On Wed, Nov 23, 2022 at 1:34 PM Eli Zaretskii <eliz@gnu.org> wrote:

> > From: João Távora <joaotavora@gmail.com>
> > Date: Wed, 23 Nov 2022 12:36:37 +0000
> > Cc: Danny Freeman <danny@dfreeman.email>, Felician Nemeth <
> felician.nemeth@gmail.com>,
> >       58790@debbugs.gnu.org, Stefan Kangas <stefankangas@gmail.com>,
> >       Dmitry Gutov <dgutov@yandex.ru>, Eli Zaretskii <eliz@gnu.org>,
> >       Michael Albinus <michael.albinus@gmx.de>
> >
> > Yes,I think I follow.  To be clear I think the problem is somewhere in
> >
> > (defun eglot--path-to-uri (path)
> >   "URIfy PATH."
> >   (let ((truepath (file-truename path)))
> >     (if (url-type (url-generic-parse-url truepath))
> >         ;; Path is already a URI, so forward it to the LSP server
> >         ;; untouched.  The server should be able to handle it, since
> >         ;; it provided this URI to clients in the first place.
> >         truepath
> >       ...)
> >
> > So either url-generic-parse-url and url-type is fixed in url-parse.el,
> or
> > we must add some Windows-specific guards in eglot.el.  Or likely
> > both, since url-parse.el is not a :core ELPA package.
> >
> > Richard/Danny, can you perhaps come up with some patch?
>
> From where I stand, the solution was already proposed here:
>
>   https://debbugs.gnu.org/cgi/bugreport.cgi?bug=59338#53


Right. When I wrote that, I was unaware of this second bug report.


> You said you wanted to augment it with something that didn't depend on
> url-parse.el, because it wasn't in Emacs < 29, but I don't think I
> understand the concern, since AFAICT url-parse.el is in all versions of
> Emacs since at least Emacs 23.  So what am I missing?
>

Maybe I misunderstood, but it seemed like you were proposing that
we change url-parse.el so that  it isn't fooled into thinking windows path
names are URLs with the drive letter as the 'url-type'.

If that were the correct solution (I tend to think it is), then it would
_not_ suffice to fix Eglot on Emacsen older than 29, because url-parse.el
is not distributed separately as a :core ELPA package.

So unless we made url-parse.el such a package, we would need a Danny's
"kludgy" patch to eglot.el anyway.

In any case, I suggest that a solution be based on the patch shown in that
> bug#59338 discussion.
>

Having now read Danny's patch, I think it's good to commit.

João