[GitHub] [beam] TheNeuralBit edited a comment on pull request #16066: [BEAM-13335] Use shorter, numerical indices for dataframe reads.

[GitBox <gi...@apache.org>]

TheNeuralBit edited a comment on pull request #16066:
URL: https://github.com/apache/beam/pull/16066#issuecomment-982775542


   Looks like this broke TaxirideIT:
   
   ```
   File "/usr/local/lib/python3.6/site-packages/apache_beam/dataframe/io.py", line 211, in _shift_range_index
   E             return df.set_index(df.index + offset)
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/frame.py", line 4614, in set_index
   E             index._cleanup()
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 552, in _cleanup
   E             self._engine.clear_mapping()
   E           File "pandas/_libs/properties.pyx", line 33, in pandas._libs.properties.CachedProperty.__get__
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 560, in _engine
   E             target_values = self._get_engine_target()
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 3903, in _get_engine_target
   E             return self._values
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/indexes/base.py", line 3897, in _values
   E             return self._data
   E           File "/usr/local/lib/python3.6/site-packages/pandas/core/indexes/range.py", line 164, in _data
   E             self.start, self.stop, self.step, dtype=np.int64
   E         OverflowError: Python int too large to convert to C long [while running '_ReadFromPandas/ParDo(_ReadFromPandasDoFn)/ParDo(_ReadFromPandasDoFn)
   ```
   
   https://ci-beam.apache.org/job/beam_PostCommit_Python36/4664/testReport/junit/apache_beam.examples.dataframe.taxiride_it_test/TaxirideIT/test_aggregation/
   
   The issue seems to be that this implementation maps on to an unsigned int64, but pandas indices are signed int64 by default. I can send a patch to fix.


-- 
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: github-unsubscribe@beam.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org