JingsongLi commented on code in PR #8380:
URL: https://github.com/apache/paimon/pull/8380#discussion_r3490548952
##########
paimon-core/src/main/java/org/apache/paimon/schema/SchemaValidation.java:
##########
@@ -894,9 +894,6 @@ private static void validateRowTracking(TableSchema schema,
CoreOptions options)
checkArgument(
rowTrackingEnabled,
"Data evolution config must enabled with
row-tracking.enabled");
- checkArgument(
- !options.deletionVectorsEnabled(),
- "Data evolution config must disabled with
deletion-vectors.enabled");
checkArgument(
Review Comment:
Before removing this guard, can we also make data-evolution compaction carry
or apply the deletion vectors? Snapshot reads now attach range-based DVs to
`DataSplit`, but `DataEvolutionCompactTask` still builds a `DataSplit` from
`compactBefore` without `dataEvolutionDeletionFiles` and writes compacted files
without replacement DV metadata. After compacting the normal anchor file, the
existing DV is still keyed by the removed file name, so later snapshot reads of
the new compacted file will not find it and previously deleted rows can
reappear. Please either keep rejecting this option combination until compaction
is updated, or update compaction to read existing DVs and preserve/filter them.
--
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]
