MichaIng

If it is the same with Raspberry Pi OS Bookworm, maybe the RPi community can help at their forum. Though when using the image with preinstalled RPi desktop, it uses...

It previously seems to have worked because the fragments were not actually checked: #1675 I also recognized that it fails for fragments in GitHub markdown files in some cases. E.g.:...

Since, AFAIK, there is no issue open regarding GitHub markdown fragment check issues, IMO best we keep this one open. Maybe rename it to "GitHub Markdown fragment checks fail" or...

Oh yeah, in the served HTML source, the HTML document is embedded there, in escaped form, like the HTML ID literally as `id=\"user-content-intro\"` etc. Even if parsing that one, the...

> both fragment styles should work according to the debugger Yes, but the HTML parser doesn't know about the script which makes the fragment without `user-content-` functional. But just in...

Hmm, quite a good idea. I wonder whether there are cases where the fragment is explicitly not intended to point to content in the Markdown document. I mean these pages...

> what does that mean? I mean that the logic would need to be implemented here: https://github.com/lycheeverse/lychee/blob/696a7ca/lychee-lib/src/checker/website.rs#L122 - Currently, it searches the HTML document for HTML element IDs. Instead if...

But that one has no `x-raw-download` header either. So yeah, the method described does not work for default index/readme pages in GitHub repos, only for pages of a particular document...

But probably easier to implement than the embedded JSON>HTML parser.

Elegant, I didn't know this remap feature 👍. Indeed we often use fragments for GitHub repo URLs, where the readme is shown by default and fragments hence work (in browser,...