Gabriella Lotz has posted comments on this change. ( http://gerrit.cloudera.org:8080/24845 )
Change subject: [tserver] Validate deprecated range predicate column type. ...................................................................... Patch Set 5: (2 comments) http://gerrit.cloudera.org:8080/#/c/24845/5/src/kudu/tserver/tablet_server-test.cc File src/kudu/tserver/tablet_server-test.cc: http://gerrit.cloudera.org:8080/#/c/24845/5/src/kudu/tserver/tablet_server-test.cc@3384 PS5, Line 3384: for (int i = 0; i < 4; i++) { Does the loop still do what the comment says? I guess the rationale was that dereferencing a random pointer wouldn't hit an allocated address 4 times in a row, but with the type check the value bytes aren't read at all, so I believe all 4 iterations take the same rejection path. Could we drop the loop (and the random generator) and just send one request? http://gerrit.cloudera.org:8080/#/c/24845/5/src/kudu/tserver/tablet_service.cc File src/kudu/tserver/tablet_service.cc: http://gerrit.cloudera.org:8080/#/c/24845/5/src/kudu/tserver/tablet_service.cc@2791 PS5, Line 2791: if (PREDICT_FALSE(idx == Schema::kColumnNotFound)) { Does the new test cover both failure paths? I think it exercises the type mismatch, but the kColumnNotFound branch doesn't seem to have a case. Would it be worth adding one that sends a predicate for a column name that isn't in the tablet schema? -- To view, visit http://gerrit.cloudera.org:8080/24845 To unsubscribe, visit http://gerrit.cloudera.org:8080/settings Gerrit-Project: kudu Gerrit-Branch: master Gerrit-MessageType: comment Gerrit-Change-Id: I82ad53ad2263ca587e81fd37d64cc7982060b0f1 Gerrit-Change-Number: 24845 Gerrit-PatchSet: 5 Gerrit-Owner: Zoltan Martonka <[email protected]> Gerrit-Reviewer: Alexey Serbin <[email protected]> Gerrit-Reviewer: Gabriella Lotz <[email protected]> Gerrit-Reviewer: Kudu Jenkins (120) Gerrit-Reviewer: Zoltan Martonka <[email protected]> Gerrit-Comment-Date: Wed, 16 Sep 2026 13:50:02 +0000 Gerrit-HasComments: Yes
