Re: multipart/related test

[Hrvoje Niksic <hniksic@srce.hr>]

Lars Magne Ingebrigtsen <larsi@gnus.org> writes:

> Kai.Grossjohann@CS.Uni-Dortmund.DE writes:
> 
> > Just installed pgnus-0.56 and w3-4.0pre.30 on Emacs 20.3, and this is
> > what happened:
> > 
> > ,-----
> > | Signaling: (wrong-number-of-arguments #<subr buffer-string> 3)
> > |   (buffer-string nil nil (mm-handle-buffer part))
> 
> My bad.  Emacs 20.3's `buffer-string' doesn't take a BUFFER
> parameter.
> 
> Er.  How weird.  `buffer-string' is an alias for `buffer-substring' in
> XEmacs 21.2.  Has it always been that way?  And jere I just convinced
> RMS to let `buffer-string' take an optional BUFFER parameter in Emacs
> 20.4, all for naught, it seems...
> 
> To clarify: In Emacs 20.4, `buffer-substring' takes mandatory START
> and END params, while `buffer-string' takes an optional BUFFER
> param.  In XEmacs 21.2, there's only one function, which takes three
> optional params -- start, end, and buffer.  I think the Emacs 20.4
> thing makes more sense.

So things were changed incompatibly -- again?  :-(

We can implement XEmacs `buffer-string' like this, but that's
monstrosity:

(defun buffer-string (&optional x y z)
  "@#$#@%#@$%#@$$@#$@#$@#"
  (if (or (bufferp x)
          (stringp x))
      (buffer-substring nil nil x)
    (buffer-substring x y z)))

After the `point-at-bol' fiasco, I'm not sure I want to see this code
in XEmacs.  :-(

-- 
Hrvoje Niksic <hniksic@srce.hr> | Student at FER Zagreb, Croatia
--------------------------------+--------------------------------
Be nice to your kids.
They'll choose your nursing home.