-
-
Notifications
You must be signed in to change notification settings - Fork 5.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
The Doc Knight Rises #12435
The Doc Knight Rises #12435
Conversation
Could this wrap long lines in the output? Generally a lot nicer to edit that way. |
Yeah that would be better. Will look for a pandoc option to do that. |
Oops sorry about the assigning myself and then unassigning, stupid vimium tricked me. Also, @one-more-minute is the hero we |
There's still some (256) ````rst` in the docstrings. Is the objective to convert these too in this PR? This is great work!! |
@shashi Or at least, the one you need right now ;) Thanks @hayd. Not all docstrings can be converted because they contain cross references ( |
Do we have doctest capabilities in the base docsystem, or standard markdown syntax for them? I see Lexicon has something for that? I think it would be fine to have the doctest-running eventually dependent on a Julia package as a replacement for Sphinx and the JuliaDoc shim. |
Is the huge number of deletions in |
I think a lot of that is line count differences due to markdown not breaking long lines where RST was. You can do |
e11a5a5
to
30eff4b
Compare
@one-more-minute, what's going on this? |
@one-more-minute This is one of the things that is preventing the 0.4 release. Would it be useful for some of us to get on a hangout and figure out how to get things done? |
This is essentially already there, I just need to regenerate the manual from it and make sure nothing's going wrong in the round trip. The main remaining issue is the text-wrapping thing – I played around with pandoc to get that working but no permutation of the |
f585767
to
5bab547
Compare
I polished this up and regenerated the manual from the new md docstrings. I also fixed the code block syntax issues, and added a warning for autogenerated docs, which shouldn't show up in the rendered manual. Once this is in I'll put up examples for how to move methods to the right place. |
@one-more-minute this diff is what I get when running |
The whitespace thing is annoying. I'll run the conversion on a unix soon and see if that helps. I'll also avoid auto-converting things that pandoc isn't recognising correctly; thanks for looking over the diff. |
Perhaps having |
7b8eec5
to
7dfd5c6
Compare
Another day, another rebase. How is this looking now? |
Why are all the builds failing? |
7dfd5c6
to
a9a2e64
Compare
Pandoc seems to introduce some whitespace issues – fixed. |
@tkelman What next here? Should we rebase and get it ready? |
Let's merge this. The remaining issues are not going to get fixed until this gets merged. |
Reflecting my own view here but I am not sure what the remaining issues are and how I can help. Can I start making PRs moving documentation to method definitions now? Tell me what to do hehe. |
a9a2e64
to
4adfd31
Compare
Rebased. @KristofferC Working on that now ;) |
4adfd31
to
a2bedf3
Compare
@one-more-minute, thanks for all the effort you've put into this one! |
@MichaelHatherly Very close now! |
Wow, this will be nice! |
|
Hurray! |
What's the deal with the thousands of lines of |
This is a first (WIP) pass at converting as many of the docstrings as possible to Markdown. There's also an RST render backend for the Markdown parser so that we can keep the RST docs as they are while having docstrings in Markdown.
Once I've sorted out some code-block related issues I'll push the manual regeneration step so that people can check for information loss.