Re: file name parsing to get the original file name

Some quick benchmarking shows your solution to be about half as fast compared to mine.

#!/usr/bin/perl

use strict;
use warnings;

use Benchmark qw /cmpthese/;

our @files = qw {
    /etc/passwd
    one/two/three/four/five/six.a
    file
    a/very/deep/file/indeed/deeper/than/you/may/think/really
};


cmpthese -5 => {
    abigail => 'foreach my $f (@files) {
                    my $fn = (split m{/} => $f) [-1]
                }',
    markm   => 'foreach my $f (@files) {
                    my ($fn) = $f =~ /([^\/]+)\z/
                }',
};
__END__
Benchmark: running abigail, markm for at least 5 CPU seconds...
   abigail:  5 wallclock secs ( 5.19 usr +  0.00 sys =  5.19 CPU) @ 68
+404.24/s (n=355018)
     markm:  6 wallclock secs ( 5.23 usr +  0.00 sys =  5.23 CPU) @ 34
+427.53/s (n=180056)
           Rate   markm abigail
markm   34428/s      --    -50%
abigail 68404/s     99%      --
[download]

Abigail

Comment on Re: file name parsing to get the original file name Download Code

Replies are listed 'Best First'.
Re: Re: file name parsing to get the original file name by MarkM (Curate) on Aug 20, 2003 at 20:30 UTC
Interesting. It looks like you've found yet another piece of Perl that isn't implemented in the most optimal manner. :-) Playing around, I found that on my system, the following tweak allows the 'single regexp match' to beat the 'split into a temporary list, and grab the last entry' approach by ~15%: `$f =~ /(?:.*\/)?(.+)/s; my $fn = $1;` I'm still surprised that Perl can match against '/' several times (split) faster than it can skip to the last '/' with a single rather simple match. It seems the Perl regular expression engine could still use a few optimizations. Until then, I suppose explicit optimization isn't that bad. Cheers, mark	[reply] [d/l]
Re: file name parsing to get the original file name by Abigail-II (Bishop) on Aug 20, 2003 at 21:22 UTC
I'm still surprised that Perl can match against '/' several times (split) faster than it can skip to the last '/' with a single rather simple match. I'm not surprised. `/(^\/+)\z/` is a rather complicated regex due to the character class being used. The optimizer can't figure out that is the same as "looking for the last slash". `m!/!` on the other hand is so simple, the optimizer recognizes it as searching for a fixed string. Abigail	[reply]