Something is fishy. I just ran this simple-minded thing and I'm,
again, getting better times for ord() than I am for unpack() on a
2.8GHz OSX iMac with 2.5.1. This is the iterate so many times you can
use your wristwatch method:
----
#! /usr/bin/env python
from os import system
from struct import unpack
print "unpack 1"
system("date")
for x in xrange(0, 100000000):
Value = unpack(">B", "a")
system("date")
print
print "ord 1"
system("date")
for x in xrange(0, 100000000):
Value = ord("a")
system("date")
print
print "unpack 3"
system("date")
for x in xrange(0, 100000000):
Value = unpack(">BBB", "abc")
system("date")
print
print "ord 3"
system("date")
for x in xrange(0, 100000000):
Value = (ord("a") << 16)+(ord("b") << 8)+ord("c")
system("date")
----
Unpack 1 is about 1m20s
Ord 1 is about 20s
Unpack 3 is about 1m20s
Ord 3 is about 1m03s
after averaging a few runs.
--
http://mail.python.org/mailman/listinfo/python-list
