hcrosse opened a new pull request, #1613:
URL: https://github.com/apache/arrow-go/pull/1613

   ### Rationale for this change
   
   `SeekToRow` can find no data page for the row, for example in a damaged 
file. It then returns `nil`, so the seek looks successful and reads return 
nothing. If the page reader ends right after a dictionary page, `SeekToRow` 
instead panics because it uses the missing page as a `DataPage`.
   
   ### What changes are included in this PR?
   
   When the seek finds no data page, `SeekToRow` returns the page reader's 
error if there is one, and otherwise an error saying it found no data page for 
the row. Either error stops the reader until a seek succeeds.
   
   ### Are these changes tested?
   
   Yes. I added tests for chunks with no data page, through both the column 
reader and the record reader and for an encrypted chunk, and for a page reader 
that ends or fails after a dictionary page.
   
   ### Are there any user-facing changes?
   
   Yes. `SeekToRow` returns an error when it finds no data page for the row, 
instead of succeeding or panicking. There is no API change.
   


-- 
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