Re: [Devel] Use of qsort

[Just van Rossum]

David Turner wrote:

> The easy thing is to provide our own implementation of quick sort
> (or heap sort, shell sort, whatever seems simpler and not too quadratic),
> since this completely avoids the problem. Actually, there already is
> so rather "reusable" quick sort algorithm in the smooth rasterizer
> (using a custom algorithm is critical for performance here)

The sort used for the kerning table (which I think I also meant to add to FT
2.0, but I'm pretty sure I didn't) can be pretty simple: the main thing is that
it performs fast on already sorted tables, which is the usual case. Hm, I think
I actually check whether the table is already sorted before doing the sort at
all. It's a workaround for buggy fonts, so performance is not too important
here: you can always blame the font ;-)

Just