Re: [I] Partition specific pointed lookup/reading strategy for metadata table [hudi]

["nsivabalan (via GitHub)" <gi...@apache.org>]

nsivabalan commented on issue #15007:
URL: https://github.com/apache/hudi/issues/15007#issuecomment-4795299798

   This appears to already be implemented on master. `HoodieBackedTableMetadata.isFullScanAllowedForPartition()` ([HoodieBackedTableMetadata.java#L733-L746](https://github.com/apache/hudi/blob/master/hudi-common/src/main/java/org/apache/hudi/metadata/HoodieBackedTableMetadata.java#L733-L746)) selects per-partition strategy: `files` uses eager full-scan (so the external spillable map caches records across lookups), while `column_stats`, `bloom_filters`, and any other partition fall back to point-lookups via the log-record reader — matching the request here. The related fix landed under HUDI-3301 (commit f8028a4 / PR #6782, `Fixing HoodieMetadataLogRecordReader to avoid flushing cache for every lookup`). If there's a remaining sub-case (e.g. additional partitions like `record_index`, `secondary_index`, `partition_stats`) where the current default isn't right, please reopen with that specifics; otherwise this can likely be closed.


-- 
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: commits-unsubscribe@hudi.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org