auctex-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Drop tarball releases and go ELPA only


From: Arash Esbati
Subject: Re: Drop tarball releases and go ELPA only
Date: Fri, 16 Aug 2024 09:13:39 +0200
User-agent: Gnus/5.13 (Gnus v5.13)

Hi Keita,

Ikumi Keita <ikumi@ikumi.que.jp> writes:

> OK, done.

🙏

> I suppose that usres of `use-package' who impose immediate loading on
> purpose think that
> "I always use package foo in my emacs session and want to use it without
> autoloading delay when I begin to work on matters requiring foo. For me,
> loading foo.el always happens sooner or later, and I don't care the
> increment of emacs startup time. Thus foo should be loaded at emacs
> startup."
> (For AUCTeX, such users might as well do
> (use-package latex
>   :after (tex))
> additionally?)

FTR, with my usual init file, `M-x emacs-init-time RET' returns 0.34 sec.
With (use-package tex) it becomes 0.55 sec.

My only concern is why AUCTeX should recommend users to pass other
package names to `use-package' instead of auctex which can be confusing?
So things like

  (use-package auctex
    :config
    (require 'tex))

would also work, right?

>> We could replace that with the FIXME added by Stefan M.:
>
>> (require 'tex-site
>>          (expand-file-name "tex-site.el"
>>                            (file-name-directory load-file-name)))
>
>> which is probably redundant: The feature tex-site is t when use-package
>> loads auctex.el and the form does nothing, but it is still much less
>> heavy than loading tex.el.
>
> It couldn't actually be in that form when we had to ensure that
> (unload-feature 'tex-site)
> to work for configure&make installation, but now we can.

Ok.

> Hmm, it would be better for local Git repo users who still have
> (load "~/Development/auctex/auctex.el nil t t)

Does this form work at all?  I tried that in my init file, restarted
Emacs, opened a .tex file and get:

  File mode specification error: (void-function LaTeX-mode)

> in their config, than just removing the relevant code as I
> suggested...?

In case I'm not wrong, can you show a patch about what you have on your
mind?

Best, Arash



reply via email to

[Prev in Thread] Current Thread [Next in Thread]