[GitHub] [arrow] jorisvandenbossche commented on pull request #10241: ARROW-12645: [Python] Fix numpydoc validation

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

jorisvandenbossche commented on pull request #10241:
URL: https://github.com/apache/arrow/pull/10241#issuecomment-832462864


   Ensuring that all keywords are documented sounds useful to check for. So indeed, if there is one of the validation chekcs we find useful and is passing, we could at that point add it to CI to keep it passing?
   
   Looking at the actual output above:
   
   > pyarrow.parquet.write_table
   PR01: Parameters {'use_compliant_nested_type', 'where', 'row_group_size', 'use_byte_stream_split', 'compression_level', '**kwargs'} not documented
   
   It seems there are some false positives for this check. In this case, all those keywords (except for `**kwargs`) are actually documented, but they are missing a space before the colon (`keyword: type` instead of `keyword : type`). Numpydoc, which is used for the validation, is picky about this. But the doc build itself looks fine, for which we use sphinx.ext.napoleon.
   


-- 
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:
users@infra.apache.org