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

Re: emacs fan help with packaging .el



2009/1/4 Matthew Saltzman <mjs clemson edu>:
> On Sun, 2009-01-04 at 16:30 +0000, Jonathan Underwood wrote:
>> 2009/1/4 Chitlesh GOORAH <chitlesh goorah gmail com>:
>>
>> Another aside: Other the past few months I've noticed that many
>> packages don't create a separate emacs-foo subpackage but rather use a
>> trick with %triggerin to drop elisp files into place when emacs is
>> installed. I also think this should be discussed and documented in the
>> packaging guidelines as an alternative when a package only has 1 or 2
>> elisp files.
>>
>> A further aside: I think the emacs guidelines should only insist on
>> separate emacs-foo-el sub-packages when there are a large number of
>> .el files. Perhaps. Maybe.
>
> What happens when one installs a package with .el files and installs
> emacs later?  How would one go about getting the .el files for your
> package in that case?
>

See the triggerin scriptlet in eg. rpmdevtools - on emacs installation
symlinks aer created in the /usr/share/emacs/site-lisp tree to the
.el(c) files as needed.


> In the case of subpackages, I'd be able to just yum list \*emacs\* to
> see what I need to add after the fact.

Yep, agreed, and that's one of the reasons why, when i drafted the
emacs packaging guidelines I detailed creating a separate
emacs-fool-[el] subpackage. However, the ubiquity of the triggerin
approach in current packages suggests there is some merit. Personally,
I think it needs further discussion, and was planning to take it to
the packaging list when I have a free moment.


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