JingsongLi opened a new pull request, #890: URL: https://github.com/apache/paimon-rust/pull/890
## What changed - add deterministic fixed-live-row chunk planning to the Rust table scan - keep append chunk boundaries as native file-local ranges and data-evolution boundaries as global row-id ranges - account for deletion vectors while cutting chunks and preserve exact visible row counts - match CPython integer-seed shuffle semantics, including negative and arbitrarily large seeds - shard the shuffled chunk list directly for parallel workers - let the data-evolution reader process multiple disjoint row-id groups carried by one native chunk - expose chunk planning and native-only split metadata through the Python binding ## Compatibility Java SplitSerializer v1 cannot encode file-local ranges. Regular serialization therefore fails loudly for those native splits, while the Python metadata view intentionally omits the native-only fields and retains the opaque Rust split for reading. ## Verification - cargo test -p paimon --lib - cargo test -p pypaimon_rust --lib - cargo clippy -p paimon -p pypaimon_rust --all-targets -- -D warnings - cargo fmt --all -- --check - companion PyPaimon native-plan/native-read matrix: 20 passed -- 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]
