Infer metadata's title from markdown h1?

[roll]

What version of starlight are you using?

0.35.2

What is your idea?

Hi! Is it possible to infer metadata's title from the first markdown h1 (or filename?)?

(I scanned the issues and discussions but could not find relevant request. I might be missing it)

Why is this feature necessary?

I work a lot with document conversions (e.g. jsonschema -> markdown) + link many documents from the repo (e.g. CONTRIBUTING.md) and every time runs into the missing title problem.

In my opinion it would be nice if standard markdown files (without frontmatter) are supported as well.

Do you have examples of this feature in other projects?

No response

Participation

• I am willing to submit a pull request for this feature.

[delucis]

Thanks for the feedback @roll!

I did some experiments to see if I could quickly unblock this use case, and while not impossible, it’s pretty ugly to do in userland (let me know if ugly is still better than nothing and I can share more 😅)

This would require some somewhat larger technical changes which we’ll need to think about, so I’ve kicked off a separate discussion to encompass all of those with more detail and context: #3461