Re: Caching merged diffs

[Piotr Piastucki <leech miranda gmail com>]

Kai Willadsen wrote:
> I'd be interested to know how you did these tests. I've had real
> problems getting reliable results when testing performance
> improvements.
>
>   

I just measured the time taken by _update_highlighting(). This time it 
was quite easy as the results are 100% reproducible, which is not always 
the case. E.g. sometimes I need to add some "warm-up" calls as the first 
executions of methods tend to be much slower.

One thing I have just noticed about my patch - it is a bit more flexible 
when the highlighting code changes (remember 
http://bugzilla.gnome.org/show_bug.cgi?id=587410 ?) as caching does not 
rely on the sequence matching algorithm.

Regards,
Piotr