XiaoHongbo-Hope opened a new pull request, #11266:
URL: https://github.com/apache/paimon/pull/11266

   ### Purpose
   
   When BLOB files in a logical read range have disjoint row IDs, different 
sequence numbers do not represent competing versions. The Python fallback 
reader nevertheless checks every file state on every batch and searches 
sequence dictionaries for each output row.
   
   Detect disjoint file ranges, visit them in row order with a forward-only 
active cursor, and resolve their values through one group. Keep the existing 
fallback behavior for overlapping files and preserve selected ranges, deletion 
vectors, gaps, NULLs, and placeholders. This addresses the Python analogue of 
#11225 without changing the file format or Java reader.
   
   ### Tests
   
   - 115 tests passed, 1 skipped, 7 large-data tests deselected across the BLOB 
table, deletion-vector, and batch-reader suites.
   - The new 128-file regression fails on the original reader with 16,384 state 
checks; the patched reader stays within 256 checks and returns identical rows.
   - Coverage includes selected gaps, placeholders, explicit NULLs, deletion 
vectors, and existing descriptor/update/rolling-file reads.
   - Flake8 with the repository configuration and `git diff --check` 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]

Reply via email to