Re: Fwd: Serious performace problems on Windows XP with new(!) GNU Emacs v22 (both patched and unpatched EmacsW32 were tried)

[Peter Tury]

2006/10/31, martin rudalics <address@hidden>:
> I'll send you my latest copy separately.

Thanks! Now I've tried (with patched EmacsW32 version CvsP061023 ) it
and found the followings:

1. starting Emacs with -Q and using the official whitespace.el (not
the .elc to be fair) whitespace-buffer on slowtst.el takes ~ 23
seconds on my PC.

2. The same with your latest whitespace.el takes ~ 8 seconds.

3. starting Emacs normally (no -Q) and using the official
whitespace.el: whitespace-buffer on slowtst.el takes ~18 seconds even
if "everything is turned off" (= in my .emacs:
'(whitespace-auto-cleanup nil)
'(whitespace-check-ateol-whitespace nil)
'(whitespace-check-leading-whitespace nil)
'(whitespace-check-trailing-whitespace t)
'(whitespace-global-mode t nil (whitespace))
also I see in customize-group whitespace:
Whitespace Check Spacetab Whitespace is off (nil)
Whitespace Check Indent Whitespace is off (nil)
Whitespace Auto Cleanup is off (nil)
Whitespace Display Spaces In Color is off (nil)
), so I would expect that whitespace-buffer does practically nothing
-- for 18 seconds?! (Also follows: all the regex searches + colorizing
takes alltogether 5 seconds?) What happens in that 18 seconds?

4. normal (= not -Q) Emacs run with your latest whitespace.el produces
immediate answer: "Whitespace: this file is clean." with the settings
detailed in 3. on slowtst.el (for whitespace-buffer).

So for me it seems your patches improve whitespace.el a lot and fix
another bug (=doing nothing for 18 seconds?).

Br,
P