peterolkhov opened a new pull request, #1021:
URL: https://github.com/apache/arrow-nanoarrow/pull/1021

   AI assistance: Codex generated the guard and regression tests and ran the 
local checks below.
   
   `iter_py()` raises `IndexError` on an empty LIST or LARGE_LIST chunk because 
`_list_iter()` indexes empty start/end slices. A zero-row record batch with a 
list column also aborts direct stream iteration before later rows are returned.
   
   Return immediately for a zero-length list slice, matching `_map_iter()`. The 
regression tests cover both list offset widths, empty arrays and slices, and a 
real `RecordBatchReader` with empty batches before, between and after populated 
batches.
   
   Validation on macOS arm64 / Python 3.13.14 using a rebuilt and reinstalled 
Meson/Cython wheel:
   
   - New regression cases fail on the unchanged implementation and pass with 
the guard.
   - Python suite: 288 passed, 4 skipped.
   - Package doctests: 118 passed; compiled-module stubtests: all 8 passed.
   - Black 22.3.0, Flake8 7.1.1 and isort 5.12.0 pass for both changed files.
   


-- 
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