szehon-ho commented on code in PR #9251:
URL: https://github.com/apache/iceberg/pull/9251#discussion_r1424802087
##########
core/src/main/java/org/apache/iceberg/DeleteFileIndex.java:
##########
@@ -474,68 +421,52 @@ private Collection<DeleteFile> loadDeleteFiles() {
DeleteFileIndex build() {
Iterable<DeleteFile> files = deleteFiles != null ? filterDeleteFiles() :
loadDeleteFiles();
- boolean useColumnStatsFiltering = false;
+ EqualityDeletes globalDeletes = new EqualityDeletes();
+ PartitionMap<EqualityDeletes> eqDeletesByPartition =
PartitionMap.create(specsById);
+ PartitionMap<PositionDeletes> posDeletesByPartition =
PartitionMap.create(specsById);
+ CharSequenceMap<PositionDeletes> posDeletesByPath =
CharSequenceMap.create();
Review Comment:
It is possible we'll see a memory growth with using path map, example if we
have a million data files, can be ~100 mbs, if i calculate correctly?. (We use
DeleteFileIndex centrally if i understand correctly). I wonder if we
considered doing some common-prefix encoding, or other encoding, of paths
anywhere. Not sure if its worth it , given the typical memory size of planner
--
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: issues-unsubscr...@iceberg.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@iceberg.apache.org
For additional commands, e-mail: issues-h...@iceberg.apache.org
