nph opened a new issue, #38770:
URL: https://github.com/apache/arrow/issues/38770

   ### Describe the bug, including details regarding any error messages, 
version, and platform.
   
   Filtering a record batch with a boolean mask in the form of a `ChunkedArray` 
results in a segmentation fault.
   
   Reproduction on PyArrow 14.0.1:
   ```
   import pandas as pd
   import pyarrow as pa
   
   # Create a simple record batch
   df = pd.DataFrame(range(1, 7), columns=['x'])
   record_batch = pa.RecordBatch.from_pandas(df)
   
   # Generate a boolean mask as a chunked array
   mask = pa.chunked_array([[True, False], [True, False], [True, False]])
   
   # Flatten the chunked array mask and use this to filter the RecordBatch - 
all good
   print(record_batch.filter(mask.combine_chunks()))
   # pyarrow.RecordBatch
   # x: int64 
   # ----
   # x: [1,3,5]
   
   # Try filtering with the original chunked array - segfaults
   print(record_batch.filter(mask))
   # libc++abi: terminating with uncaught exception of type 
std::bad_variant_access: bad_variant_access
   # Abort trap: 6
   ```
   
   Note - this problem doesn't occur when filtering a table.
   
   ### Component(s)
   
   Python


-- 
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