mf2199 edited a comment on issue #8457: [BEAM-3342] Create a Cloud Bigtable IO connector for Python URL: https://github.com/apache/beam/pull/8457#issuecomment-590610828 @chamikaramj PTAL. The long standing issue now seems to be resolved. The reason was the namespace conflict between the existing `bigtableio.py` file and the extra package necessary for running the integration test. As it turns out, the two must have different names, otherwise Dataflow discards the package and uses the existing file, which of course does not have the newly added classes and hence the `AttributeError: 'module' object has no attribute...` error. Apparently this was not the case when this PR was originally created, so now we also have a caveat: - Until the code is merged, the only way to run the test is to change the package name in the import directive, `from bigtableio import ReadFromBigtable`, to something different, and use the external tarball package named accordingly. Upon merge, using an extra package will no longer be necessary and the test should run as-is. This was confirmed by running a sequence of nearly identical tests back-to-back, the instructions to which I can provide separately. In an attempt to reduce the code changes to a bare minimum, the write part of the test has been discarded, as it would test an already accepted code anyway. Finally, I'd also suggest closing this PR and opening a new one, so to make things cleaner. Let me know if this is a viable option.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
