houqp commented on pull request #1905: URL: https://github.com/apache/arrow-datafusion/pull/1905#issuecomment-1059724553
I also think we should try to avoid forcing mutex at the api level to avoid its overheads. > In order to change ChunkReader to &mut self to cut of Mutex, it seems I need to change the full code path along parquet reading. Since they all share an immutability interface. What if we only use Arc and call `get_mut`? This is similar to what's happening with FileSource in the parquet crate. In the long run though, I feel like it might be worth looking into getting rid of the `ChunkReader` abstraction in the Parquet crate entirely and replace with Read + Seek. -- 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: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
