sassoftware / python-sasctl

Python package and CLI for user-friendly integration with SAS Viya
https://sassoftware.github.io/python-sasctl
Apache License 2.0
45 stars 40 forks source link

Documentation empty for model_repository, reports, report_images #175

Open yjching opened 1 year ago

yjching commented 1 year ago

The API documentation for the model_repository, reports & report_images services seems to be missing: https://sassoftware.github.io/python-sasctl/api/services/model_repository.html https://sassoftware.github.io/python-sasctl/api/services/report_images.html https://sassoftware.github.io/python-sasctl/api/services/reports.html

In the case of the model_repository service, the docstring on line 7 in services/model_repository.py is still read but not the rest of the code.

I tried cloning this repo to build the documentation using: tox -e py38-doc but sphinx returned the error: Extension error: Could not import extension numpydoc (exception: No module named 'numpydoc') Changing from numpydoc to napoleon then returned: Theme error: no theme named 'pydata_sphinx_theme' found (missing theme.conf?)

Even installing both of the above packages with pip install failed to resolve these sphinx errors.

In addition to understanding why the docs are missing it would be good to get assistance in resolving the above sphinx errors,

Thank you!

SilvestriStefano commented 11 months ago

Hello, for the error the simple solution would be forcing tox to install those packages before running the command. That is to say add the following to the tox.ini file in the deps section

doc: numpydoc
doc: pydata_sphinx_theme

The two packages are not Sphinx extensions so they must be installed separately.

The documentation of those services not being generated is due to numpydoc. I am not sure what goes wrong but if instead of numpydocwe use the spinx extension sphinx.ext.napoleon then the documentation gets build image

Note: If i run tox -e py38-doc I get hundreds of warning from Sphinx mostly due to referencing issues . If i don't use numpydoc i get about 60 errors less.