Fix an issue with multiple short list rowgroups using the Parquet chunked reader. #15342
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
First, row groups of lists being loaded together were not getting their end row counts computed correctly on a per-rowgroup basis. The main consequence there was that we could potentially have generated splits that were larger than we would have liked. But downstream from this was a second bug where we were generating incorrect page indices to be decoded, causing corruption in the decode kernels.
nvdbaranec
added
bug
|
Leaving as a draft until running it through all the Spark tests. |
ttnghia
approved these changes
Mar 20, 2024
vuule
reviewed
Mar 20, 2024
|
Adding do-not-merge until pending changes run through the spark integration tests. |
vuule
reviewed
Mar 20, 2024
…oid unnecessarily looping in some cases.
vuule
approved these changes
Mar 20, 2024
|
/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Outdated suggestions cannot be applied.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.
Fixes #15306
The core issue here was that under certain conditions, the chunked reader could generate invalid page indices for list columns when using the chunked reader. This led to corruption in the decode kernels. The fix is fairly simple, but there's a decent amount of delta in this PR that is just name changes for clarity and some more comments/docs.
This affected the number of chunks generated in some of the very (unrealistically) constrained tests.
Checklist