hcrosse opened a new pull request, #1560:
URL: https://github.com/apache/arrow-go/pull/1560

   ### Rationale for this change
   
   When a `file` column reader cannot read a page, `ReadBatch` and `Skip` 
return no error, and the next call skips the failed page's values.
   
   ### What changes are included in this PR?
   
   `HasNext` stays false after a page error, `ReadBatch` and `Skip` return the 
error, and a successful `SeekToRow` clears it. This also covers `processPage` 
rejections such as decoder `SetData` errors.
   
   This changes the `ReadBatch` contract. It returned a nil error for a page it 
could not open, with the reason available only through `Err()`. It now returns 
the error with any values read before that page. 
`TestALPOnIntegerColumnIsRejected` now expects the returned error.
   
   ### Are these changes tested?
   
   `TestReadBatchStopsAtCorruptPage`, `TestSkipStopsAtCorruptPage` and 
`TestSeekToRowAfterCorruptPage` cover partial results with a persistent error, 
`Skip`, and reading after a seek. All three pass with this change, and the 
`ReadBatch` and `Skip` tests fail on main.
   
   With Go 1.26.10, vet and race tests for `parquet/file` and 
`parquet/pqarrow`, full `go test -race ./parquet/...` with pinned test data, 
`golangci-lint` 2.13.2 and the Release Audit Tool passed.
   
   ### Are there any user-facing changes?
   
   Yes. `ReadBatch` and `Skip` return page errors instead of moving past the 
failed page. There is no API change.
   
   Closes https://github.com/apache/arrow-go/issues/1559.
   


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to