JingsongLi opened a new pull request, #10408: URL: https://github.com/apache/paimon/pull/10408
### Purpose Enable Native `_SEQUENCE_NUMBER` reads and predicates over partial Data Evolution files, and align the Python reader's tracking versions with Java. Depends on https://github.com/apache/paimon-rust/pull/1051. **Merge Rust first**: CI continues to install `paimon-rust` from `main`. - Remove the obsolete Native partial-file sequence capability fallback. - Evaluate row-tracking metadata predicates after metadata assignment instead of passing missing sequence columns to Parquet decoding. Cover mixed user/metadata predicates and limits. - Keep non-null physical sequence versions and fill null entries from manifest metadata, matching Java `ColumnarRowIterator`. - Include sequence projections in existing Native Data Evolution REST commit tests. - Remove obsolete lost-file recovery configuration cases from ROW-sidecar tests and retain strict missing/corrupt sidecar assertions. - Add real pure-Python/Native cross-read regressions with Python/Native writers, deletion vectors, repeated partial updates, physical versions and case-sensitive metadata names. ### Tests - Local Native regression: **1,413 passed, 31 skipped**, covering planning, reading, writing, updating and committing with the paired Rust binding. - Final focused Native system-column, ROW-sidecar, merge, sequence and read tests: **271 passed**; 365 Native plans, 384 reads and 265 writes exercised. - `pypaimon/tests/py36`: **79 passed** in normal Python mode on the local Python 3.13 runtime. - Flake8 with `paimon-python/dev/cfg.ini` for all changed Python files. No CI source pin, storage-format change or new dependency. -- 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]
