westonpace commented on pull request #11106:
URL: https://github.com/apache/arrow/pull/11106#issuecomment-924286237
Ok. Given the comments I am going to close this. I believe the correct fix
would be to fix things on the pandas end so that a pandas UUID type could be
used (and this could convert to Arrow automatically).
> Although we could add extension types for python builtin types which would
properly roundtrip as well.
This is not a bad idea though I think it would only be needed in the
`from_pydict` case. When creating arrays you can always specify the type and
pandas has its own mechanisms. Maybe we could get away with a type mapper?
I'll make a JIRA for this idea.
> Maybe it's time to actually add a UUID C++ extension type .. (Parquet also
supports it, some SQL engines as well)
That might be nice. It would be good to have some example C++ extension
types. I don't see myself working on it anytime soon though so I'll probably
not create a JIRA.
--
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]
