Re: eglot-tests fail for emacs-31 branch

João Távora <[email protected]>
Newsgroups gmane.emacs.devel
Message-ID <CALDnm52DjphJu6EHNF5tS5QEQErhcw0RjXgnRTjydzqB19TEoA@mail.gmail.com>
Hi Michael,

On Tue, Jul 21, 2026 at 3:33 PM Michael Albinus <[email protected]> wrote:

> > I suspect that the typescript-language-server version was upgraded and
> > now requires something (namely typescript itself) that used to be there,
> > but now isn't.  If I'm right then the correct fix is to install
> > typescript.
>
> That's not the intention of the tests on emba. They shall prove, that
> there's no error even in case of an incomplete environment.

Well, I don't know how you draw the line between an incomplete
and broken environment, but here this seems very clearly the latter
to me: a typescript-language-server installation without typescript
is manifestly useless.  I don't know what happened to it, but my
local EMBA copy (cloned some months ago following the instructions
you pointed me to) has this:

root@59a84580aeb9:/checkout# apt search typescript
node-tslib/now 2.4.1-1 all [installed,local]
  Implementation of tslib for javascript

node-typescript/now 4.9.5+ds1-2 all [installed,local]
  TypeScript is a language for application scale JavaScript development

In it, the javascript Eglot tests pass without a problem.  So something
regressed in this environment somehow, no?

> > Or or delete typescript-language-server entirely (it's not very useful
> > without typescript, is it?).  The test will be skipped.
>
> Thanks.

Hope you're not misunderstanding here.  This test will be skipped
once you (or someone) deletes the defective typescript-language-server
installation.  We will lose a signal, but that's life.  I don't have
plans for changes to eglot-tests: I think the code is fine as is.

João
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.