This is an automated email from the ASF dual-hosted git repository.
dkuzmenko pushed a commit to branch branch-4.1
in repository https://gitbox.apache.org/repos/asf/hive.git
The following commit(s) were added to refs/heads/branch-4.1 by this push:
new 8583b373e70 HIVE-29080: Fix Incorrect results for queries on Decimal
column Partitioned table with Vectorization enabled (#5965)
8583b373e70 is described below
commit 8583b373e7026ae284cab92e0cd4472536a38ab9
Author: Butao Zhang <zhangbu...@apache.org>
AuthorDate: Sun Jul 13 02:54:27 2025 +0800
HIVE-29080: Fix Incorrect results for queries on Decimal column Partitioned
table with Vectorization enabled (#5965)
Co-authored-by: Indhumathi <in...@visa.com>
---
.../hive/ql/exec/vector/VectorizedRowBatchCtx.java | 2 +-
.../clientpositive/vector_decimal_partition.q | 15 ++++-
.../llap/vector_decimal_partition.q.out | 76 ++++++++++++++++++++++
.../hive/ql/exec/vector/Decimal64ColumnVector.java | 7 ++
4 files changed, 98 insertions(+), 2 deletions(-)
diff --git
a/ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatchCtx.java
b/ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatchCtx.java
index 0abbd59b9e9..a0906cfb033 100644
---
a/ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatchCtx.java
+++
b/ql/src/java/org/apache/hadoop/hive/ql/exec/vector/VectorizedRowBatchCtx.java
@@ -576,7 +576,7 @@ public void addPartitionColsToBatch(ColumnVector col,
Object value, int colIndex
dv.isNull[0] = true;
dv.isRepeating = true;
} else {
- dv.fill(((HiveDecimal) value).longValue());
+ dv.fill((HiveDecimal) value);
}
} else {
DecimalColumnVector dv = (DecimalColumnVector) col;
diff --git a/ql/src/test/queries/clientpositive/vector_decimal_partition.q
b/ql/src/test/queries/clientpositive/vector_decimal_partition.q
index dc8e667a47c..2cb4f4e15fd 100644
--- a/ql/src/test/queries/clientpositive/vector_decimal_partition.q
+++ b/ql/src/test/queries/clientpositive/vector_decimal_partition.q
@@ -13,4 +13,17 @@ INSERT INTO decimal_part PARTITION (nr_bank = 88) VALUES (1,
'test');
INSERT INTO decimal_part PARTITION (nr_bank = 8801) VALUES (1, '8801');
EXPLAIN VECTORIZATION EXPRESSION SELECT count(*), nr_bank FROM decimal_part
GROUP BY nr_bank;
-SELECT count(*), nr_bank FROM decimal_part GROUP BY nr_bank;
\ No newline at end of file
+SELECT count(*), nr_bank FROM decimal_part GROUP BY nr_bank;
+
+DROP TABLE IF EXISTS decimal_part1;
+
+CREATE EXTERNAL TABLE decimal_part1 (quantity INT) PARTITIONED BY
(sales_percent DECIMAL(10,2)) STORED AS ORC;
+INSERT INTO decimal_part1 VALUES (1, 24518.01);
+INSERT INTO decimal_part1 VALUES (2, 24518.02);
+
+set hive.auto.convert.join=true;
+
+SELECT count(*), sales_percent FROM decimal_part1 GROUP BY sales_percent;
+SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1 JOIN decimal_part1
d2 ON d1.sales_percent=d2.sales_percent;
+SET hive.vectorized.execution.enabled=false;
+SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1 JOIN decimal_part1
d2 ON d1.sales_percent=d2.sales_percent;
diff --git
a/ql/src/test/results/clientpositive/llap/vector_decimal_partition.q.out
b/ql/src/test/results/clientpositive/llap/vector_decimal_partition.q.out
index 83e7d2a0502..7d1651b1077 100644
--- a/ql/src/test/results/clientpositive/llap/vector_decimal_partition.q.out
+++ b/ql/src/test/results/clientpositive/llap/vector_decimal_partition.q.out
@@ -178,3 +178,79 @@ POSTHOOK: Input: default@decimal_part@nr_bank=8801
#### A masked pattern was here ####
1 88
1 8801
+PREHOOK: query: DROP TABLE IF EXISTS decimal_part1
+PREHOOK: type: DROPTABLE
+PREHOOK: Output: database:default
+POSTHOOK: query: DROP TABLE IF EXISTS decimal_part1
+POSTHOOK: type: DROPTABLE
+POSTHOOK: Output: database:default
+PREHOOK: query: CREATE EXTERNAL TABLE decimal_part1 (quantity INT) PARTITIONED
BY (sales_percent DECIMAL(10,2)) STORED AS ORC
+PREHOOK: type: CREATETABLE
+PREHOOK: Output: database:default
+PREHOOK: Output: default@decimal_part1
+POSTHOOK: query: CREATE EXTERNAL TABLE decimal_part1 (quantity INT)
PARTITIONED BY (sales_percent DECIMAL(10,2)) STORED AS ORC
+POSTHOOK: type: CREATETABLE
+POSTHOOK: Output: database:default
+POSTHOOK: Output: default@decimal_part1
+PREHOOK: query: INSERT INTO decimal_part1 VALUES (1, 24518.01)
+PREHOOK: type: QUERY
+PREHOOK: Input: _dummy_database@_dummy_table
+PREHOOK: Output: default@decimal_part1
+POSTHOOK: query: INSERT INTO decimal_part1 VALUES (1, 24518.01)
+POSTHOOK: type: QUERY
+POSTHOOK: Input: _dummy_database@_dummy_table
+POSTHOOK: Output: default@decimal_part1
+POSTHOOK: Output: default@decimal_part1@sales_percent=24518.01
+POSTHOOK: Lineage: decimal_part1 PARTITION(sales_percent=24518.01).quantity
SCRIPT []
+PREHOOK: query: INSERT INTO decimal_part1 VALUES (2, 24518.02)
+PREHOOK: type: QUERY
+PREHOOK: Input: _dummy_database@_dummy_table
+PREHOOK: Output: default@decimal_part1
+POSTHOOK: query: INSERT INTO decimal_part1 VALUES (2, 24518.02)
+POSTHOOK: type: QUERY
+POSTHOOK: Input: _dummy_database@_dummy_table
+POSTHOOK: Output: default@decimal_part1
+POSTHOOK: Output: default@decimal_part1@sales_percent=24518.02
+POSTHOOK: Lineage: decimal_part1 PARTITION(sales_percent=24518.02).quantity
SCRIPT []
+PREHOOK: query: SELECT count(*), sales_percent FROM decimal_part1 GROUP BY
sales_percent
+PREHOOK: type: QUERY
+PREHOOK: Input: default@decimal_part1
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.01
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+POSTHOOK: query: SELECT count(*), sales_percent FROM decimal_part1 GROUP BY
sales_percent
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@decimal_part1
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.01
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+1 24518.02
+1 24518.01
+PREHOOK: query: SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1 JOIN
decimal_part1 d2 ON d1.sales_percent=d2.sales_percent
+PREHOOK: type: QUERY
+PREHOOK: Input: default@decimal_part1
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.01
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+POSTHOOK: query: SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1
JOIN decimal_part1 d2 ON d1.sales_percent=d2.sales_percent
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@decimal_part1
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.01
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+1 24518.01
+2 24518.02
+PREHOOK: query: SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1 JOIN
decimal_part1 d2 ON d1.sales_percent=d2.sales_percent
+PREHOOK: type: QUERY
+PREHOOK: Input: default@decimal_part1
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.01
+PREHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+POSTHOOK: query: SELECT d1.quantity,d1.sales_percent FROM decimal_part1 d1
JOIN decimal_part1 d2 ON d1.sales_percent=d2.sales_percent
+POSTHOOK: type: QUERY
+POSTHOOK: Input: default@decimal_part1
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.01
+POSTHOOK: Input: default@decimal_part1@sales_percent=24518.02
+#### A masked pattern was here ####
+1 24518.01
+2 24518.02
diff --git
a/storage-api/src/java/org/apache/hadoop/hive/ql/exec/vector/Decimal64ColumnVector.java
b/storage-api/src/java/org/apache/hadoop/hive/ql/exec/vector/Decimal64ColumnVector.java
index a06ce39b46c..3753d337d8b 100644
---
a/storage-api/src/java/org/apache/hadoop/hive/ql/exec/vector/Decimal64ColumnVector.java
+++
b/storage-api/src/java/org/apache/hadoop/hive/ql/exec/vector/Decimal64ColumnVector.java
@@ -41,6 +41,13 @@ public Decimal64ColumnVector(int size, int precision, int
scale) {
scratchHiveDecWritable = new HiveDecimalWritable();
}
+ // Fill the vector entries with provided value
+ public void fill(HiveDecimal value) {
+ isRepeating = true;
+ isNull[0] = false;
+ set(0, value);
+ }
+
/**
* Set a Decimal64 field from a HiveDecimalWritable.
*
