feat: add implementation pages in the guidebook to support peer review

[lwasser]

In our slack, @coatless made some great suggestions as our current Editor in Chief regarding gaps in our guidebook related to our peer review process and how things could be more seamless as we work through peer review.

These guides are conceptually strong; but, leave some of the implementation examples MIA: https://www.pyopensci.org/python-package-guide/documentation/index.html https://www.pyopensci.org/python-package-guide/tests/index.html

I think that it would be better for the EiC to have more concrete examples in the guide so we can link to the guide (and we can also link out from the guide if there are better things to just point to in some cases).

But in the spirit of "one way to do things" having working examples (that are in our copier template) would be invaluable.

@coatless are examples of how to setup docs and how to setup tests the place where we should start? this would also be documentation for our copier template

or do you have anything to add / did I miss things that you think are missing here?

[pllim]

FWIW

• OpenAstronomy has packaging guide that uses cookiecutter at https://github.com/OpenAstronomy/packaging-guide
• Scientific Python has some guides here as well: https://learn.scientific-python.org/development/guides/packaging-simple/

[lwasser]

thank you for this @pllim !! I haven't looked closely at the openAstronomy template yet! The feedback we keep getting is that people want one way to do things. So while the scientific python's template and guide is great - people don't want all of the tabs and options.

So that is what our guide tries to do along with explaining things for users who are not packaging experts.

One middle ground is we create examples from our template BUT then cross link to Scientific Python's guide for multi tool options.

I did talk to Henry a bit about cross linking at pycon us! thoughts?

[pllim]

I seldom have to create a package from scratch but for the 1-2 times I had to do it, cookiecutter was nice. It had prompts and then give you a somewhat functional skeleton. Still have to go in and clean things up but better than writing up everything from scratch. Perhaps @Cadair could comment more on the OpenAstronomy one. 🙏 Hope this helps!

[Cadair]

The goal of the OpenAstronomy template is to provide exactly 0 options, and for you to end up with a package which looks like astropy/sunpy when you are done. I appreciate that's not the most modern way of doing things any more but that's the goal.