simple regex

Anonymous Monk has asked for the wisdom of the Perl Monks concerning the following question:

Replies are listed 'Best First'.
Re: simple regex by Paladin (Vicar) on Mar 03, 2006 at 17:51 UTC
The reason your RE is failing is because of the `/x` which makes the RE engine ignore all unescaped whitespace in your RE, like the space between `name=\"VERIFIER\"` and `value=`. Check out perlre for more info.	[reply] [d/l] [select]
Re: simple regex by McDarren (Abbot) on Mar 03, 2006 at 17:54 UTC
"always wondered if there's a list that contains all the special characters to watch out for" This is pretty much covered in perldoc perlre	[reply]
Re: simple regex by wfsp (Abbot) on Mar 03, 2006 at 18:29 UTC
I find that "simple regex" and HTML very rarely fit nicely together! You can never be sure what the HTML will actually look like as ptum mentions above. If your HTML is part of a file it would be well worth considering a parser. `#!/bin/perl5 use strict; use warnings; use HTML::TokeParser::Simple; my $html = '<input type="hidden" name="VERIFIER" value="076-62">'; my $attribute = 'value'; my $value; my $tp = HTML::TokeParser::Simple->new(\$html) or die "Couldn't parse string: $!"; while (my $t = $tp->get_token) { if ( $t->is_start_tag('input') and $value = $t->get_attr($attribute) ) { print "$value\n"; } }` [download]	[reply] [d/l]
Re: simple regex by explorer (Chaplain) on Mar 03, 2006 at 17:42 UTC
Try: `$page =~ m/name="VERIFIER" value="([-0-9]+?)"/;` [download]	[reply] [d/l]
Re^2: simple regex by ptum (Priest) on Mar 03, 2006 at 17:51 UTC
You may also wish to use the m/regex/si flags in case the HTML spans multiple lines or the VERIFIER name attribute is not upper-cased. No good deed goes unpunished. -- (attributed to) Oscar Wilde	[reply]
Re^2: simple regex by lo_tech (Scribe) on Mar 03, 2006 at 19:37 UTC
Who says 'Fun with regexes' is not a contact sport? `$text .= 'blah<input type="hidden" name="FOO" value="123-01">blah'; $text .= "\nblah\n"; $text .= 'blah<input type="hidden" name="VERIFIER" value="076-62">blah +'; $text .= "\nblah\n"; $page = (($text =~ m/name="VERIFIER"\ value="([-0-9]+?)">/xg),$1)[0]; print "$page";` [download] So, $page will end up as "076-62" in your example... and to hell with $1. If you wanted make it more interesting, you could replace the appropriate line with: `$page = (($text =~ m/<input\ type="hidden"\ name="VERIFIER"\ value="(\ +d+)\-(\d+)">/xg),$1.'-'.$2)[0];` [download] or `$page = (($text =~ m/<input\ type="hidden"\ name="VERIFIER"\ value="(\ +d+)\-(\d+)">/xg),$1.'-'.$2)[1];` [download] if you only wanted a portion of the value. (The '072' or '062' in your example.) FYI, the 'special character' that's goofing with you (at least as written in your example) is the space between `"VERIFIER"` and `value`. The `"` characters do not have to be backwhack-escaped, but that non-escaped space in the middle of your regex will cause the match to fail. Well, that's my $.02 worth. No, for refunds you'll have to check our customer service department.	[reply] [d/l] [select]
Re: simple regex by gube (Parson) on Mar 03, 2006 at 17:53 UTC
Hi try this, You have tried well it's not special characters problem. `$page =~ m/name="VERIFIER" value="(\d+-\d+)"/g; print $1;` [download]	[reply] [d/l]