in reply to Benchmark, -s versus schwartzian
On my Windows XP Perl 5.8 system the Schwartzian runs about 5 times as fast as the sort (with the glob inside the loop).
Benchmark: running Ordinary, Schwartzian for at least 2 CPU seconds... Ordinary: 2 wallclock secs ( 0.42 usr + 1.75 sys = 2.17 CPU) @ 5 +.06/s (n=11) Schwartzian: 2 wallclock secs ( 0.68 usr + 1.45 sys = 2.13 CPU) @ 2 +3.91/s (n=51)
CountZero
"If you have four groups working on a compiler, you'll get a 4-pass compiler." - Conway's Law
|
---|
In Section
Seekers of Perl Wisdom