This is an automated email from the ASF dual-hosted git repository.
comphead 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 6f1ca298fe fix: metadata of join schema (#16221)
6f1ca298fe is described below
commit 6f1ca298fe120328a75371e074fd52bf5936b725
Author: Chen Chongchen <chenkov...@qq.com>
AuthorDate: Tue Jun 3 03:04:28 2025 +0800
fix: metadata of join schema (#16221)
---
datafusion/expr/src/logical_plan/builder.rs | 38 +++++++++++++++++++++++++++--
datafusion/physical-plan/src/joins/utils.rs | 35 ++++++++++++++++++++++++--
2 files changed, 69 insertions(+), 4 deletions(-)
diff --git a/datafusion/expr/src/logical_plan/builder.rs
b/datafusion/expr/src/logical_plan/builder.rs
index fabdfecef1..f75e79cd66 100644
--- a/datafusion/expr/src/logical_plan/builder.rs
+++ b/datafusion/expr/src/logical_plan/builder.rs
@@ -1626,12 +1626,19 @@ pub fn build_join_schema(
join_type,
left.fields().len(),
);
- let metadata = left
+
+ let (schema1, schema2) = match join_type {
+ JoinType::Right | JoinType::RightSemi | JoinType::RightAnti => (left,
right),
+ _ => (right, left),
+ };
+
+ let metadata = schema1
.metadata()
.clone()
.into_iter()
- .chain(right.metadata().clone())
+ .chain(schema2.metadata().clone())
.collect();
+
let dfschema = DFSchema::new_with_metadata(qualified_fields, metadata)?;
dfschema.with_functional_dependencies(func_dependencies)
}
@@ -2870,4 +2877,31 @@ mod tests {
Ok(())
}
+
+ #[test]
+ fn test_join_metadata() -> Result<()> {
+ let left_schema = DFSchema::new_with_metadata(
+ vec![(None, Arc::new(Field::new("a", DataType::Int32, false)))],
+ HashMap::from([("key".to_string(), "left".to_string())]),
+ )?;
+ let right_schema = DFSchema::new_with_metadata(
+ vec![(None, Arc::new(Field::new("b", DataType::Int32, false)))],
+ HashMap::from([("key".to_string(), "right".to_string())]),
+ )?;
+
+ let join_schema =
+ build_join_schema(&left_schema, &right_schema, &JoinType::Left)?;
+ assert_eq!(
+ join_schema.metadata(),
+ &HashMap::from([("key".to_string(), "left".to_string())])
+ );
+ let join_schema =
+ build_join_schema(&left_schema, &right_schema, &JoinType::Right)?;
+ assert_eq!(
+ join_schema.metadata(),
+ &HashMap::from([("key".to_string(), "right".to_string())])
+ );
+
+ Ok(())
+ }
}
diff --git a/datafusion/physical-plan/src/joins/utils.rs
b/datafusion/physical-plan/src/joins/utils.rs
index 4be1437424..3abeff6621 100644
--- a/datafusion/physical-plan/src/joins/utils.rs
+++ b/datafusion/physical-plan/src/joins/utils.rs
@@ -314,12 +314,18 @@ pub fn build_join_schema(
JoinType::RightSemi | JoinType::RightAnti => right_fields().unzip(),
};
- let metadata = left
+ let (schema1, schema2) = match join_type {
+ JoinType::Right | JoinType::RightSemi | JoinType::RightAnti => (left,
right),
+ _ => (right, left),
+ };
+
+ let metadata = schema1
.metadata()
.clone()
.into_iter()
- .chain(right.metadata().clone())
+ .chain(schema2.metadata().clone())
.collect();
+
(fields.finish().with_metadata(metadata), column_indices)
}
@@ -1498,6 +1504,7 @@ pub(super) fn swap_join_projection(
#[cfg(test)]
mod tests {
use super::*;
+ use std::collections::HashMap;
use std::pin::Pin;
use arrow::array::Int32Array;
@@ -2495,4 +2502,28 @@ mod tests {
assert_eq!(col.name(), name);
assert_eq!(col.index(), index);
}
+
+ #[test]
+ fn test_join_metadata() -> Result<()> {
+ let left_schema = Schema::new(vec![Field::new("a", DataType::Int32,
false)])
+ .with_metadata(HashMap::from([("key".to_string(),
"left".to_string())]));
+
+ let right_schema = Schema::new(vec![Field::new("b", DataType::Int32,
false)])
+ .with_metadata(HashMap::from([("key".to_string(),
"right".to_string())]));
+
+ let (join_schema, _) =
+ build_join_schema(&left_schema, &right_schema, &JoinType::Left);
+ assert_eq!(
+ join_schema.metadata(),
+ &HashMap::from([("key".to_string(), "left".to_string())])
+ );
+ let (join_schema, _) =
+ build_join_schema(&left_schema, &right_schema, &JoinType::Right);
+ assert_eq!(
+ join_schema.metadata(),
+ &HashMap::from([("key".to_string(), "right".to_string())])
+ );
+
+ Ok(())
+ }
}
---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscr...@datafusion.apache.org
For additional commands, e-mail: commits-h...@datafusion.apache.org
