[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC] refactoring DEFUN
From: |
Alan Mackenzie |
Subject: |
Re: [RFC] refactoring DEFUN |
Date: |
Mon, 25 Mar 2013 13:54:05 +0000 |
User-agent: |
Mutt/1.5.21 (2010-09-15) |
Hello Dmitry.
On Mon, Mar 25, 2013 at 03:56:29PM +0400, Dmitry Antipov wrote:
> Although C preprocessor can't transform the text too much, it's
> still possible to use concatenation to avoid silly typing like:
> DEFUN ("foo", Ffoo, Sfoo, ...)
> and use:
> DEFUN ("foo", foo, ...)
> instead. The core change is simple, but the obvious rest is ~450K
> uncompressed (99.9% was generated by elisp program, BTW).
I'm against this change. We don't seem to need it. I think it will
cause more work than it will save, given how seldomly we write new
DEFUNs.
I think it is best to keep Ffoo (at least) explicit. It would be strange
indeed if grep finds uses of a DEFUN, but not the DEFUN itself. This
would be suboptimal.
This explicit Ffoo is used by (amongst others?) c-defun-name, a component
of `add-change-log-entry-other-window', C-x 4 a, which is used
frequently.
After this change, c-defun-name would need to be significantly
complicated in order to parse both the traditional and new versions of
DEFUN.
I am against this change.
> Dmitry
--
Alan Mackenzie (Nuremberg, Germany).