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

Excuse a newbie question, but I'm baffled. Assuming I have a single line of XML, how do I pull out all values of the same tag name? ie. 
my $s = '<blah>red</blah><blah>white</blah><blah>blue</blah>';
print $s, $/;
while (/<blah>(.*)<\/blah>/) {   # this RE isn't right...
    print $1, $/;
}

[download]
Thanks for any help you can provide.

Replies are listed 'Best First'.
Re: non-greedy incremental regular expression?
by virtualsue (Vicar) on May 14, 2005 at 08:03 UTC
    If you intend to do much with XML, you will probably find the XML modules on the CPAN very useful. Here is a little sample of what XML::Simple can do: 
    #!/usr/bin/perl
use warnings;
use strict;

use XML::Simple;
use Data::Dumper;

my $s = '<foo><color>red</color><color>white</color><color>blue</color
+></foo>';
print $s,"\n";
my $parsed = XMLin($s);
print Dumper($parsed);

    [download]
    Voilà: 
    <foo><color>red</color><color>white</color><color>blue</color></foo>
$VAR1 = {
          'color' => [
                       'red',
                       'white',
                       'blue'
                     ]
        };

    [download]
    In english, $parsed is a hash containing one key, 'color', whose value is a reference to an array containing the values 'red', 'white', and 'blue'. E.g. 
    print $_,"\n" for ( @{$parsed->{color}} )

    [download]
[reply]
[d/l]
[select]
Re: non-greedy incremental regular expression?
by gopalr (Priest) on May 14, 2005 at 04:11 UTC

    Use .*? for immediate match

    See the changes below of your code: 

    my $s = '<blah>red</blah><blah>white</blah><blah>blue</blah>';

while ($s=~/<blah>(.*?)<\/blah>/)
{   # this RE isn't right...
    print "\n$1";sleep 1;
}

    [download]

    Update:: Remove /g

[reply]
[d/l]
[select]
Re: non-greedy incremental regular expression?
by eibwen (Friar) on May 14, 2005 at 04:02 UTC

    You are right to suspect that your regex is greedy: /<blah>(.*)<\/blah>/ matches the entire string, capturing:

    red</blah><blah>white</blah><blah>blue

    Read perlre for more about greedy regexes, but suffice it to say, I think you want /<blah>(.*?)<\/blah>/

[reply]
[d/l]
[select]
Re: non-greedy incremental regular expression?
by prasadbabu (Prior) on May 14, 2005 at 04:17 UTC

    TIMTOWTDI

    while ($s =~ /<blah>((?:(?!<\/blah>).)*)<\/blah>/)
 {       print $1, $/;
 }

    [download]

    Prasad

[reply]
[d/l]
Re: non-greedy incremental regular expression?
by TedPride (Priest) on May 14, 2005 at 15:41 UTC 
    my $s = '<blah>red</blah><blah>white</blah><blah>blue</blah>';
print $s, $/;
print $1, $/ while $s =~ /<blah>(.*?)<\/blah>/g;

    [download]
[reply]
[d/l]

Back to Seekers of Perl Wisdom