Hi! We got a copy of a derivation of the GNU sort(3) program from Alexa. They've done some improvements to handle large collections of files better, it seems, and use a specialized qsort. But most interestingly, they have a merging algorithm that's O(N*log(M)) rather than O(N*M), where N is the number of lines output, and M is the number of input files. You may want to consider that part for addition to future versions of sort().
I'm attaching the source (too much has been specialized for diff to be useful), which includes the standard GNU header. If you want to contact the author before looking at the source, it's [EMAIL PROTECTED] Hope this is useful. -Lars
_______________________________________________ Bug-textutils mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/bug-textutils
