Dandandan commented on a change in pull request #291:
URL: https://github.com/apache/arrow-datafusion/pull/291#discussion_r628876661
##########
File path: datafusion/src/physical_plan/hash_join.rs
##########
@@ -1515,6 +1589,43 @@ mod tests {
Ok(())
}
+ #[tokio::test]
+ async fn join_outer_one() -> Result<()> {
+ 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(
+ ("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)?;
+
+ let columns = columns(&join.schema());
+ assert_eq!(columns, vec!["a1", "b1", "c1", "a2", "c2"]);
+
+ let stream = join.execute(0).await?;
+ let batches = common::collect(stream).await?;
+
+ let expected = vec![
+ "+----+----+----+----+----+",
+ "| a1 | b1 | c1 | a2 | c2 |",
+ "+----+----+----+----+----+",
+ "| | | | 30 | 90 |",
Review comment:
The `b1` column is here from the left side. I can make the test a bit
easier to follow.
I believe this should be covered by the work that @houqp is working on to
support qualified names in tables (and disambiguating using the names).
--
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]
