fix: support nullable columns in pre-sorted data sources #16783
Merged
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
github-actions
bot
added
sqllogictest
crepererum
force-pushed
the
crepererum/handle-null-cols-in-files
branch
from
b5fd354 to
e5c1777
Compare
crepererum
commented
Jul 15, 2025
| @@ -130,8 +130,7 @@ STORED AS PARQUET; | |||
| ---- | |||
| 3 | |||
|
|
|||
| # Check output plan again, expect no "output_ordering" clause in the physical_plan -> ParquetExec, | |||
| # due to there being more files than partitions: | |||
| # Check output plan again | |||
There was a problem hiding this comment.
I think the actual reason why the output_ordering was missing here wasn't the number of files, but because DF had trouble with a column that wasn't marked as NOT NULL (i.e. is nullable).
There was a problem hiding this comment.
I agree -- I reviewed the test and the table definition explicitly says
WITH ORDER (string_col ASC NULLS LAST, int_col ASC NULLS LAST)So I would expect this plan not to have additional sorting
alamb
approved these changes
Jul 15, 2025
There was a problem hiding this comment.
Thank you @crepererum -- I reviewed the code and test changes carefully and I think they all make sense
| @@ -130,8 +130,7 @@ STORED AS PARQUET; | |||
| ---- | |||
| 3 | |||
|
|
|||
| # Check output plan again, expect no "output_ordering" clause in the physical_plan -> ParquetExec, | |||
| # due to there being more files than partitions: | |||
| # Check output plan again | |||
There was a problem hiding this comment.
I agree -- I reviewed the test and the table definition explicitly says
WITH ORDER (string_col ASC NULLS LAST, int_col ASC NULLS LAST)So I would expect this plan not to have additional sorting
| @@ -230,14 +230,7 @@ impl MinMaxStatistics { | |||
| .zip(sort_columns.iter().copied()) | |||
| .map(|(sort_expr, column)| { | |||
| let schema = values.schema(); | |||
|
|
|||
crepererum
added a commit
to influxdata/arrow-datafusion
that referenced
this pull request
Jul 17, 2025
crepererum
added a commit
to influxdata/arrow-datafusion
that referenced
this pull request
Jul 17, 2025
crepererum
added a commit
to influxdata/arrow-datafusion
that referenced
this pull request
Jul 17, 2025
|
Thanks @crepererum |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Which issue does this PR close?
-
Rationale for this change
There are valid cases where files are pre-sorted on nullable columns. Arrow and DataFusion have proper semantic for that.
What changes are included in this PR?
Removes an exception / special-handling for nullable columns. The Arrow row encoder that is used in the respective code fully supports nullable columns (or at least it will cleanly fail if it doesn't), INCLUDING nulls-first&last.
Are these changes tested?
split_groups_by_statisticsAre there any user-facing changes?
Some queries are now faster.