emacs-devel
[Top][All Lists]
Advanced

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

Re: best practices for implementing Emms configuration


From: Björn Bidar
Subject: Re: best practices for implementing Emms configuration
Date: Sat, 10 Dec 2022 01:46:22 +0200
User-agent: Gnus/5.13 (Gnus v5.13)

Stefan Monnier <monnier@iro.umontreal.ca> writes:

>> Write the configuration to the user's ~/.emacs in the same manner as "
>> ;; custom-set-variables was added by Custom." or ";; Added by
>> Package.el."  I'm not a huge fan of this option for the sole reason that
>> I don't appreciate packages automatically editing a file I manually
>> edit.
>
> AFAIK people who don't like/want Emacs to edit their manually-edited
> config file (such as myself) should and do set `custom-file` accordingly.
>
> So the better option seems very definitely to just save the config via
> Custom: that's what it's for.

I agree however those people that don't like Emacs to edit their
configuration file might also setup emms manually.

>> Write the configuration to ~/.emacs.d/emms/auto-config.el. Emms would
>> then look for that configuration file there during startup. This is
>> cleaner, and can have the benefit of Emms being configured without a
>> single line needing to be added by the user to their ~/.emacs.
>
> I personally dislike it when packages do such things because it's never
> completely clear how it interacts with the settings I put into my own
> manually-edited init file.
>
> Custom suffers from the same problems, of course, but once you've
> learned how it works, then this carries over to all the packages
> that use it.

If the package writes the file with setopt the behavior should be the
same as using custom, however no released version support that so far.

>> I know that a bunch of packages store information in ~/.emacs.d/, just
>> as Emms does.  But is it acceptable practice to store configuration
>> there too?
>
> There's no law against it.  But as a user, I'd vote against it.

Sometimes it makes sense, several packages to it e.g. by proving an
org-mode file setup some configuration value.

As long as the location of the file can be configured it does sound fine
to me, packages such as no-littering exist for a reason.

Br,

Björn



reply via email to

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