Re: multipart/related test

[Lars Magne Ingebrigtsen <larsi@gnus.org>]

Hrvoje Niksic <hniksic@srce.hr> writes:

> As strange as it may sound, XEmacs has a user-base of its own, and
> IMHO we cannot afford to change documented interfaces on Stallman's
> whims, at least not in areas where XEmacs implemented stuff before FSF
> Emacs.

Well, this is my whim, and not Stallman's.  :-)

Here's how I think it happened:  I was programming under XEmacs, and I 
wrote `(buffer-string some-buffer)'.  I then switched to Emacs, and
the thing bugged out, of course.  (As it would have done under XEmacs, 
but I guess I never tried.)  I grumbled and rewrote it as
`(save-excursion (set-buffer some-buffer) (buffer-string))'.

And then I sent a patch to Stallman to make Emacs XEmacs-compliant (as
I, in a particularly dizzy moment, thought it to be), and he accepted
it.

I still do think that my dizzyness yielded a better function than the
hard truth.  `buffer-string' and `buffer-substring' have different
feeling, and I think it makes sense to have them as separate
functions.  I mean -- you seldom want to give a buffer argument to
`buffer-substring', but you often want to give it to `buffer-string',
so having to write `(buffer-string nil nil some-buffer)' feels weird
to me. 

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@gnus.org * Lars Magne Ingebrigtsen