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

[martin rudalics]

> It looks like we could speed it up even more
> by producing this list
>
>     !      (setq whitespace-retval (cons (match-beginning 0) 
whitespace-retval))
>
> only if it is actually wanted.  Calling from whitespace-buffer
> could pass t as an optional argument saying do produce this list.

We could at least shorten such lists to a reasonably length.

>       The overhead for maintaining
>     these overlays may grow non-linearly with respect to their number.  Your
>
>      > 3. unzip and then visit the attached slowtst.el (I drag&dropped it)
>
>     has some 3500 lines which will produce approximately 7000 overlays.
>
> Is that still true after the changes you sent?

Yes.  But the file is pathological.

>     In addition, `whitespace-buffer' may unconditionally trigger a non-idle
>     timer which may trigger further executions of `whitespace-buffer'.  The
>     doc-string of the timer says:
>
>     "Timer object used to rescan the files in buffers that have been 
modified."
>
>     That's false, whitespace doesn't check whether a file is modified.
>
> Can you fix that up too?

I think (1) the timer should become an idle-timer, (2) quitting must be
permitted, and (3) one could use `buffer-chars-modified-tick' - to check
whether the latter works;-).  On the other hand, most of this could be
done via font-lock which would ask for a completely different design.