JingsongLi opened a new pull request, #11318:
URL: https://github.com/apache/paimon/pull/11318

   ### Purpose
   
   An eager Data Evolution scan could evaluate the global index against one 
snapshot and then plan data splits against a newer snapshot. An append between 
those steps could make both `full` and `detail` omit a matching row.
   
   Prepare the batch reader once. Use its selected snapshot for scalar index 
and BTree Top-N evaluation, then reuse the prepared reader for split planning. 
Scans without a single selected snapshot use the ordinary data path. The 
search-mode settings and coverage calculations stay as defined today.
   
   ### Tests
   
   - Added a deterministic concurrent-append regression for `full` and 
`detail`. Both cases reproduced the missing row before the fix and passed after 
it.
   - Ran 202 targeted `paimon-core` tests with the standard build checks 
enabled, covering global indexes, snapshot selection, incremental scans, and 
audit log reads.
   


-- 
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