rgyhuang opened a new pull request, #56249:
URL: https://github.com/apache/spark/pull/56249
<!--
Thanks for sending a pull request! Here are some tips for you:
1. If this is your first time, please read our contributor guidelines:
https://spark.apache.org/contributing.html
2. Ensure you have added or run the appropriate tests for your PR:
https://spark.apache.org/developer-tools.html
3. If the PR is unfinished, add '[WIP]' in your PR title, e.g.,
'[WIP][SPARK-XXXX] Your PR title ...'.
4. Be sure to keep the PR description updated to reflect all changes.
5. Please write your PR title to summarize what this PR proposes.
6. If possible, provide a concise example to reproduce the issue for a
faster review.
7. If you want to add a new configuration, please read the guideline first
for naming configurations in
'core/src/main/scala/org/apache/spark/internal/config/ConfigEntry.scala'.
8. If you want to add or modify an error type or message, please read the
guideline first in
'common/utils/src/main/resources/error/README.md'.
-->
### What changes were proposed in this pull request?
<!--
Please clarify what changes you are proposing. The purpose of this section
is to outline the changes and how this PR fixes the issue.
If possible, please consider writing useful notes for better and faster
reviews in your PR. See the examples below.
1. If you refactor some codes with changing classes, showing the class
hierarchy will help reviewers.
2. If you fix some SQL features, you can provide some references of other
DBMSes.
3. If there is design documentation, please add the link.
4. If there is a discussion in the mailing list, please add the link.
-->
This PR fixes a whole-stage codegen (WSCG) correctness bug in ArrayJoin
(array_join) where the generated code computes the correct joined string but
discards it as NULL.
ArrayJoin.doGenCode initializes ev.isNull = true whenever the expression is
nullable (which is the case when the optional nullReplacement argument is a
nullable column). The actual join is then produced by
genCodeForArrayAndDelimiter, which has two branches:
When array or delimiter is nullable, the body is wrapped in nullSafeExec and
explicitly emits ev.isNull = false before building the result. When both array
and delimiter are non-nullable, the else branch builds the result but never
resets ev.isNull, leaving it at its initialized true.
A minimal reproduction:
SET spark.sql.codegen.wholeStage = true;
SET spark.sql.codegen.factoryMode = CODEGEN_ONLY;
-- Returns NULL for every row (buggy):
SELECT array_join(
array('a', 'b'),
',',
CASE WHEN id % 2 = 0 THEN 'NR' ELSE CAST(NULL AS STRING) END
) AS r
FROM range(4);
SET spark.sql.codegen.wholeStage = false;
SET spark.sql.codegen.factoryMode = NO_CODEGEN;
-- Returns ['a,NR,b', NULL, 'a,NR,b', NULL] (correct):
SELECT array_join(
array('a', 'b'),
',',
CASE WHEN id % 2 = 0 THEN 'NR' ELSE CAST(NULL AS STRING) END
) AS r
FROM range(4);
### Why are the changes needed?
<!--
Please clarify why the changes are needed. For instance,
1. If you propose a new API, clarify the use case for a new API.
2. If you fix a bug, you can clarify why it is a bug.
-->
This is a silent correctness bug: `array_join(arr, delimiter, repl)` returns
`NULL` for every row instead of the joined string, but only under a specific
(and realistic) combination:
- The third argument nullReplacement is a nullable, non-foldable column, so
`ArrayJoin.nullable` is true.
- An upstream `Filter` containing `IsNotNull(array)` (and/or
`IsNotNull(delimiter)`) tightens those children to non-nullable.
`FilterExec.output` marks `IsNotNull`-referenced attributes as non-nullable,
and `UpdateAttributeNullability` propagates this downstream, so
`genCodeForArrayAndDelimiter` takes the non-nullable else branch.
- The query stays in whole-stage codegen over a materialized source (e.g.
`FileScan parquet`, or an `InMemoryRelation` from `CACHE TABLE`). Inline
`VALUES / WITH` sources are folded by `ConvertToLocalRelation` to interpreted
`eval()` and therefore do not hit the bug.
Interpreted `eval()` returns the correct result, so the same query produces
different answers depending on whether codegen kicks in.
### Does this PR introduce _any_ user-facing change?
<!--
Note that it means *any* user-facing change including all aspects such as
new features, bug fixes, or other behavior changes. Documentation-only updates
are not considered user-facing changes.
If yes, please clarify the previous behavior and the change this PR proposes
- provide the console output, description and/or an example to show the
behavior difference if possible.
If possible, please also clarify if this is a user-facing change compared to
the released Spark versions or within the unreleased branches such as master.
If no, write 'No'.
-->
Yes. It fixes incorrect results. Previously, `array_join(arr, delimiter,
nullReplacement)` could return `NULL` for every row under whole-stage codegen
when nullReplacement was a nullable column and an upstream `IsNotNull` filter
made the array/delimiter non-nullable. After this change, such queries return
the correctly joined string, matching interpreted execution. Queries that were
already correct (2-arg form, literal non-null `nullReplacement`, no upstream
`IsNotNull` filter, or non-codegen execution) are unaffected.
### How was this patch tested?
<!--
If tests were added, say they were added here. Please make sure to add some
test cases that check the changes thoroughly including negative and positive
cases if possible.
If it was tested in a way different from regular unit tests, please clarify
how you tested step by step, ideally copy and paste-able, so that other
reviewers can test and check, and descendants can verify in the future.
If tests were not added, please describe why they were not added and/or why
it was difficult to add.
If benchmark tests were added, please run the benchmarks in GitHub Actions
for the consistent environment, and the instructions could accord to:
https://spark.apache.org/developer-tools.html#github-workflow-benchmarks.
-->
Unit testing in `CollectionExpressionsSuite` and `DataFrameFunctionsSuite`
### Was this patch authored or co-authored using generative AI tooling?
<!--
If generative AI tooling has been used in the process of authoring this
patch, please include the
phrase: 'Generated-by: ' followed by the name of the tool and its version.
If no, write 'No'.
Please refer to the [ASF Generative Tooling
Guidance](https://www.apache.org/legal/generative-tooling.html) for details.
-->
Generated-by: Claude Opus 4.8
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
