Re: What to do about unmaintained ELPA packages

["João Távora" <joaotavora@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 3586 bytes --]

On Tue, May 31, 2022, 17:42 Philip Kaludercic <philipk@posteo.net> wrote:

To be specific, my issue was mentioned here:
> https://github.com/joaotavora/yasnippet/issues/1121 ("Avoid
> false-positives when expanding snippets").  In my case, I have managed
> to circumvent the issue by unbinding the tab key and relying in
> hippie-expand, but the solution is not elegant and I would have
> appreciated a discussion on the issue tracker.
>

Does it really matter where the discussion happens? The problem
here is not the place, it's the unavailability of brain-hours to dedicate
to your request.

The point was not to say that yasnippet has been abandoned, just to give
> an example where the liveliness of development could be questioned.
>

You don't have to sugar-coat it. Yasnippet's bug reports have gone
unanswered
at least for a year. They don't make it into my inbox, I'm not the
maintainer
anymore. Noam did a thorough and excellent job while he was maintaining it,
he's not available anymore. It's de facto unmaintained, though rather
stable
in its intended use case.  Of course Emacs is gigantic and everyone wants
everything, and so it might not be doing what you want.  I understand that.

In that sense, it'd be nice if yasnippet wasn't' the monolithic package it
is,
and 'snippet.el' is a step in that direction.  Then you could perhaps opt
out
(or in) to the multi-file, keybinding, abbrev-like functionality of
yasnippet.el.
Or simply compose 'snippet.el' bare-bones expansion/navigation with
whatever other productivity tool where you think  snippets make sense.
Eglot is one of those tools and it works well with yasnippet today.  I'd
love to
make it depend on simply 'snippet.el'.

> Is it just the general idea of abandonment of the fact that people's
> > issue reports go unanswered?  If the latter, then I think archiving to
> > the repo would make sense.  No more issue reports would come in and
> > one could advertise the Emacs bug tracker.  Would that improve things?
>
> I guess, but I think it would be preferable to call for someone to fork
> the package instead of shifting the burden of maintenance onto the emacs
> bug tracker.
>

Again, i don't quite understand how that magically solves the root problem
You're free to fork when you want, of course, but that normally happens
when you want to take the software in a new irreconcilable new direction.
Currently, yasnippet just needs (1) a maintainer that understands the
current functionality and "protects" it, fixing bugs, not introducing new
ones.
New features are probably not a great idea. (2) someone to develop
'snippet.el'
into production-ready state and make 'yasnippet.el' depend on it, then may
integrate make 'snippet.el' into a new package. The person of (1) and (2)
can
be the same.

> it motivates someone to pick up the maintenance of the existing
> > Yasnippet.
>
> Due to a sudden increase in free time available to me, I might consider
> this.
>

Great, if you're serious about it drop me a line stating more or less what
your plan is. I don't know if you've read the code: it's not exactly a work
of art (at least my parts aren't) :-) If you'd prefer a more palatable
challenge,
you could opt to simply develop the somewhat nicer 'snippet.el'.  It'd be
fine
to create your own repo to host and develop it (say in that shiny SourceHut
account where we discussed a bug the other day).

João

PS: in the meantime, it seems that Dmitry suggested what seems like a
reasonable way to address your request.

[-- Attachment #2: Type: text/html, Size: 5729 bytes --]