XiaoHongbo-Hope opened a new pull request, #1460:
URL: https://github.com/apache/paimon-rust/pull/1460

   ### Purpose
   
   BLOB files with different sequence numbers currently enter separate fallback 
groups even when their row-ID ranges are disjoint. Each output batch then 
traverses those unrelated groups, including groups whose files have already 
been exhausted.
   
   Coalesce disjoint file ranges into one row-ordered group after 
selected-range pruning. Keep the existing fallback reader to preserve gap and 
placeholder handling; any overlapping ranges retain the original newest-first 
groups. This is the Rust counterpart of apache/paimon#11225 and does not change 
the file format or sequential-reader eligibility.
   
   ### Validation
   
   - Regression coverage coalesces 128 disjoint groups and checks ordered 
output across batches.
   - Tests preserve selected gaps, explicit NULLs, placeholders, and 
overlapping-version fallback without reading overridden payloads.
   - The table-reader integration test now exercises both same-sequence and 
different-sequence rolling BLOB files, reading actual Parquet/BLOB fixtures as 
bytes and descriptors.
   - 93 tests passed: `cargo test -p paimon --lib table::data_evolution_reader 
--no-default-features --features storage-memory,storage-fs`
   - `cargo fmt --all -- --check` and `git diff --check`.
   


-- 
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]

Reply via email to