Uh oh!
There was an error while loading. Please reload this page.
fix: TestDeltaByteArray implementation and fix - #369
Merged
Conversation
This exact test seems to be passing on v17 version. More investigation is needed
zeroshade
commented
May 7, 2025
Member
Can you provide the csv file that you're testing with so that I can try to reproduce? |
| } | ||
| require.DirExists(t, dir) | ||
| expected, err := os.ReadFile(path.Join(dir, "delta_byte_array_expect.csv")) |
| records = records[1:] // skip header | ||
| props := parquet.NewReaderProperties(memory.DefaultAllocator) | ||
| fileReader, err := file.OpenParquetFile(path.Join(dir, "delta_byte_array.parquet"), |
ContributorAuthor
There was a problem hiding this comment.
here is the Parquet
MetalBlueberry
commented
May 8, 2025
ContributorAuthor
I got the files from the parquet-testing repository. You can find the info here https://github.com/apache/parquet-testing/blob/master/data/delta_byte_array.md |
I'm unsure about what the difference is, but by looking at version v17 code, I noticed a difference between Int32 and Int64 decoder. Int32 used totalValues and Int64 used nvals. Looks like when merging implementations, the Int64 approach was used. Which seems to not be working properly for Int32. This change at least seems to be able to read the test file, which makes me think it is the right approach.
MetalBlueberry
commented
May 9, 2025
ContributorAuthor
@zeroshade see last commit for the fix. I'm just pattern matching, I do not really know what those values mean 😅 . |
MetalBlueberry
marked this pull request as ready for review
May 9, 2025 08:55
zeroshade
commented
May 9, 2025
Member
I figured out what the underlying issue was and updated this to be a better fix that won't cause the other tests to fail 😄 |
zeroshade
approved these changes
May 10, 2025
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 freeto join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
This exact test seems to be passing on v17 version. More investigation is needed
Rationale for this change
After updating from v17 to v18 I noticed a test case failing that was passing before.
What changes are included in this PR?
This PR contains a test case to reproduce the issue. A reference PR for version v17 can be found here
Are these changes tested?
The PR itself just contains the test
Are there any user-facing changes?