zhztheplayer commented on PR #13248: URL: https://github.com/apache/arrow/pull/13248#issuecomment-1216352589
> @zhztheplayer correct me if I'm wrong, but what I see as the issue is this: data comes in via the C Data Interface. You then want to unload the data into an ArrowRecordBatch and load it into another root (that happens to have its own allocator) to perform some computation on it. But since transfer is not implemented, that currently fails. Yes this could be the original issue. If we all agree on associating an allocator with the imported buffer I can start to try making a implementation then. Actually I was trying to get the actual meaning of buffer transferring for some time. In c++ we don't have this design but things seem to go on well. Also it seems in netty or java nio they don't design transfer semantic for buffer too. -- 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]
