help-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: emacs-24.1.1 slow scrolling in Fundamental mode with font lock disab


From: Eli Zaretskii
Subject: Re: emacs-24.1.1 slow scrolling in Fundamental mode with font lock disabled
Date: Mon, 12 Nov 2012 18:19:39 +0200

> From: Maxim Yegorushkin <address@hidden>
> Date: Mon, 12 Nov 2012 12:49:51 +0000
> 
> I have 9Mb text file with 90000+ lines in it. When opening this file in 
> emacs-24.1.1 and scrolling to line 5000 and onwards the scrolling 
> becomes very slow and there is a long delay after I press PageDown and 
> emacs updates the view.
> 
> I heard that font lock may cause this so I disabled font lock in that 
> buffer. Disabling font lock doesn't resolve the issue or speed things up 
> a bit.
> 
> The contents of the file look like this:
> 
> 2012-10-09 08:35:28.217216000 167.150000  44035
> 2012-10-09 08:35:28.217216000 167.200000   2859
>                                                  2012-10-09 
> 08:35:28.217216000 167.250000  54888
> 2012-10-09 08:35:28.233155000 167.150000  66035
>                                                  2012-10-09 
> 08:35:28.233155000 167.250000  44404
> 
> (long line wrapped in the above).
> 
> I.e. there is a timestamp on each line of it in the form of YYYY-MM-DD 
> HH:MM:SS.nnnnnnnnn.
> 
> I almost accidentally replaced all 2012-10-09 with YYYY-MM-DD in the 
> file and that fixed the issue, the scrolling became swift again.
> 
> Do you guys have any clues what may be causing this scrolling slowness 
> and why replacing ISO dates with YYYY-MM-DD makes scrolling so much 
> faster? (once again I have font lock disabled in that buffer).

Does scrolling become faster if you type this:

  M-x set-variable RET bidi-paragraph-direction RET left-to-right RET

If it does, then I believe this is an inefficiency in the
bidirectional display engine that is already fixed in the development
sources, and will be part of the upcoming Emacs 24.3 release.  If you
can try the development sources, I think you should see that this
problem is gone.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]