I am pretty hopeless at regexps; my eyes glaze over at anything more complicated than (?:foo). That said, I have a performance problem. I'm looking at several gigabytes of log files and I want to extract "interesting" lines. Stripping the problem down to the absolute barebones, the problem looks like this: 

#! /usr/bin/perl -w

my $target = shift || 'word';
my $re = qr/a=<(.*?$target.*?)>/;

while( <DATA> ) {
    print "[$1]\n" if /$re/;
}

__DATA__
a=<wordy>
b=<rappinghood>
a=<thisword>
b=<thatword>
a=<foreword>
b=<junk>
a=<nothing>
b=<word>
b=<wordplay>
a=<swords>
b=<end>

[download]

In other words, I'm interesting in a= lines that contain word (or whatever I specify) anywhere between angle brackets. If I do see it, then I want everything between the angle brackets. (And in other variants I might be interesting in a= and b=, but that's another story). The above program emits:

[wordy]
[thisword]
[foreword]
[swords]

[download]

I'm using .*? to elide the remaining characters between the my target and delimiters, however, when I look at the behaviour of the match engine (with use re 'debug') I see lots of backtracking.

My question is this: is it possible to write a regexp for this problem that does not involve backtracking? The other question is whether it improves performance to any significant degree...

The corollary is that, hopefully, given I understand the problem domain, the solution might help me understand regexps better, and possibly be able to transpose it to other problems in the future.

Thanks.

Update: (meta-note: hooray for being able to update SoPWs!) Duh! Seconds after creating this node, it of course dawns on me that I can write

my $re = qr/a=<(.*?$target[^>]*)>/;

This is probably half the issue... but what about the leading characters between the opening delimiter and the target?

In reply to Removing backtracking from a .*? regexp by grinder

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.