wangyum opened a new pull request, #18002:
URL: https://github.com/apache/iceberg/pull/18002

   `BaseContentScanTask.split()` wraps every splittable file in a 
`SplitScanTask` even when the file has no usable split offsets and already fits 
within a single split (i.e. `length() <= targetSplitSize`).
   For tables with many small files, this allocates one redundant 1:1 wrapper 
object per data file during scan planning, adding unnecessary driver heap 
pressure.
   
   This change short-circuits that case in `BaseContentScanTask.split()` and 
returns the task itself instead of constructing a
   `FixedSizeSplitScanTaskIterator` that would only ever produce a single, 
functionally-identical wrapper (same `start()`/`length()`/ 
`estimatedRowsCount()`).
   
   Closes #17999
   


-- 
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]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to