I do not think it measures what you think it measures.
Ah, good point. . . so we were actually measuring how long it takes to make zero replacements on the modified list, which explains the downward trend of the timings (as the modified list gets shorter).
Adding the list copy (using lst[:] in the call to the remove() method) gives me these numbers:
.1% lc: 2410 usec try: 617 usec 10% lc: 2020 usec try: 7340 usec 50% lc: 1780 usec try: 34300 usec 90% lc: 1450 usec try: 61000 usec
Right, that's *far* more like the numbers I was expecting :)
And now they demonstrate what they were meant to demonstrate - that there's a reason filtration is the recommended approach!
Cheers, Nick.
--
Nick Coghlan | [EMAIL PROTECTED] | Brisbane, Australia
---------------------------------------------------------------
http://boredomandlaziness.skystorm.net
--
http://mail.python.org/mailman/listinfo/python-list
