Thanks to everyone.

I did my own benchmark based on your suggestions, and found the following:

#!perl -w

use strict;
use warnings;
use Benchmark 'cmpthese';

my $test = join '', map { ' ' x  (0 == int rand 9) . ',' . ' ' x ((0 =
+= int rand 3) * int rand 3) . $_ } 0 .. 1_000;

cmpthese(-10,{
  'test1' => sub { my $f = $test; $f =~ s/\s+,\s+|\s+,|,\s+/,/g; },
  'testA' => sub { my $f = $test; $f =~ s/\s+,\s+/,/g; $f =~ s/\s+,/,/
+g; $f =~ s/,\s+/,/g; },
  'test2' => sub { my $f = $test; $f =~ s/\s+,\s*|,\s+/,/g; },
  'testB' => sub { my $f = $test; $f =~ s/\s+,\s*/,/g; $f =~ s/,\s+/,/
+g; },
  'test3' => sub { my $f = $test; $f =~ s/\s*,\s*/,/g; },
});

[download]

Results using ActivePerl 5.10.0 on WinXP-SP3:

1_000:
        Rate test1 test2 test3 testA testB
test1  974/s    --  -22%  -56%  -66%  -71%
test2 1256/s   29%    --  -43%  -56%  -63%
test3 2221/s  128%   77%    --  -22%  -35%
testA 2852/s  193%  127%   28%    --  -16%
testB 3412/s  250%  172%   54%   20%    --

10_000:
        Rate test1 test2 test3 testA testB
test1   79/s    --  -24%  -62%  -72%  -77%
test2  103/s   31%    --  -50%  -63%  -70%
test3  205/s  161%   99%    --  -26%  -39%
testA  277/s  252%  169%   35%    --  -18%
testB  339/s  331%  228%   65%   22%    --

100_000:
        Rate test1 test2 test3 testA testB
test1  6.5/s    --  -26%  -67%  -76%  -81%
test2  8.9/s   36%    --  -55%  -68%  -74%
test3 19.6/s  200%  122%    --  -28%  -42%
testA 27.3/s  319%  209%   39%    --  -19%
testB 33.6/s  414%  279%   71%   23%    --

[download]

Results are listed from slower to faster regexp(s).

Tests 1 and 3 are regexps from my original question, and the intermediate 2's regexp was supplied by ig. A and B are 1 and 2 respectively, but splitted on alternations (replace in phases).

Possible quick conclusions:

  1. It's better to use backtracking than alternations.
  2. It's better to split a regexp with an alternation whenever is possible.
  3. Itīs better to replace a pattern by the same thing instead of doing some effort to avoid that.

Comments?

In reply to Re: Comparing Regular Expressions by vitoco
in thread Comparing Regular Expressions by vitoco

Title:
Use:  <p> text here (a paragraph) </p>
and:  <code> code here </code>
to format your post, it's "PerlMonks-approved HTML":


  • Posts are HTML formatted. Put <p> </p> tags around your paragraphs. Put <code> </code> tags around your code and data!
  • Titles consisting of a single word are discouraged, and in most cases are disallowed outright.
  • Read Where should I post X? if you're not absolutely sure you're posting in the right place.
  • Please read these before you post! —
  • Posts may use any of the Perl Monks Approved HTML tags:
    a, abbr, b, big, blockquote, br, caption, center, col, colgroup, dd, del, details, div, dl, dt, em, font, h1, h2, h3, h4, h5, h6, hr, i, ins, li, ol, p, pre, readmore, small, span, spoiler, strike, strong, sub, summary, sup, table, tbody, td, tfoot, th, thead, tr, tt, u, ul, wbr
  • You may need to use entities for some characters, as follows. (Exception: Within code tags, you can put the characters literally.)
            For:     Use:
    & &amp;
    < &lt;
    > &gt;
    [ &#91;
    ] &#93;
  • Link using PerlMonks shortcuts! What shortcuts can I use for linking?
  • See Writeup Formatting Tips and other pages linked from there for more info.