Abhi2939 commented on issue #41346:
URL: https://github.com/apache/beam/issues/41346#issuecomment-6896510748

   Hi @jk-shah, I'd like to work on this one.
   
   I reproduced both problems with the snippet in the description. For the 
`KeyError` message, I'll interpolate `self._row_key` instead of `row_key_str`.
   
   For the `None` key, I'd check for `None` before the `str()` call and skip 
the Bigtable RPC. I have one question about the behavior. I'm planning to 
follow the existing missing-row handling, so:
   - `ExceptionLevel.WARN` logs a warning and returns `request, beam.Row()`
   - `ExceptionLevel.RAISE` raises a `ValueError` with a clear message
   
   Or would you prefer that a `None` key always raises, regardless of 
`exception_level`? I'll go with the first option unless you say otherwise.
   
   I'll add unit tests for a `None` key at each exception level and for the 
missing-column error message, using the mocked table as in the reproducer.


-- 
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