drin commented on PR #13487: URL: https://github.com/apache/arrow/pull/13487#issuecomment-1362341154
hmm, so I have implemented more test coverage now, but when testing `ListArray` I realized that my naive approach for nested types is too naive. I have figured out what I think is correct, but making the updates is a bit slow (easy to confuse myself). The gist of the change is that for list types I need to flatten the structure to a binary-like layout that has offsets and variable length data. In simple cases, (list and map), this mostly means flattening offsets and recursing into child arrays as before. No change for struct types. The fixed size list type should be relatively similar to other list types. Union types seem like they will require a different approach; though, they may only require something like a flattened sparse union representation. -- 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
