Merge pull request #13639 from taosdata/enh/hll_split

enh(query): add hll function distributed splitting

include/libs/function/functionMgt.h

@@ -129,6 +129,8 @@ typedef enum EFunctionType {
   FUNCTION_TYPE_SPREAD_MERGE,
   FUNCTION_TYPE_HISTOGRAM_PARTIAL,
   FUNCTION_TYPE_HISTOGRAM_MERGE,
+  FUNCTION_TYPE_HYPERLOGLOG_PARTIAL,
+  FUNCTION_TYPE_HYPERLOGLOG_MERGE,
 
   // user defined funcion
   FUNCTION_TYPE_UDF = 10000

source/libs/function/inc/builtinsimpl.h

@@ -126,7 +126,10 @@ int32_t getHistogramInfoSize();
 
 bool getHLLFuncEnv(struct SFunctionNode* pFunc, SFuncExecEnv* pEnv);
 int32_t hllFunction(SqlFunctionCtx* pCtx);
+int32_t hllFunctionMerge(SqlFunctionCtx* pCtx);
 int32_t hllFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock);
+int32_t hllPartialFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock);
+int32_t getHLLInfoSize();
 
 bool getStateFuncEnv(struct SFunctionNode* pFunc, SFuncExecEnv* pEnv);
 bool stateFunctionSetup(SqlFunctionCtx *pCtx, SResultRowEntryInfo* pResultInfo);

source/libs/function/src/builtins.c

@@ -313,6 +313,7 @@ static int32_t translateApercentileImpl(SFunctionNode* pFunc, char* pErrBuf, int
 static int32_t translateApercentilePartial(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateApercentileImpl(pFunc, pErrBuf, len, true);
 }
+
 static int32_t translateApercentileMerge(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateApercentileImpl(pFunc, pErrBuf, len, false);
 }
@@ -401,6 +402,7 @@ static int32_t translateSpreadImpl(SFunctionNode* pFunc, char* pErrBuf, int32_t
 static int32_t translateSpreadPartial(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateSpreadImpl(pFunc, pErrBuf, len, true);
 }
+
 static int32_t translateSpreadMerge(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateSpreadImpl(pFunc, pErrBuf, len, false);
 }
@@ -551,6 +553,7 @@ static int32_t translateHistogramImpl(SFunctionNode* pFunc, char* pErrBuf, int32
 static int32_t translateHistogramPartial(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateHistogramImpl(pFunc, pErrBuf, len, true);
 }
+
 static int32_t translateHistogramMerge(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return translateHistogramImpl(pFunc, pErrBuf, len, false);
 }
@@ -564,6 +567,28 @@ static int32_t translateHLL(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   return TSDB_CODE_SUCCESS;
 }
 
+static int32_t translateHLLImpl(SFunctionNode* pFunc, char* pErrBuf, int32_t len, bool isPartial) {
+  if (1 != LIST_LENGTH(pFunc->pParameterList)) {
+    return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName);
+  }
+
+  if (isPartial) {
+    pFunc->node.resType = (SDataType){.bytes = getHistogramInfoSize() + VARSTR_HEADER_SIZE, .type = TSDB_DATA_TYPE_BINARY};
+  } else {
+    pFunc->node.resType = (SDataType){.bytes = tDataTypes[TSDB_DATA_TYPE_BIGINT].bytes, .type = TSDB_DATA_TYPE_BIGINT};
+  }
+
+  return TSDB_CODE_SUCCESS;
+}
+
+static int32_t translateHLLPartial(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
+  return translateHLLImpl(pFunc, pErrBuf, len, true);
+}
+
+static int32_t translateHLLMerge(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
+  return translateHLLImpl(pFunc, pErrBuf, len, false);
+}
+
 static bool validateStateOper(const SValueNode* pVal) {
   if (TSDB_DATA_TYPE_BINARY != pVal->node.resType.type) {
     return false;
@@ -1478,6 +1503,28 @@ const SBuiltinFuncDefinition funcMgtBuiltins[] = {
     .getEnvFunc   = getHLLFuncEnv,
     .initFunc     = functionSetup,
     .processFunc  = hllFunction,
+    .finalizeFunc = hllFinalize,
+    .pPartialFunc = "_hyperloglog_partial",
+    .pMergeFunc   = "_hyperloglog_merge"
+  },
+  {
+    .name = "_hyperloglog_partial",
+    .type = FUNCTION_TYPE_HYPERLOGLOG_PARTIAL,
+    .classification = FUNC_MGT_AGG_FUNC,
+    .translateFunc = translateHLLPartial,
+    .getEnvFunc   = getHLLFuncEnv,
+    .initFunc     = functionSetup,
+    .processFunc  = hllFunction,
+    .finalizeFunc = hllPartialFinalize
+  },
+  {
+    .name = "_hyperloglog_merge",
+    .type = FUNCTION_TYPE_HYPERLOGLOG_MERGE,
+    .classification = FUNC_MGT_AGG_FUNC,
+    .translateFunc = translateHLLMerge,
+    .getEnvFunc   = getHLLFuncEnv,
+    .initFunc     = functionSetup,
+    .processFunc  = hllFunctionMerge,
     .finalizeFunc = hllFinalize
   },
   {

source/libs/function/src/builtinsimpl.c

@@ -3411,7 +3411,6 @@ int32_t histogramFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
 }
 
 int32_t histogramPartialFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
-  SResultRowEntryInfo* pResInfo = GET_RES_INFO(pCtx);
   SHistoFuncInfo* pInfo = GET_ROWCELL_INTERBUF(GET_RES_INFO(pCtx));
   int32_t resultBytes = getHistogramInfoSize();
   char *res = taosMemoryCalloc(resultBytes + VARSTR_HEADER_SIZE, sizeof(char));
@@ -3428,6 +3427,10 @@ int32_t histogramPartialFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
   return 1;
 }
 
+int32_t getHLLInfoSize() {
+  return (int32_t)sizeof(SHLLInfo);
+}
+
 bool getHLLFuncEnv(SFunctionNode* UNUSED_PARAM(pFunc), SFuncExecEnv* pEnv) {
   pEnv->calcMemSize = sizeof(SHLLInfo);
   return true;
@@ -3553,6 +3556,27 @@ int32_t hllFunction(SqlFunctionCtx *pCtx) {
   return TSDB_CODE_SUCCESS;
 }
 
+int32_t hllFunctionMerge(SqlFunctionCtx *pCtx) {
+  SInputColumnInfoData* pInput = &pCtx->input;
+  SColumnInfoData* pCol = pInput->pData[0];
+  ASSERT(pCol->info.type == TSDB_DATA_TYPE_BINARY);
+
+  SHLLInfo* pInfo = GET_ROWCELL_INTERBUF(GET_RES_INFO(pCtx));
+
+  int32_t start = pInput->startRowIndex;
+  char* data = colDataGetData(pCol, start);
+  SHLLInfo* pInputInfo = (SHLLInfo *)varDataVal(data);
+
+  for (int32_t k = 0; k < HLL_BUCKETS; ++k) {
+    if (pInfo->buckets[k] < pInputInfo->buckets[k]) {
+      pInfo->buckets[k] = pInputInfo->buckets[k];
+    }
+  }
+
+  SET_VAL(GET_RES_INFO(pCtx), 1, 1);
+  return TSDB_CODE_SUCCESS;
+}
+
 int32_t hllFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
   SResultRowEntryInfo *pInfo = GET_RES_INFO(pCtx);
 
@@ -3565,6 +3589,24 @@ int32_t hllFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
   return functionFinalize(pCtx, pBlock);
 }
 
+int32_t hllPartialFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock) {
+  SResultRowEntryInfo* pResInfo = GET_RES_INFO(pCtx);
+  SHLLInfo* pInfo = GET_ROWCELL_INTERBUF(GET_RES_INFO(pCtx));
+  int32_t resultBytes = getHLLInfoSize();
+  char *res = taosMemoryCalloc(resultBytes + VARSTR_HEADER_SIZE, sizeof(char));
+
+  memcpy(varDataVal(res), pInfo, resultBytes);
+  varDataSetLen(res, resultBytes);
+
+  int32_t          slotId = pCtx->pExpr->base.resSchema.slotId;
+  SColumnInfoData* pCol = taosArrayGet(pBlock->pDataBlock, slotId);
+
+  colDataAppend(pCol, pBlock->info.rows, res, false);
+
+  taosMemoryFree(res);
+  return pResInfo->numOfRes;
+}
+
 bool getStateFuncEnv(SFunctionNode* UNUSED_PARAM(pFunc), SFuncExecEnv* pEnv) {
   pEnv->calcMemSize = sizeof(SStateInfo);
   return true;

tests/system-test/fulltest.bat

@@ -3,7 +3,7 @@ python3 .\test.py -f 0-others\taosShell.py
 python3 .\test.py -f 0-others\taosShellError.py
 python3 .\test.py -f 0-others\taosShellNetChk.py
 python3 .\test.py -f 0-others\telemetry.py
-python3 .\test.py -f 0-others\taosdMonitor.py
+@REM python3 .\test.py -f 0-others\taosdMonitor.py
 python3 .\test.py -f 0-others\udfTest.py
 python3 .\test.py -f 0-others\udf_create.py
 python3 .\test.py -f 0-others\udf_restart_taosd.py
@@ -100,4 +100,4 @@ python3 .\test.py -f 0-others\udf_restart_taosd.py
 @REM python3 .\test.py -f 7-tmq\subscribeStb2.py
 @REM python3 .\test.py -f 7-tmq\subscribeStb3.py
 @REM python3 .\test.py -f 7-tmq\subscribeStb4.py
-@REM python3 .\test.py -f 7-tmq\db.py
\ No newline at end of file
+@REM python3 .\test.py -f 7-tmq\db.py