00
code = colDataSetVal(pColInfoData, 0, taosVariantGet(&pExpr[k].base.pParam[0].param, type), false);
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 04:32:40 +00:00
}
}
pResult->info.rows = 1;
2024-07-04 03:04:59 +00:00
goto _exit;
2022-11-28 04:32:40 +00:00
}
if (pResult != pSrcBlock) {
pResult->info.id.groupId = pSrcBlock->info.id.groupId;
memcpy(pResult->info.parTbName, pSrcBlock->info.parTbName, TSDB_TABLE_NAME_LEN);
}
// if the source equals to the destination, it is to create a new column as the result of scalar
// function or some operators.
bool createNewColModel = (pResult == pSrcBlock);
if (createNewColModel) {
2024-07-24 09:08:08 +00:00
code = blockDataEnsureCapacity(pResult, pResult->info.rows);
if (code) {
goto _exit;
}
2022-11-28 04:32:40 +00:00
}
int32_t numOfRows = 0;
for (int32_t k = 0; k < numOfOutput; ++k) {
int32_t outputSlotId = pExpr[k].base.resSchema.slotId;
SqlFunctionCtx* pfCtx = &pCtx[k];
SInputColumnInfoData* pInputData = &pfCtx->input;
if (pExpr[k].pExpr->nodeType == QUERY_NODE_COLUMN) { // it is a project query
SColumnInfoData* pColInfoData = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pColInfoData == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
if (pResult->info.rows > 0 && !createNewColModel) {
2024-07-25 01:55:27 +00:00
int32_t ret = 0;
2022-12-20 02:15:14 +00:00
if (pInputData->pData[0] == NULL) {
int32_t slotId = pfCtx->param[0].pCol->slotId;
SColumnInfoData* pInput = taosArrayGet(pSrcBlock->pDataBlock, slotId);
2024-07-24 09:08:08 +00:00
if (pInput == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-12-20 02:15:14 +00:00
2024-07-25 01:55:27 +00:00
ret = colDataMergeCol(pColInfoData, pResult->info.rows, (int32_t*)&pResult->info.capacity, pInput,
pSrcBlock->info.rows);
2022-12-20 02:15:14 +00:00
} else {
2024-07-25 01:55:27 +00:00
ret = colDataMergeCol(pColInfoData, pResult->info.rows, (int32_t*)&pResult->info.capacity,
pInputData->pData[0], pInputData->numOfRows);
}
if (ret < 0) {
code = ret;
2022-12-20 02:15:14 +00:00
}
2024-07-25 01:55:27 +00:00
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 04:32:40 +00:00
} else {
2022-12-20 02:15:14 +00:00
if (pInputData->pData[0] == NULL) {
int32_t slotId = pfCtx->param[0].pCol->slotId;
2022-11-28 04:32:40 +00:00
2022-12-20 02:15:14 +00:00
SColumnInfoData* pInput = taosArrayGet(pSrcBlock->pDataBlock, slotId);
2024-07-24 09:08:08 +00:00
if (pInput == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
2024-07-24 09:08:08 +00:00
code = colDataAssign(pColInfoData, pInput, pSrcBlock->info.rows, &pResult->info);
2022-12-20 02:15:14 +00:00
numOfRows = pSrcBlock->info.rows;
} else {
2024-07-24 09:08:08 +00:00
code = colDataAssign(pColInfoData, pInputData->pData[0], pInputData->numOfRows, &pResult->info);
2022-12-20 02:15:14 +00:00
numOfRows = pInputData->numOfRows;
}
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _exit);
2022-12-20 02:15:14 +00:00
}
2022-11-28 04:32:40 +00:00
} else if (pExpr[k].pExpr->nodeType == QUERY_NODE_VALUE) {
SColumnInfoData* pColInfoData = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pColInfoData == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
int32_t offset = createNewColModel ? 0 : pResult->info.rows;
int32_t type = pExpr[k].base.pParam[0].param.nType;
if (TSDB_DATA_TYPE_NULL == type) {
2023-02-20 02:04:08 +00:00
colDataSetNNULL(pColInfoData, offset, pSrcBlock->info.rows);
2022-11-28 04:32:40 +00:00
} else {
char* p = taosVariantGet(&pExpr[k].base.pParam[0].param, type);
2022-11-28 06:36:55 +00:00
for (int32_t i = 0; i < pSrcBlock->info.rows; ++i) {
2024-07-24 09:08:08 +00:00
code = colDataSetVal(pColInfoData, i + offset, p, false);
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 06:36:55 +00:00
}
2022-11-28 04:32:40 +00:00
}
numOfRows = pSrcBlock->info.rows;
} else if (pExpr[k].pExpr->nodeType == QUERY_NODE_OPERATOR) {
SArray* pBlockList = taosArrayInit(4, POINTER_BYTES);
2024-07-24 09:08:08 +00:00
if (pBlockList == NULL) {
code = terrno;
goto _exit;
}
void* px = taosArrayPush(pBlockList, &pSrcBlock);
if (px == NULL) {
code = terrno;
taosArrayDestroy(pBlockList);
goto _exit;
}
2022-11-28 04:32:40 +00:00
SColumnInfoData* pResColData = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pResColData == NULL) {
code = terrno;
taosArrayDestroy(pBlockList);
goto _exit;
}
2022-11-28 04:32:40 +00:00
SColumnInfoData idata = {.info = pResColData->info, .hasNull = true};
SScalarParam dest = {.columnData = &idata};
2024-07-04 03:04:59 +00:00
code = scalarCalculate(pExpr[k].pExpr->_optrRoot.pRootNode, pBlockList, &dest);
2022-11-28 04:32:40 +00:00
if (code != TSDB_CODE_SUCCESS) {
taosArrayDestroy(pBlockList);
2024-07-04 03:04:59 +00:00
goto _exit;
2022-11-28 04:32:40 +00:00
}
int32_t startOffset = createNewColModel ? 0 : pResult->info.rows;
if (pResult->info.capacity <= 0) {
qError("project failed at: %s:%d", __func__, __LINE__);
code = TSDB_CODE_INVALID_PARA;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
2024-07-25 01:55:27 +00:00
int32_t ret = colDataMergeCol(pResColData, startOffset, (int32_t*)&pResult->info.capacity, &idata, dest.numOfRows);
if (ret < 0) {
code = ret;
}
2022-11-28 04:32:40 +00:00
2024-07-25 01:55:27 +00:00
colDataDestroy(&idata);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _exit);
2024-07-25 01:55:27 +00:00
2022-11-28 04:32:40 +00:00
numOfRows = dest.numOfRows;
taosArrayDestroy(pBlockList);
} else if (pExpr[k].pExpr->nodeType == QUERY_NODE_FUNCTION) {
// _rowts/_c0, not tbname column
if (fmIsPseudoColumnFunc(pfCtx->functionId) && (!fmIsScanPseudoColumnFunc(pfCtx->functionId))) {
// do nothing
} else if (fmIsIndefiniteRowsFunc(pfCtx->functionId)) {
SResultRowEntryInfo* pResInfo = GET_RES_INFO(pfCtx);
code = pfCtx->fpSet.init(pfCtx, pResInfo);
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 04:32:40 +00:00
pfCtx->pOutput = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pfCtx->pOutput == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
pfCtx->offset = createNewColModel ? 0 : pResult->info.rows; // set the start offset
// set the timestamp(_rowts) output buffer
if (taosArrayGetSize(pPseudoList) > 0) {
int32_t* outputColIndex = taosArrayGet(pPseudoList, 0);
2024-07-24 09:08:08 +00:00
if (outputColIndex == NULL) {
code = terrno;
goto _exit;
}
2022-11-28 04:32:40 +00:00
pfCtx->pTsOutput = (SColumnInfoData*)pCtx[*outputColIndex].pOutput;
}
// link pDstBlock to set selectivity value
if (pfCtx->subsidiaries.num > 0) {
pfCtx->pDstBlock = pResult;
}
2024-07-04 03:04:59 +00:00
code = pfCtx->fpSet.process(pfCtx);
2022-11-28 04:32:40 +00:00
if (code != TSDB_CODE_SUCCESS) {
2024-07-04 03:04:59 +00:00
goto _exit;
2022-11-28 04:32:40 +00:00
}
2024-07-24 09:08:08 +00:00
2022-11-28 04:32:40 +00:00
numOfRows = pResInfo->numOfRes;
2024-07-04 03:04:59 +00:00
if (fmIsProcessByRowFunc(pfCtx->functionId)) {
2024-07-08 12:38:23 +00:00
if (NULL == processByRowFunctionCtx) {
processByRowFunctionCtx = taosArrayInit(1, sizeof(SqlFunctionCtx*));
if (!processByRowFunctionCtx) {
2024-07-04 03:04:59 +00:00
code = terrno;
goto _exit;
}
}
2024-07-24 09:08:08 +00:00
void* px = taosArrayPush(processByRowFunctionCtx, &pfCtx);
if (px == NULL) {
code = terrno;
goto _exit;
}
2024-07-04 03:04:59 +00:00
}
2022-11-28 04:32:40 +00:00
} else if (fmIsAggFunc(pfCtx->functionId)) {
// selective value output should be set during corresponding function execution
if (fmIsSelectValueFunc(pfCtx->functionId)) {
continue;
}
// _group_key function for "partition by tbname" + csum(col_name) query
SColumnInfoData* pOutput = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pOutput == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
int32_t slotId = pfCtx->param[0].pCol->slotId;
// todo handle the json tag
SColumnInfoData* pInput = taosArrayGet(pSrcBlock->pDataBlock, slotId);
2024-07-24 09:08:08 +00:00
if (pInput == NULL) {
code = terrno;
TSDB_CHECK_CODE(code, lino, _exit);
}
2022-11-28 04:32:40 +00:00
for (int32_t f = 0; f < pSrcBlock->info.rows; ++f) {
bool isNull = colDataIsNull_s(pInput, f);
if (isNull) {
2023-02-20 02:04:08 +00:00
colDataSetNULL(pOutput, pResult->info.rows + f);
2022-11-28 04:32:40 +00:00
} else {
char* data = colDataGetData(pInput, f);
2024-07-24 09:08:08 +00:00
code = colDataSetVal(pOutput, pResult->info.rows + f, data, isNull);
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 04:32:40 +00:00
}
}
} else {
SArray* pBlockList = taosArrayInit(4, POINTER_BYTES);
2024-07-24 09:08:08 +00:00
if (pBlockList == NULL) {
code = terrno;
goto _exit;
}
void* px = taosArrayPush(pBlockList, &pSrcBlock);
if (px == NULL) {
code = terrno;
goto _exit;
}
2022-11-28 04:32:40 +00:00
SColumnInfoData* pResColData = taosArrayGet(pResult->pDataBlock, outputSlotId);
2024-07-24 09:08:08 +00:00
if (pResColData == NULL) {
taosArrayDestroy(pBlockList);
code = terrno;
goto _exit;
}
2022-11-28 04:32:40 +00:00
SColumnInfoData idata = {.info = pResColData->info, .hasNull = true};
SScalarParam dest = {.columnData = &idata};
2024-07-05 00:07:00 +00:00
code = scalarCalculate((SNode*)pExpr[k].pExpr->_function.pFunctNode, pBlockList, &dest);
2022-11-28 04:32:40 +00:00
if (code != TSDB_CODE_SUCCESS) {
taosArrayDestroy(pBlockList);
2024-07-04 03:04:59 +00:00
goto _exit;
2022-11-28 04:32:40 +00:00
}
int32_t startOffset = createNewColModel ? 0 : pResult->info.rows;
if (pResult->info.capacity <= 0) {
qError("project failed at: %s:%d", __func__, __LINE__);
code = TSDB_CODE_INVALID_PARA;
TSDB_CHECK_CODE(code, lino, _exit);
}
2024-07-25 01:55:27 +00:00
int32_t ret = colDataMergeCol(pResColData, startOffset, (int32_t*)&pResult->info.capacity, &idata, dest.numOfRows);
if (ret < 0) {
code = ret;
}
2022-11-28 04:32:40 +00:00
colDataDestroy(&idata);
numOfRows = dest.numOfRows;
taosArrayDestroy(pBlockList);
2024-07-24 09:08:08 +00:00
TSDB_CHECK_CODE(code, lino, _exit);
2022-11-28 04:32:40 +00:00
}
} else {
return TSDB_CODE_OPS_NOT_SUPPORT;
}
}
2024-07-08 12:38:23 +00:00
if (processByRowFunctionCtx && taosArrayGetSize(processByRowFunctionCtx) > 0){
SqlFunctionCtx** pfCtx = taosArrayGet(processByRowFunctionCtx, 0);
2024-07-24 09:08:08 +00:00
if (pfCtx == NULL) {
code = terrno;
2024-07-04 07:56:12 +00:00
goto _exit;
}
2024-07-24 09:08:08 +00:00
code = (*pfCtx)->fpSet.processFuncByRow(processByRowFunctionCtx);
TSDB_CHECK_CODE(code, lino, _exit);
2024-07-04 03:04:59 +00:00
numOfRows = (*pfCtx)->resultInfo->numOfRes;
}
2024-07-25 01:55:27 +00:00
2022-11-28 04:32:40 +00:00
if (!createNewColModel) {
pResult->info.rows += numOfRows;
}
2024-07-24 09:08:08 +00:00
2024-07-04 03:04:59 +00:00
_exit:
2024-07-08 12:38:23 +00:00
if(processByRowFunctionCtx) {
taosArrayDestroy(processByRowFunctionCtx);
2024-07-04 03:04:59 +00:00
}
return code;
2022-11-28 04:32:40 +00:00
}