rbdpngn a90d675613 Move ecore_file_ls from an O(n^2) insertion sort to an O(n log n) heap sort, ... since we already have it available. Worked fine on OS X and had dj2 check on Linux to ensure filesystem ordering wasn't skewing the results. He ran a couple quick benchmarks on a 10k file directory. Insertion sort results 23:38 <@dj2> real 0m2.134s 23:38 <@dj2> user 0m1.880s 23:38 <@dj2> sys 0m0.120s Heap sort results 23:35 <@dj2> real 0m0.223s 23:35 <@dj2> user 0m0.072s 23:35 <@dj2> sys 0m0.052s 23:38 <@dj2> and thats on a 3ghz box, heh SVN revision: 18410		2005-11-10 05:44:41 +00:00
..
ecore	Move ecore_file_ls from an O(n^2) insertion sort to an O(n log n) heap sort,	2005-11-10 05:44:41 +00:00
edje	we dont need no steenkin move signals	2005-11-08 10:06:59 +00:00
eet	LEAK! fix.	2005-11-07 06:16:26 +00:00
embryo	Don't use strlen to check if a string is empty.	2005-11-04 08:16:24 +00:00
emotion	Don't use strlen to check for empty string.	2005-11-04 08:21:28 +00:00
evas	i think this will fix the layer thing nathan found	2005-11-08 11:59:21 +00:00