englefly commented on PR #68314:
URL: https://github.com/apache/doris/pull/68314#issuecomment-5772521965

   Pushed `63572a9fa58` with the review follow-ups:
   
   1. **BE production fix** (`column_reader.cpp`): 
`FileColumnIterator::read_by_rowids()` in `NULL_MAP_ONLY` mode filled `count` 
placeholders into the nested column before the rowid loop and appended 
`total_read_count` more at the end, leaving the nested column longer than the 
null map. Only the post-loop insertion is kept now (matches master). This is 
what the new sparse-rowid test was catching.
   2. **`segment_iterator_lazy_pruned_test.cpp`**: the fixture initializes 
`_schema_block_id_map` the way `_vec_init_lazy_materialization()` does instead 
of indexing an empty vector.
   3. **`column_reader_test.cpp`**: the skipped-read test starts from a truly 
empty destination and now asserts the append semantics with a second read 
(`MapReadByRowidsSkipReadingAppendsPlaceholders`).
   4. **FE `PruneNestedColumnTest`**: kept the verified `[s, f]` DATA 
expectation (see the thread reply — the collected set for that query is exactly 
that path, and the FIELD is nullable, so the suggested `[s, f, NULL]` META 
expectation fails). Added `testNotNullStructIsNullOnOuterJoinNullableSide`, 
which is the shape that actually reaches the `hasPhysicalNullMap` guard and 
asserts no root `[s, NULL]` META path is emitted, so the suppression this PR 
adds is really covered.
   
   Verification on the pushed commit:
   * `./run-fe-ut.sh --run 
org.apache.doris.nereids.rules.rewrite.PruneNestedColumnTest` → 64 tests, 0 
failures; `mvn checkstyle:check -pl fe-core` → 0 violations.
   * `BUILD_TYPE_UT=Release ./run-be-ut.sh --run 
--filter='ColumnReaderTest.NullMapOnlyReadBySparseRowidsAcrossPages:ColumnReaderTest.MapReadByRowidsSkipReadingAppendsPlaceholders:SegmentIteratorLazyPrunedTest.*'`
 → 4 tests, 0 failures.


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to