[SPARK-49110][SQL] Fix reading metadata columns for tables with CHAR columns #53846
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
JIRA Issue Information=== Improvement SPARK-49110 === This comment was automatically generated by GitHub Actions |
tomvanbussel
force-pushed
the
SPARK-49110-fix
branch
from
502a294 to
61e96a8
Compare
cloud-fan
reviewed
Jan 19, 2026
|
|
||
| def readSideCharPadding: Boolean = getConf(SQLConf.READ_SIDE_CHAR_PADDING) | ||
|
|
||
| def readSideCharPaddingAfterAlias: Boolean = |
Contributor
There was a problem hiding this comment.
we usually don't add a method if it's only called once.
cloud-fan
approved these changes
Jan 19, 2026
vladimirg-db
approved these changes
Jan 19, 2026
cloud-fan
approved these changes
Jan 19, 2026
Instead of adding char padding Project between SubqueryAlias and the data source scan then moving it afterwards, use a two-pass approach: - First pass: match SubqueryAlias with data source scan, add padding Project AFTER SubqueryAlias to preserve metadata column access - Second pass: match data source scan alone (for cases like spark.read.format(...).load()) For idempotence, when adding char padding Project, the rule also clears out char type in the output attributes, so the second pass does nothing if the first pass matches. This removes the need for the READ_SIDE_PADDING_PROJECT_TAG.
Use two-pass approach for char padding with SubqueryAlias
cloud-fan
reviewed
Jan 19, 2026
| sql(s"CREATE TABLE $tbl (id bigint, data char(1)) PARTITIONED BY (bucket(4, id), id)") | ||
| sql(s"INSERT INTO $tbl VALUES (1, 'a'), (2, 'b'), (3, 'c')") | ||
| val sqlQuery = sql(s"SELECT id, data, index, _partition FROM $tbl") | ||
| val sqlQueryWithAlias = sql(s"SELECT t.id, t.data, t.index, t._partition FROM $tbl t") |
Contributor
There was a problem hiding this comment.
user specified subquery alias always work, can we test SELECT $tbl.id, ... FROM $tbl?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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?
This PR modifies
SubqueryAliasto propagate the metadata output of its child if the is aProjectinjected byApplyCharTypePadding.Why are the changes needed?
This is needed because with these changes it is not possible to read metadata columns when reading a table with a CHAR column when read-side padding is enabled. In the case the plan is
SubqueryAlias(Project(LeafNode))), so without this patch the metadata output is not propagated.Does this PR introduce any user-facing change?
Yes, there will be more cases in which the metadata columns can be accessed, and users will no longer get an exception in these cases.
How was this patch tested?
Added a case to
MetadataColumnSuite.Was this patch authored or co-authored using generative AI tooling?
No