wjones127 commented on pull request #12216:
URL: https://github.com/apache/arrow/pull/12216#issuecomment-1029112711


   > In the PR description you say that validating the table catches the issue. 
Is it sufficient or is the custom MemoryPool actually needed?
   
   @pitrou I felt that was a very indirect way of testing the issue. It's only 
one possible consequence of the buffer overflow; most of the time another 
validity bitmap doesn't seem to be immediately after. I tried for several days 
to come up with another (shorter) record batch that produces this issue, but 
all I could find is the one in that parquet file and that one `.Slice(159, 
160)`. This is a lot of code, but I seemed like the most direct way to test 
this.
   
   > I'm working on https://issues.apache.org/jira/browse/ARROW-15550 which 
will allow exercising this bug without additional test.
   
   Yeah I think if we implemented that then this test would be obsolete.


-- 
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