iajoiner edited a comment on pull request #9702:
URL: https://github.com/apache/arrow/pull/9702#issuecomment-986184490
@jorisvandenbossche I have added the `batch_size` option and set it to 1024
by default. As for the `stripe_size` issue it is more complex than what's in
your example which I do manage to replicate.
```python
# Shall this table be written as 41 batches?
>>> table = pa.table({'a': np.random.randn(1024 * 40 + 1)})
# with small stripe size, you actually get only 5 stripes
>>> orc.write_table(table, "test_orc_size.orc", batch_size=1024,
stripe_size=1, compression="zlib")
>>> orc.ORCFile("test_orc_size.orc").nstripes
5
```
Hence I don't really think we can easily test whether the `stripe_size` and
`batch_size` actually work properly by manually computing `nstripes` from the
table, `stripe_size` and `batch_size`. However what we can do is importing
`pyorc` if we are willing to add to `requirements-test.txt`, use the same
settings to write an ORC file and then compare the `nstripes`. `pyorc` and
`pyarrow.orc` should produce the same outcome.
--
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]
