Joe McDonnell has posted comments on this change. ( http://gerrit.cloudera.org:8080/22666 )
Change subject: IMPALA-13882: Fix Iceberg v2 deletes with tuple caching ...................................................................... Patch Set 2: (2 comments) http://gerrit.cloudera.org:8080/#/c/22666/2/be/src/exec/iceberg-delete-builder.cc File be/src/exec/iceberg-delete-builder.cc: http://gerrit.cloudera.org:8080/#/c/22666/2/be/src/exec/iceberg-delete-builder.cc@131 PS2, Line 131: // Tuple caching can place a TupleCacheNode above the scan node. Look past > I think this would make sense as a helper that could also be used in fragme I turned this into a static helper function on PlanNode to look past a single tuple cache node. Something that looks up a child with an option to bypass a tuple cache node would work here, but fragment-state.cc isn't looking for a child. http://gerrit.cloudera.org:8080/#/c/22666/2/be/src/exec/iceberg-delete-builder.cc@138 PS2, Line 138: if (delete_scan_node->tnode_->node_type != TPlanNodeType::HDFS_SCAN_NODE) { > Should this be a DCHECK? I think it would only happen if we've done somethi Changed this to a DCHECK. I think the code in fragment-state.cc would fail at runtime before we would get to this DCHECK. -- To view, visit http://gerrit.cloudera.org:8080/22666 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: Impala-ASF Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I162e738c4e4449a536701a740272aaac56ce8fd8 Gerrit-Change-Number: 22666 Gerrit-PatchSet: 2 Gerrit-Owner: Joe McDonnell <joemcdonn...@cloudera.com> Gerrit-Reviewer: Impala Public Jenkins <impala-public-jenk...@cloudera.com> Gerrit-Reviewer: Joe McDonnell <joemcdonn...@cloudera.com> Gerrit-Reviewer: Kurt Deschler <kdesc...@cloudera.com> Gerrit-Reviewer: Michael Smith <michael.sm...@cloudera.com> Gerrit-Reviewer: Yida Wu <wydbaggio...@gmail.com> Gerrit-Comment-Date: Mon, 07 Apr 2025 22:04:09 +0000 Gerrit-HasComments: Yes
