pure julia mode markdown comment highlighting

[gabrielgellner]

Is it feasible to add support for the pure Julia form so that the markdown is highlighted nicely using the #' syntax in a .jl file?

[mpastell]

I'm not sure how it could be done. Let me know if you know of an Atom-package that does something similar and I can look into the code.

[gabrielgellner]

Could it be similar to how Markdown uses code fences to change the highlighting grammar? You would do some kind of match on the starts with #' and then change the grammar to markdown?
Maybe a region is needed .. in which case would it then be maybe possible to do a match on something like #=' so it would kind of work like inverse Weave mode ... you would do fences for the markdown documentation, with the default mode being julia.

I say this just because I like to have my scripts be valid julia so I don't need to weave at computers without latex/pandoc ... but maybe this is beyond the scope of this package.

A similar issue is:
zargony/atom-language-rust#36

which argues for looking at how html code highlights css and javascript.

Though I really have not idea how atom does its highlighting! Thanks for even considering it!

[mpastell]

Sorry for the delay in my reply.

The problem is that if you change grammar of comment lines to markdown they all get highlighted as markdown headers (they start with #).

Anyway I will keep this in mind if I find some free time.

[gabrielgellner]

That makes sense. I have moved to using #== , ==# blocks to enclose markdown. I have been trying to figure out how to change the grammar to just do the embedded highlighting for blocks of that form.

[tpoisot]

I'd like to revive this issue -- right now, I am writing the package of a manual using the #' syntax and it's OK, but not ideal because a lot of formatting is lost.
I was not aware of #== -- is it rendered as markdown when calling weave?
I'd be happy to work on a PR for this.

[mpastell]

It is currently not rendered as markdown, it would be great if you can contribute a PR. You just need to implement a new reader for the format and come up with a syntax for chunk options