From 267ed293d7ba3b20770e71c17fdb7add26187b75 Mon Sep 17 00:00:00 2001
From: Ganlin Zhao <ganlinzhao@hotmail.com>
Date: Mon, 16 May 2022 22:43:06 +0800
Subject: [PATCH] feat(query): add HYPERLOGLOG function

---
 .gitmodules                             |  6 +++---
 include/libs/function/functionMgt.h     |  1 +
 source/libs/function/inc/builtinsimpl.h |  4 ++++
 source/libs/function/src/builtins.c     | 20 ++++++++++++++++++++
 source/libs/function/src/builtinsimpl.c |  2 +-
 tools/taos-tools                        |  2 +-
 6 files changed, 30 insertions(+), 5 deletions(-)

diff --git a/.gitmodules b/.gitmodules
index bc38453f19..9419dcd100 100644
--- a/.gitmodules
+++ b/.gitmodules
@@ -13,6 +13,6 @@
 [submodule "examples/rust"]
 	path = examples/rust
 	url = https://github.com/songtianyi/tdengine-rust-bindings.git
-[submodule "tools/taos-tools"]
-	path = tools/taos-tools
-	url = https://github.com/taosdata/taos-tools
+#[submodule "tools/taos-tools"]
+    #path = tools/taos-tools
+    #url = https://github.com/taosdata/taos-tools
diff --git a/include/libs/function/functionMgt.h b/include/libs/function/functionMgt.h
index 2b58ed7c0b..aec1476663 100644
--- a/include/libs/function/functionMgt.h
+++ b/include/libs/function/functionMgt.h
@@ -41,6 +41,7 @@ typedef enum EFunctionType {
   FUNCTION_TYPE_SUM,
   FUNCTION_TYPE_TWA,
   FUNCTION_TYPE_HISTOGRAM,
+  FUNCTION_TYPE_HYPERLOGLOG,
 
   // nonstandard SQL function
   FUNCTION_TYPE_BOTTOM = 500,
diff --git a/source/libs/function/inc/builtinsimpl.h b/source/libs/function/inc/builtinsimpl.h
index c25d74911c..99313675a5 100644
--- a/source/libs/function/inc/builtinsimpl.h
+++ b/source/libs/function/inc/builtinsimpl.h
@@ -90,6 +90,10 @@ bool histogramFunctionSetup(SqlFunctionCtx *pCtx, SResultRowEntryInfo* pResultIn
 int32_t histogramFunction(SqlFunctionCtx* pCtx);
 int32_t histogramFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock);
 
+bool getHLLFuncEnv(struct SFunctionNode* pFunc, SFuncExecEnv* pEnv);
+int32_t hllFunction(SqlFunctionCtx* pCtx);
+int32_t hllFinalize(SqlFunctionCtx* pCtx, SSDataBlock* pBlock);
+
 bool getStateFuncEnv(struct SFunctionNode* pFunc, SFuncExecEnv* pEnv);
 bool stateFunctionSetup(SqlFunctionCtx *pCtx, SResultRowEntryInfo* pResultInfo);
 int32_t stateCountFunction(SqlFunctionCtx* pCtx);
diff --git a/source/libs/function/src/builtins.c b/source/libs/function/src/builtins.c
index e41e3c7c39..cc50348397 100644
--- a/source/libs/function/src/builtins.c
+++ b/source/libs/function/src/builtins.c
@@ -263,6 +263,16 @@ static int32_t translateHistogram(SFunctionNode* pFunc, char* pErrBuf, int32_t l
   return TSDB_CODE_SUCCESS;
 }
 
+static int32_t translateHLL(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
+  if (1 != LIST_LENGTH(pFunc->pParameterList)) {
+    return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName);
+  }
+
+  pFunc->node.resType = (SDataType){.bytes = tDataTypes[TSDB_DATA_TYPE_BIGINT].bytes, .type = TSDB_DATA_TYPE_BIGINT};
+  return TSDB_CODE_SUCCESS;
+}
+
+
 static int32_t translateStateCount(SFunctionNode* pFunc, char* pErrBuf, int32_t len) {
   if (3 != LIST_LENGTH(pFunc->pParameterList)) {
     return invaildFuncParaNumErrMsg(pErrBuf, len, pFunc->functionName);
@@ -829,6 +839,16 @@ const SBuiltinFuncDefinition funcMgtBuiltins[] = {
     .processFunc  = histogramFunction,
     .finalizeFunc = histogramFinalize
   },
+  {
+    .name = "hyperloglog",
+    .type = FUNCTION_TYPE_HYPERLOGLOG,
+    .classification = FUNC_MGT_AGG_FUNC,
+    .translateFunc = translateHLL,
+    .getEnvFunc   = getHLLFuncEnv,
+    .initFunc     = functionSetup,
+    .processFunc  = hllFunction,
+    .finalizeFunc = hllFinalize
+  },
   {
     .name = "state_count",
     .type = FUNCTION_TYPE_STATE_COUNT,
diff --git a/source/libs/function/src/builtinsimpl.c b/source/libs/function/src/builtinsimpl.c
index 556015a2ac..de63567011 100644
--- a/source/libs/function/src/builtinsimpl.c
+++ b/source/libs/function/src/builtinsimpl.c
@@ -2745,7 +2745,7 @@ bool getHLLFuncEnv(SFunctionNode* UNUSED_PARAM(pFunc), SFuncExecEnv* pEnv) {
   return true;
 }
 
-static uint8_t hllCountNum(void* data, int32_t bytes, int32_t *buk) {
+static uint8_t hllCountNum(char *data, int32_t bytes, int32_t *buk) {
   uint64_t hash = MurmurHash3_64(data, bytes);
   int32_t index = hash & HLL_BUCKET_MASK;
   hash >>= HLL_BUCKET_BITS;
diff --git a/tools/taos-tools b/tools/taos-tools
index 0aad27d725..2f3dfddd4d 160000
--- a/tools/taos-tools
+++ b/tools/taos-tools
@@ -1 +1 @@
-Subproject commit 0aad27d725f4ee6b18daf1db0c07d933aed16eea
+Subproject commit 2f3dfddd4d9a869e706ba3cf98fb6d769404cd7c