ARROW-17256 [Python] Can't call combine_chunks on empty ChunkedArray #13757
Conversation
|
|
|
IMO: yes, combine_chunks (which boils down to Concatenate in C++) should just take a datatype, though that would be a larger change |
python/pyarrow/tests/test_array.py
Outdated
| @@ -3192,3 +3192,13 @@ def test_to_pandas_timezone(): | |||
| arr = pa.chunked_array([arr]) | |||
| s = arr.to_pandas() | |||
| assert s.dt.tz is not None | |||
|
|
|||
|
|
|||
| def test_chunked_array_can_combine_chunks_with_no_chunks(): | |||
There was a problem hiding this comment.
combine_chunks is tested in test_table.py, could you move your test there too?
…-combine-chunks-should-work-with-no-chunks
|
Benchmark runs are scheduled for baseline = d171b6c and contender = 210cf06. 210cf06 is a master commit associated with this PR. Results will be available as each benchmark for each run completes. |
https://issues.apache.org/jira/browse/ARROW-17256
Not 100% sure it should be fixed at this level. Should
combine_chunksreceive aDataTypeinstead?