CC:@zware@Haypo@methane

@methane That line is now folded.

Has anyone tested to make sure the docs build doesn't pick up and build README.rst? I'm not sure that it would be harmful, but would be unnecessary.

@zware nice catch!

@rogersachan There are two problem.

• make check reports error.-i README.rst should be added to rstlint argument in Makefile.
• make html reports warningREADME.rst:: WARNING: document isn't included in any toctree. Please addREADME.rst toexclude_patterns in conf.py.

(I feel markdown is better for README now. But I was too late.)

@methane I have made the requested changes.

@Haypo I have made the requested changes.

Thanks, I merged your PR.

Right nowhttp://sphinx-doc.org/ link displays "Diese Präsenz ist leider nicht verfügbar." error message in german, whereashttp://www.sphinx-doc.org/ redirects tohttp://www.sphinx-doc.org/en/stable/ which is the expected website. But I hope that the error is only temporary. Otherwise, we can fix that later. I reported this issue on IRC.

The Windows installer build picks up this readme file and distributes it asreadme.txt (seeTools/msi/exe/exe_files.wxs, which now needs fixing).

I vote we should keep both so that we can distribute the file without rst formatting.

(As an aside, we also read the /LICENSE file, so if someone wants to come in and "fix" that too then they'll also need to update the installer.)

The Windows installer build picks up this readme file and distributes it as readme.txt (see Tools/msi/exe/exe_files.wxs, which now needs fixing).

This issue is about Doc/README.txt, but Tools/msi/exe/exe_files.wxs seems to refer to README.txt (root directory).

See PR#2 which renamed README.txt to README.rst. This PR was followed by PR#70, PR#73 and#21.

Is it common to provide README as plain text on Windows?

Would it be possible to provide a README formatted as HTML? It would look nicer than plain text, no? It would also allow to keep new features like images (well, "badges"). For example, on Linux I'm running manually "rst2html README.rst README.html".

Ah, I see. This PR just modified the one at the root, which is why it showed up as the last commit for the file.

It doesn't matter whether it's common or not - it's a functional change that should have been on bpo and signed off by the RMs (who are responsible for keeping at least the header in the file up to date). This may just be Ned, depending on where the change was backported to.

I openedhttp://bugs.python.org/issue29579

Thanks @rogersachan for the PR, and@Haypo for merging it 🌮🎉.. I'm working now to backport this PR to: 2.7.
🐍🍒⛏🤖

Sorry, @rogersachan and@Haypo, I could not cleanly backport this to2.7 due to a conflict.
Please backport usingcherry_picker on command line.

I don't think that we need to backport this change to Python 2.7, so I removed the "needs backport to 2.7".