lidavidm commented on PR #34079: URL: https://github.com/apache/arrow/pull/34079#issuecomment-1423360477
Then it seems weird to get a scalar by logical index as we do here, but changing the meaning of getting a scalar would also be inconsistent. On Wed, Feb 8, 2023, at 18:06, Felipe Oliveira Carvalho wrote: > > > ***@***.**** commented on this pull request. > > > In go/arrow/scalar/scalar.go <https://github.com/apache/arrow/pull/34079#discussion_r1100774478>: > > > + runEnds := arr.RunEndsArr() > + switch ends := runEnds.(type) { > + case *array.Int16: > + runLength = int64(ends.Value(physicalIndex)) > + case *array.Int32: > + runLength = int64(ends.Value(physicalIndex)) > + case *array.Int64: > + runLength = int64(ends.Value(physicalIndex)) > + } > > @zeroshade <https://github.com/zeroshade> Antoine has suggested that I add a RunEndEncoded scalar type instead of adding hacks to make things work without it. > >> To me, this scalar makes sense as representing the physical value for a physical index, but not the logical value for a logical index (if that makes...logical sense?) >> > If you look at the REE array as an array of runs, the scalar that is typed by the REE type is a single run. > > > — > Reply to this email directly, view it on GitHub <https://github.com/apache/arrow/pull/34079#discussion_r1100774478>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AACQB35OEZYPF2SRHZUF65TWWQRIHANCNFSM6AAAAAAUVUO7GM>. > You are receiving this because you were mentioned.Message ID: ***@***.***> > -- 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]
