tanmayrauth commented on PR #2444:
URL: https://github.com/apache/iceberg-go/pull/2444#issuecomment-6988600605
Thanks @laskoviymishka, all four are in the latest push.
Unknown-typed columns are now skipped by default like geo, and naming one
returns ErrInvalidArgument; TestComputeNDVsSkipsUnknownColumns pins the V3 {id
int, v unknown} case, which failed with unsupported value type: <nil> before
the fix. Once the snapshot is resolved, the projection and the scan both use
WithSnapshotID(snap.SnapshotID), so columns come from the snapshot's schema, as
in Java's table.schemas().get(snapshot.schemaId());
TestComputeNDVsUsesSnapshotSchema covers an AddColumn with no new snapshot,
where extra is no longer reported and naming it is an error. The type matrix
now has timestamp_ns and timestamptz_ns rows on V3 tables, with values that
differ only below the microsecond so any truncation would show up, and I
applied your doc suggestion on empty values.
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]