vibhatha commented on code in PR #33969:
URL: https://github.com/apache/arrow/pull/33969#discussion_r1099410974
##########
python/pyarrow/tests/test_dataset.py:
##########
@@ -5084,10 +5084,8 @@ def test_dataset_partition_with_slash(tmpdir):
read_table = ds.dataset(
source=path,
format='ipc',
- partitioning='hive',
- schema=pa.schema([pa.field("exp_id", pa.int32()),
- pa.field("exp_meta", pa.utf8())])
- ).to_table().combine_chunks()
+ schema=dt_table.schema,
Review Comment:
In a previous PR I added this test case. It works without providing the
schema (with the current change) and we don't need the change I did in the
`_ensure_scheme`. But if a user provide a schema in this method, I get nulls
for the `exp_meta` column after making this change. That was the bug
encountered. I wasn't quite confident changing the test case to make things
work. Do you agree with the C++ change to fix this bug? If so, for Python we
could just leave the `schema`, but in case someone provides it there, we could
run into trouble.
--
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]
