Dandandan commented on a change in pull request #291:
URL: https://github.com/apache/arrow-datafusion/pull/291#discussion_r628887212
##########
File path: datafusion/src/physical_plan/hash_join.rs
##########
@@ -1410,6 +1413,45 @@ mod tests {
assert_batches_sorted_eq!(expected, &batches);
}
+ #[tokio::test]
+ async fn join_outer_multi_batch() {
+ let left = build_table(
+ ("a1", &vec![1, 2, 3]),
+ ("b1", &vec![4, 5, 7]), // 7 does not exist on the right
+ ("c1", &vec![7, 8, 9]),
+ );
+ let right = build_table_two_batches(
+ ("a2", &vec![10, 20, 30]),
+ ("b1", &vec![4, 5, 6]),
+ ("c2", &vec![70, 80, 90]),
+ );
+ let on = &[("b1", "b1")];
+
+ let join = join(left, right, on, &JoinType::Full).unwrap();
+
+ let columns = columns(&join.schema());
+ assert_eq!(columns, vec!["a1", "b1", "c1", "a2", "c2"]);
+
+ let stream = join.execute(0).await.unwrap();
+ let batches = common::collect(stream).await.unwrap();
+
+ let expected = vec![
Review comment:
To me the results look correct given two equivalent right batches. I
included `b2` as column here as well to make it a bit more obvious:
```
"+----+----+----+----+----+----+",
"| a1 | b1 | c1 | a2 | b2 | c2 |",
"+----+----+----+----+----+----+",
"| | | | 30 | 6 | 90 |",
"| | | | 30 | 6 | 90 |",
"| 1 | 4 | 7 | 10 | 4 | 70 |",
"| 1 | 4 | 7 | 10 | 4 | 70 |",
"| 2 | 5 | 8 | 20 | 5 | 80 |",
"| 2 | 5 | 8 | 20 | 5 | 80 |",
"| 3 | 7 | 9 | | | |",
"+----+----+----+----+----+----+",
```
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
