[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: lisp/generic.el patch.
From: |
Lute Kamstra |
Subject: |
Re: lisp/generic.el patch. |
Date: |
Tue, 15 Mar 2005 22:47:11 +0100 |
User-agent: |
Gnus/5.11 (Gnus v5.11) Emacs/22.0.50 (gnu/linux) |
Stefan Monnier <address@hidden> writes:
>>> When loading a byte-compiled file that used define-generic-mode, I
>>> think you'll get an error that generic-mode-list is not defined
>>> (because generic.el doesn't get loaded: the autoload caused
>>> generic.el to be loaded during byte-compilation, but it doesn't
>>> have any such effect when loading the macro-expanded code).
>
>> Good point. The same goes for calling generic-mode-internal. I'll
>> add autoload cookies to generic-mode-list and generic-mode-internal.
>
> Maybe a better choice is to put a (require 'generic) in the macro
> expanded code.
I thought about that. The difference is that adding (require
'generic) causes generic.el to be loaded the moment a file with a
compiled define-generic-mode call gets loaded. Using autoload cookies
delays loading generic.el until the generic-mode function is actually
called.
I prefer the latter. Maybe the mode function will never be called.
For example, it makes it possible to remove (require 'generic) from
generic-x.el. So you can just load generic-x.elc to define a bunch of
major modes (and add them to auto-mode-alist) and generic.el will only
be loaded when you actually use one of them.
Why do you prefer adding (require 'generic)?
Lute.
Re: lisp/generic.el patch., Richard Stallman, 2005/03/16
Re: lisp/generic.el patch., Lute Kamstra, 2005/03/17