yihua opened a new pull request, #981:
URL: https://github.com/apache/hudi-rs/pull/981

   ## Description
   
   Concurrent reads on one `Table` handle shared the file system view's 
partition to file group map, so a filtered read's stats-pruned file groups 
could replace an unfiltered read's before it collected its slices, and that 
read returned part of the table. The map was reloaded on every call and never 
served as a cache, so each `get_file_slices` call now builds its own and passes 
it to slice collection. Partitions are sorted after listing so every read plans 
them in the same order.
   
   Closes #921
   
   ## How are the changes test-covered
   
   - [ ] N/A
   - [x] Automated tests (unit and/or integration tests)
   - [ ] Manual tests
     - [ ] Details are described below
   
   `test_concurrent_reads_on_one_handle_plan_independently` runs unfiltered and 
filtered reads concurrently on clones of one handle; on main the unfiltered 
read returns 300 to 340 of 400 rows.
   


-- 
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