GitHub action to normalize MyST content

[kdeldycke]

Description / Summary

I'm looking for a way to normalize a large batch of MyST content.

Value / benefit

The idea is to integrate that normalization as a GitHub action, so I'll have an automated linting stage that will reduce the maintenance burden on my documentation.

Implementation details

So far, I came up with this pandoc CLI:

$ find ./ -iname "*.md" -exec pandoc "{}" --output="{}" --from-gfm --to=gfm --standalone --columns=79 --tab-stop=2\;

It is already integrated in my workflow there: https://github.com/kdeldycke/meta-package-manager/blob/78dd5effd7776607adf08a086d1f387b8f989e00/.github/workflows/autofix.yaml#L52-L58

It works great but still has a bunch of edge-cases (see below).

Tasks to complete

I'll be happy to contribute my experiments to the MyST-Parser documentation. Is that project the best place to contribute this?

Still, there are a couple edge-cases on pandoc side to be addresses:

• Extra-space between fenced code block start and attributes jgm/pandoc#7616
• MyST directives CodeBlock nodes normalized to Code nodes jgm/pandoc#7622

[chrisjsewell]

Heya yep no need for pandoc: https://github.com/executablebooks/mdformat

[kdeldycke]

Heya yep no need for pandoc: https://github.com/executablebooks/mdformat

Oh 😮. Perfect. Exactly what I need! 😁 Sorry for the noise! 😅 Let's close this issue then.