[SPARK-52311][SQL] Redefine UnionLoop output to not be duplicated if the anchor output is duplicated #51041
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…the anchor output is duplicated
cloud-fan
reviewed
May 29, 2025
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/plans/logical/cteOperators.scala
Outdated
Show resolved
Hide resolved
cloud-fan
reviewed
May 29, 2025
| cteDef.id, | ||
| anchor, | ||
| rewriteRecursiveCTERefs(recursion, anchor, cteDef.id, None), | ||
| anchor.output.map(_.newInstance().exprId), |
Contributor
There was a problem hiding this comment.
nit: we can add a new apply method in object UnionLoop, which takes the same parameters as before, and pass anchor.output.map(_.newInstance().exprId) additionally to construct UnionLoop
Contributor
Author
There was a problem hiding this comment.
just using apply in object UnionLoop doesn't work because of default arguments.
in object UnionLoop, multiple overloaded alternatives of method apply define default arguments.
There are other possibilities to make this work, but I don't think any one change can work without modifying something in ResolveWithCTE.scala
cloud-fan
approved these changes
May 29, 2025
…s/logical/cteOperators.scala Co-authored-by: Wenchen Fan <[email protected]>
Contributor
|
thanks, merging to master! |
yhuang-db
pushed a commit
to yhuang-db/spark
that referenced
this pull request
Jun 9, 2025
…the anchor output is duplicated ### What changes were proposed in this pull request? Redefine the output of the UnionLoop and UnionLoopRef to be with new expression IDs. ### Why are the changes needed? Currently rCTEs don't behave properly in case when the anchor references the same column multiple times in the anchor, leading to wrong things being identified in the recursion. For example this rCTE: ``` WITH RECURSIVE tmp(x) AS ( values (1), (2), (3), (4), (5) ), rcte(x, y) AS ( SELECT x, x FROM tmp WHERE x = 1 UNION ALL SELECT x + 1, x FROM rcte WHERE x < 5 ) SELECT * FROM rcte; ``` Will return: ``` 1 1 2 2 3 3 4 4 5 5 ``` Instead of: ``` 1 1 2 1 3 2 4 3 5 4 ``` ### Does this PR introduce _any_ user-facing change? No. ### How was this patch tested? New tests in golden file cte-recursion.sql. ### Was this patch authored or co-authored using generative AI tooling? No. Closes apache#51041 from Pajaraja/pavle-martinovic_data/UnionLoopOutput. Lead-authored-by: pavle-martinovic_data <[email protected]> Co-authored-by: Pavle Martinovic <[email protected]> Signed-off-by: Wenchen Fan <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What changes were proposed in this pull request?
Redefine the output of the UnionLoop and UnionLoopRef to be with new expression IDs.
Why are the changes needed?
Currently rCTEs don't behave properly in case when the anchor references the same column multiple times in the anchor, leading to wrong things being identified in the recursion.
For example this rCTE:
Will return:
Instead of:
Does this PR introduce any user-facing change?
No.
How was this patch tested?
New tests in golden file cte-recursion.sql.
Was this patch authored or co-authored using generative AI tooling?
No.