Gabriel39 opened a new pull request, #65548:
URL: https://github.com/apache/doris/pull/65548
### What problem does this PR solve?
Issue Number: None
Related PR: None
Problem Summary: FileScannerV2 treated valid empty splits as scan failures,
localized promoted nested struct predicates with mismatched operand types,
retained a process-global Parquet page-cache range index, and rejected whole
Parquet schemas when unprojected leaves used unsupported logical types. This
change skips and counts empty splits, casts localized nested leaves back to
table types, scopes range metadata to each file reader, and defers
unsupported-type errors until projection. Regression fixtures and commented
examples cover empty CSV input, mixed Parquet struct leaf types, repeated
multi-file scans, and an unprojected TIME_MILLIS column.
### Release note
Fix FileScannerV2 compatibility for empty files, evolved nested predicates,
and Parquet files containing unprojected unsupported columns; reduce Parquet
page-cache range-index contention and lifetime.
### Check List (For Author)
- Test: Regression test / Unit Test
- 55 targeted BE ASAN unit tests
- test_file_scanner_v2_review_fixes regression suite
- Behavior changed: Yes (valid empty splits are skipped; unsupported Parquet
logical leaves fail only when projected)
- Does this need documentation: No
--
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]
