Perl XML::Twig

rangeshram · April 9th, 2010, 11:26 AM

Dear All,

I am new to the PERL using XML::Twig, we need to navigate each reference element which is parent and compare both the <title> and <original_text> elements. While doing so in my xml there are multiple <title> tags present in the reference element. Thus our code produce wrong output for comparing each <title> text with <original_text> text.

I have tried the below code:
Perl-Twig.pl

Code:

use strict;
use XML::Twig;

my @arrTitle;
my @arrOriginalText;

my $t= XML::Twig->new( twig_roots   => { 'title' => \&title_ext, 'original_text' => \&org_ext   }
);

  $t->parsefile( 'Test.xml');

    sub title_ext 
    { my( $t, $elt)= @_;
	push(@arrTitle, $elt->text);
      $t->purge;           # frees the memory
    }

    sub org_ext 
    { my( $t, $elt)= @_;
	push(@arrOriginalText, $elt->text);
      $t->purge;           # frees the memory
    }

Our Input: Test.xml

Code:

<?xml version="1.0" encoding="UTF-8"?>
<references>
<reference id="[1]">
    <book book_type="monograph">
      <book_metadata>
	<contributors>
		<person_name sequence="first" contributor_role="author">
			<given_name>L. I.</given_name>
			<surname>Rozonoer</surname>
			<suffix></suffix>
		</person_name>
		<organization sequence="first|additional" contributor_role="author|editor|chair|translator"></organization>
	</contributors>
	<titles>
		<title>Thermodynamics and Regulation of Biological Processes</title>
	</titles>
	<series_metadata>
		<titles>
			<title></title>
		</titles>
		<isbn></isbn>
		<doi_data>
			<doi></doi>
			<resource></resource>
		</doi_data>
		<comments>
			<comment></comment>
		</comments>
	</series_metadata>
      </book_metadata>
      <original_text>[11] L. I. Rozonoer, <i>Thermodynamics and Regulation of Biological Processes</i>, de Gryter, Berlin, 1984.</original_text>
    </book>
</reference>
<reference id="[2]">
    <journal>
      <journal_metadata>
	<full_title></full_title>
	<abbrev_title>Energie, Wirtschaftswachstum und technischer Fortschritt Phys. Bl.</abbrev_title>
	<issn></issn>
	<coden></coden>
      </journal_metadata>
      <journal_article>
	<titles>
		<title></title>
	</titles>
      </journal_article>
      <original_text>[1] R. K&#x00FC;mmel, D. Lindenberger, W. Eichborn, Energie, Wirtschaftswachstum und technischer Fortschritt, Phys. Bl., <b>53</b>, 869 (1997).</original_text>
    </journal>
</reference>
</references>

Is there any way to pass the argument as XPath like 'book/titles/title' and also for the instances 'Journal/titles/title'.

Kindly advise/help us on this.

Thanks,

PeterPeiGuo · April 10th, 2010, 12:08 AM

For what you are trying to do, you can simply use XML::Simple. use the below script to load your XML and dump its internal representation of it. This should help you understand how XML::Simple can be used, and make your decision - basically all you need to do is to iterate through that data strcture.

Code:

use strict;
use warnings;
use XML::Simple;
use Data::Dumper;

print Dumper(XMLin("foo.xml"));