On Tuesday 09 December 2008 07:17:23 Joe Zeff wrote: > On 12/08/2008 [EMAIL PROTECTED] wrote: > > it seems to me that PAN could use some performance tuning. I cannot > > believe > > that it take so long to sort some 1000000 articles. It are just > > strings ? > > I take it that you either don't program, or have never studied sorting. > I don't program any more, but back when I did, I was very interested, > for some strange reason, in sorting. One thing I still remember is > this: as the number of items to sort goes up, the time goes up, and it's > never in a 1:1 ratio, although the best get gosh-darn close: triple the > number of items and the time goes up by a factor of about 3.39. And, > even if Pan does a pointer sort, eventually it's going to have to move > all those strings into the right order, and that's going to take a > little time, especially if they're on disk instead of RAM. I'm not > saying that Pan couldn't be faster, but when you're talking about a > million articles, it's going to take more than an eyeblink.
if you have a collection that is loaded and request a different sort (e.g. by date) it is fast ! It is just when you switch it is slow. So i think that the reading of the articles takes long (perhaps the reading/scanning of the data takes too long ?) or is it the putting in the GUI ? (all articles are loaded and not those that are shown/visible ? ) Anyway I have developed software for 25 years now and know that IF something is slow, there are ways to make things faster *except* if you use an algorithm or tool that has bottlenecks for the task at hand. Finally AFAIK pan1 did not have the problem. W _______________________________________________ Pan-users mailing list Pan-users@nongnu.org http://lists.nongnu.org/mailman/listinfo/pan-users