[T. V. Raman] read-only modes should be using buffer-disable-undo?

[Dave Love]

raman sent this to an rms-free list (as far as I know).  He has a good
point -- I'm surprised man doesn't already do that.  Maybe that's why
it's always seemed so slow.

I'm not convinced about this prompting about undo anyway.  I think it
should at least offer more information, like a hint that you probably
do want to discard it.  The question may be too obscure otherwise.

> --- Begin Message ---
>
>
>
> Subject: 
>
> read-only modes should be using buffer-disable-undo?
>
>
>
>
> Date: 
>
> Sun,  2 Jan 2005 08:56:03 -0800 (PST)
>
>
>
>
> I run 21.3.50 from CVS, and recent updates to simple.el 
> produce a yes/no dialog when the undo  information to save becomes
> large; this pointed out an interesting place where Emacs  libs could
> be optimizing but apparently aren't -- namely, cases like man-mode.
>
> as an exercise, try M-x man gcc
> and notice that GC gets triggered at least twice with yes/no dialogs
> for throwing away the undo information; but all this could be avoided
> if Man-mode first executed buffer-disable-undo on the buffer where the
> man page is being formatted.
>
> You could probably do something similar during the execution of a
> shell command in shell-mode/comint-mode and friends; this would win in
> the case of long running shell commands that produce lots of output a
> small chunk at  time.
>
> -- 
> Best Regards,
> --raman
>
>       
> Email:  address@hidden
> WWW:    http://emacspeak.sf.net/raman/
> AIM:    TVRaman
> PGP:    http://emacspeak.sf.net/raman/raman-almaden.asc
> IRC:    irc://irc.gnu.org/emacspeak
>
>
> --- End Message ---