Pass columns instead of Series to cudf.DataFrame in split-combine workflow
#1429
Conversation
This file contains 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
jameslamb
added
improvement
|
Is this fix needed in 24.08? |
harrism
approved these changes
Aug 5, 2024
I wouldn't necessarily call this a show-stopping bug, so IMO it's OK to target for 24.10 (especially since the Python unit tests didn't break due to this bug). But I'm happy to backport it if others think it's necessary |
|
Based on @harrism 's 👍🏻 on #1429 (comment) and our offline conversation about this, seems to me this fix belongs in 24.10. I'm going to merge this into 24.10, to help with #1424. I also am available to help if we do decide later to backport it to 24.08. |
|
/merge |
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Outdated suggestions cannot be applied.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.
Description
closes #1426
It appears there was a change in 24.08 that broke a notebook demonstrating a
mergeon two geometry columns. It seems like the merge result tries to reconstruct aGeoDataFramefrom adict[Any, GeoSeries | Series]but theSeries.indexalignment requires the types to be recognized cudf types (not"geometry")I don't think this alignment is entirely necessary though since it goes through the
_split_out_geometry_columns/_recombine_columnsmethods which appears to be used on operations that maintain row ordering so index alignment isn't required.This PR instead passes a
dict[Any, GeoColumn | Column]tocudf.DataFrame._from_datagiven that this row ordering is preserved.(This PR also includes the fix for #1427)
Checklist