From 13d4b5d4e8fdfc85c503201de13aa7fc9b74903a Mon Sep 17 00:00:00 2001
From: ziyi chen <chenziyi990424@gmail.com>
Date: Tue, 23 Apr 2024 14:19:27 +0800
Subject: [PATCH 1/2] Fix distinct hash table performance bug

---
 src/processor/operator/aggregate/aggregate_hash_table.cpp | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/src/processor/operator/aggregate/aggregate_hash_table.cpp b/src/processor/operator/aggregate/aggregate_hash_table.cpp
index 51ddac8a72..418ce77d6f 100644
--- a/src/processor/operator/aggregate/aggregate_hash_table.cpp
+++ b/src/processor/operator/aggregate/aggregate_hash_table.cpp
@@ -1,5 +1,7 @@
 #include "processor/operator/aggregate/aggregate_hash_table.h"
 
+#include <chrono>
+
 #include "common/utils.h"
 
 using namespace kuzu::common;
@@ -49,7 +51,7 @@ bool AggregateHashTable::isAggregateValueDistinctForGroupByKeys(
     }
     distinctKeyVectors[groupByFlatKeyVectors.size()] = aggregateVector;
     computeVectorHashes(distinctKeyVectors, std::vector<ValueVector*>() /* unFlatKeyVectors */);
-    hash_t hash = hashVector->getValue<hash_t>(hashVector->state->selVector->selectedPositions[0]);
+    auto hash = hashVector->getValue<hash_t>(hashVector->state->selVector->selectedPositions[0]);
     auto distinctHTEntry = findEntryInDistinctHT(distinctKeyVectors, hash);
     if (distinctHTEntry == nullptr) {
         resizeHashTableIfNecessary(1);
@@ -378,6 +380,7 @@ uint8_t* AggregateHashTable::createEntryInDistinctHT(
         factorizedTable->updateFlatCell(entry, i, groupByHashKeyVectors[i],
             groupByHashKeyVectors[i]->state->selVector->selectedPositions[0]);
     }
+    factorizedTable->updateFlatCellNoNull(entry, hashColIdxInFT, &hash);
     fillEntryWithInitialNullAggregateState(entry);
     fillHashSlot(hash, entry);
     return entry;

From 98a80a18aaf3e47a0624de917c0ac81eea88164c Mon Sep 17 00:00:00 2001
From: ziyi chen <chenziyi990424@gmail.com>
Date: Tue, 23 Apr 2024 14:20:29 +0800
Subject: [PATCH 2/2] update

---
 src/processor/operator/aggregate/aggregate_hash_table.cpp | 2 --
 1 file changed, 2 deletions(-)

diff --git a/src/processor/operator/aggregate/aggregate_hash_table.cpp b/src/processor/operator/aggregate/aggregate_hash_table.cpp
index 418ce77d6f..a0a45d4d81 100644
--- a/src/processor/operator/aggregate/aggregate_hash_table.cpp
+++ b/src/processor/operator/aggregate/aggregate_hash_table.cpp
@@ -1,7 +1,5 @@
 #include "processor/operator/aggregate/aggregate_hash_table.h"
 
-#include <chrono>
-
 #include "common/utils.h"
 
 using namespace kuzu::common;