[SPARK-56045][SQL] Add flag for ignoring Parquet UNKNOWN type annotation and revert to old behavior#54870
Closed
ZiyaZa wants to merge 2 commits intoapache:masterfrom
Closed
[SPARK-56045][SQL] Add flag for ignoring Parquet UNKNOWN type annotation and revert to old behavior#54870ZiyaZa wants to merge 2 commits intoapache:masterfrom
ZiyaZa wants to merge 2 commits intoapache:masterfrom
Conversation
Contributor
|
LGTM if CI is green, please create a new JIRA ticket as the original commit is already released. |
ZiyaZa
changed the title
Contributor
Author
CI is green, linked the new ticket in the title. |
cloud-fan
pushed a commit
that referenced
this pull request
Mar 18, 2026
…ion and revert to old behavior ### What changes were proposed in this pull request? This PR introduces a new flag `spark.sql.parquet.reader.respectUnknownTypeAnnotation.enabled` for Parquet reader to control the behavior when it reads an external file with `UNKNOWN` logical type annotation: - (Default) When false, we infer the Spark type based on the physical type used in the Parquet file, as we did before Spark 4.1. - When true, we use NullType as the Spark type. ### Why are the changes needed? To fix the regression introduced by #52922, as we have been reading files differently since then. ### Does this PR introduce _any_ user-facing change? Yes. With default flag value, when we read a Parquet file written by an external engine: - Before, we inferred NullType - Now, we'll infer a type based on the physical type (e.g. IntegerType) ### How was this patch tested? Added tests. ### Was this patch authored or co-authored using generative AI tooling? No. Closes #54870 from ZiyaZa/unknown-type-flag. Authored-by: Ziya Mukhtarov <ziya5muxtarov@gmail.com> Signed-off-by: Wenchen Fan <wenchen@databricks.com> (cherry picked from commit 50514c5) Signed-off-by: Wenchen Fan <wenchen@databricks.com>
Contributor
|
thanks, merging to master/4.1! |
| "inference and infers NullType. When disabled, ignores the UNKNOWN annotation " + | ||
| "and uses the physical type instead.") | ||
| .version("4.1.2") | ||
| .withBindingPolicy(ConfigBindingPolicy.SESSION) |
Member
There was a problem hiding this comment.
Hi, @ZiyaZa and @cloud-fan .
This broken branch-4.1. Let me revert this from branch-4.1 only for now.
[error] /home/runner/work/spark/spark/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala:1627:8: value withBindingPolicy is not a member of org.apache.spark.internal.config.ConfigBuilder
[error] possible cause: maybe a semicolon is missing before `value withBindingPolicy`?
[error] .withBindingPolicy(ConfigBindingPolicy.SESSION)
[error] ^
[error] /home/runner/work/spark/spark/sql/catalyst/src/main/scala/org/apache/spark/sql/internal/SQLConf.scala:1627:26: not found: value ConfigBindingPolicy
[error] .withBindingPolicy(ConfigBindingPolicy.SESSION)
[error] ^
[error] two errors found
Contributor
Author
There was a problem hiding this comment.
Hi, it seems withBindingPolicy doesn't exist in 4.1. So deleting that line should solve it. I can create a PR deleting that line, or re-apply this PR as a whole after your revert. Either way works.
Member
|
Yes, it's already reverted. Please make a new backporting PR to branch-4.1 now to make it sure that CI passes, @ZiyaZa . |
Member
|
BTW, thank you for the fix, @ZiyaZa . |
Contributor
Author
|
Created a new PR here: #54885 Thanks for letting me know. |
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
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 introduces a new flag
spark.sql.parquet.reader.respectUnknownTypeAnnotation.enabledfor Parquet reader to control the behavior when it reads an external file withUNKNOWNlogical type annotation:Why are the changes needed?
To fix the regression introduced by #52922, as we have been reading files differently since then.
Does this PR introduce any user-facing change?
Yes. With default flag value, when we read a Parquet file written by an external engine:
How was this patch tested?
Added tests.
Was this patch authored or co-authored using generative AI tooling?
No.