mengna-lin commented on PR #16094:
URL: https://github.com/apache/iceberg/pull/16094#issuecomment-5722727907
Thanks @xndai — agreed that field ids are the stable identity and a
name-keyed hint can bind to the wrong column after a drop + re-add.
That said, every existing per-column property is name-keyed and shares this
exact behavior: bloom-filter-{enabled,fpp,ndv}.column, stats-enabled.column,
dict-encoding-enabled.column, and write.metadata.metrics.column. Making
compression alone field-id-keyed wouldn't fix the class of problem — it'd just
make one property inconsistent.
I'd suggest keeping this convention for this feature and raising field-id
keying as a separate proposal for the whole *.column-name family.
--
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]
