jorisvandenbossche commented on pull request #11351: URL: https://github.com/apache/arrow/pull/11351#issuecomment-939047021
Tested this branch, and it seems to work nicely. I was just looking into the topic of "selecting nested columns" for ORC (https://github.com/apache/arrow/pull/11372), and selecting lists elements actually uses different "dotted paths" in ORC compared to Parquet. For example, your example in the top post of `file.read(["root.list.item.addr.that"])` needs to be written as `file.read(["root.addr.that"])` for an ORCFile. Not for this PR, but that's maybe something to consider for Parquet as well? (I personally like this simplified selecting; the "list.item" part in the path for lists is basically superfluous, since lists always have a single child) -- 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]
