This is an automated email from the ASF dual-hosted git repository.
dheres pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/datafusion.git
The following commit(s) were added to refs/heads/main by this push:
new f044bc8371 Fix bug that `COUNT(DISTINCT)` on StringView panics
(#11768)
f044bc8371 is described below
commit f044bc8371d5b4e1e51a9026f3eccac16a6d4648
Author: Xiangpeng Hao <[email protected]>
AuthorDate: Fri Aug 2 04:33:05 2024 +0800
Fix bug that `COUNT(DISTINCT)` on StringView panics (#11768)
* fix bug
* Add test showing panic on string view
---------
Co-authored-by: Andrew Lamb <[email protected]>
---
datafusion/functions-aggregate/src/count.rs | 5 +++-
datafusion/sqllogictest/test_files/string_view.slt | 34 ++++++++++++++++++++++
2 files changed, 38 insertions(+), 1 deletion(-)
diff --git a/datafusion/functions-aggregate/src/count.rs
b/datafusion/functions-aggregate/src/count.rs
index e2d59003fc..64eb7253f5 100644
--- a/datafusion/functions-aggregate/src/count.rs
+++ b/datafusion/functions-aggregate/src/count.rs
@@ -237,7 +237,7 @@ impl AggregateUDFImpl for Count {
Box::new(BytesDistinctCountAccumulator::<i32>::new(OutputType::Utf8))
}
DataType::Utf8View => {
-
Box::new(BytesViewDistinctCountAccumulator::new(OutputType::Utf8))
+
Box::new(BytesViewDistinctCountAccumulator::new(OutputType::Utf8View))
}
DataType::LargeUtf8 => {
Box::new(BytesDistinctCountAccumulator::<i64>::new(OutputType::Utf8))
@@ -245,6 +245,9 @@ impl AggregateUDFImpl for Count {
DataType::Binary =>
Box::new(BytesDistinctCountAccumulator::<i32>::new(
OutputType::Binary,
)),
+ DataType::BinaryView =>
Box::new(BytesViewDistinctCountAccumulator::new(
+ OutputType::BinaryView,
+ )),
DataType::LargeBinary =>
Box::new(BytesDistinctCountAccumulator::<i64>::new(
OutputType::Binary,
)),
diff --git a/datafusion/sqllogictest/test_files/string_view.slt
b/datafusion/sqllogictest/test_files/string_view.slt
index 3f9a4793f6..763b4e99c6 100644
--- a/datafusion/sqllogictest/test_files/string_view.slt
+++ b/datafusion/sqllogictest/test_files/string_view.slt
@@ -321,6 +321,40 @@ logical_plan
02)--Filter: CAST(test.column2_utf8 AS Utf8View) = test.column1_utf8view
03)----TableScan: test projection=[column1_utf8, column2_utf8,
column1_utf8view]
+## Test distinct aggregates
+query III
+SELECT
+ COUNT(DISTINCT column1_utf8),
+ COUNT(DISTINCT column1_utf8view),
+ COUNT(DISTINCT column1_dict)
+FROM test;
+----
+3 3 3
+
+query III
+SELECT
+ COUNT(DISTINCT column1_utf8),
+ COUNT(DISTINCT column1_utf8view),
+ COUNT(DISTINCT column1_dict)
+FROM test
+GROUP BY column2_utf8view;
+----
+1 1 1
+1 1 1
+1 1 1
+
+
+query TT
+EXPLAIN SELECT
+ COUNT(DISTINCT column1_utf8),
+ COUNT(DISTINCT column1_utf8view),
+ COUNT(DISTINCT column1_dict)
+FROM test;
+----
+logical_plan
+01)Aggregate: groupBy=[[]], aggr=[[count(DISTINCT test.column1_utf8),
count(DISTINCT test.column1_utf8view), count(DISTINCT test.column1_dict)]]
+02)--TableScan: test projection=[column1_utf8, column1_utf8view, column1_dict]
+
statement ok
drop table test;
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
