regex matching issue

rahulgsp83 has asked for the wisdom of the Perl Monks concerning the following question:

hi, i am trying to match below few lines in an xml file

<item name="pdf_link">
           <value>http://www.yahoo.com</value>

</item>
[download]

with this regex

if($line=~/(.*)(<item name=\"pdf_link\">\s*<value>)(.*)(<\/value>)(.*)
+/i)
[download]

the control flow is parsing that line ,but regex is not matching ,i need the value between <value> tags, ideally if i write print $3 it should give me that value but it printing blank , looking forward for some guidance from experts, regards, rahul

Comment on regex matching issue Select or Download Code

Replies are listed 'Best First'.
Re: regex matching issue (XML parse) by toolic (Bishop) on Mar 09, 2010 at 14:39 UTC
Using regular expressions to parse XML is more trouble than it's worth (see XML parsing vs Regular expressions). Consider using a parser such as XML::Twig: `use strict; use warnings; use XML::Twig; my $xmlStr = <<XML; <item name="pdf_link"> <value>http://www.yahoo.com</value> </item> XML my $twig= XML::Twig->new( twig_handlers => { item => \&item } ); $twig->parse($xmlStr); sub item { my ($twig, $item) = @_; if ($item->att('name') eq 'pdf_link') { print $item->first_child('value')->text(), "\n"; } } __END__ http://www.yahoo.com` [download]	[reply] [d/l]
Re: regex matching issue by LanX (Saint) on Mar 09, 2010 at 14:36 UTC
Do I understand you the `if` is true but $3 doesn't get printed? Without knowing the whole file I would recommend changing `(.*)` to `(.+?)` meaning "match at least one character but non-greedy"! In general you should consider using one of the XML-moduls on CPAN... Cheers Rolf	[reply] [d/l] [select]
Re: regex matching issue by BioLion (Curate) on Mar 09, 2010 at 14:51 UTC
Works for me: `my $line = <<'END' <item name="pdf_link"> <value>http://www.yahoo.com</value> </item> END ; if($line=~m/(.)(\<item name\=\"pdf_link\"\>\s\<value\>)(.)(\<\/valu +e\>)(.)/i){ print "3: $3\n"; } else { print "No match\n"; } __END__ 3: http://www.yahoo.com` [download] But as people have already pointed out, an XML parser is probably a more reliable bet! Just a something something...	[reply] [d/l]
Re: regex matching issue by Ratazong (Monsignor) on Mar 09, 2010 at 14:41 UTC
Just a guess: You call your variable $line, but you have a multi-line-input. Have you doublechecked the whole text is in that variable? HTH, Rata	[reply]
Re^2: regex matching issue by rahulgsp83 (Novice) on Mar 09, 2010 at 16:12 UTC
thanks for all the replies ,you are right ,i am having a multi line input ,i checked all the data is in that variable ,is there any special quantifier in case of multi which we need to care of? Regards, Rahul	[reply]
Re: regex matching issue by FunkyMonk (Bishop) on Mar 09, 2010 at 14:53 UTC
Works for me! `use Data::Dump 'pp'; my $line = q{<item name="pdf_link"> <value>http://www.yahoo.com</value> </item>}; pp $line=~/(.)(<item name=\"pdf_link\">\s<value>)(.)(<\/value>)(.) +/i; __END__ ( "", "<item name=\"pdf_link\">\n <value>", "http://www.yahoo.com", "</value>", "", )` [download]	[reply] [d/l]