You could use eg. XML::Rules, somewhat like this:

use XML::Rules;

my $parser = XML::Rules->new(
  rules => [
    Item => sub { $_[1]->{Name} => $_[1]->{_content}},
      # take only the name and content from the <Item> 
      # and make it available using the Name as the key
    Id => 'content',
    DocSum => sub {
      print "Id: $_[1]->{Id}\nName: $_[1]->{Name}\nFoo: $_[1]->{Foo1},
+ $_[1]->{Foo2}\n\n";
      return; # we are done with the <DocSum>, no need to keep the dat
+a
    },
  ]
);

$parser->parse($the_file);
[download]

Basically the Item and Id rules specify what and how do you want from those tags and the DocSum consolidates the data from the subgtags and processes it the way you need. The (posibly) important difference from XML::Simple is that you process the XML gene by gene, instead of parsing the whole file, creating a possibly huge datastructure and then processing that structure.

You may want to have a look at XML::Twig and CPAN::XML::LibXML as well.

Update 2007-6-14: fixed a typo in the code.

!/usr/local/bin/perl -w
use strict;
use warnings;

use XML::LibXML;
use XML::LibXSLT;

my $xml = q{<eSummaryResult>
  <DocSum>
   <Id>25</Id>
   <Item Name="Description" Type="String">Ableson Murine</Item>
  </DocSum>
  <DocSum>
   <Id>26</Id>
   <Item Name="Description" Type="String">yada yada</Item>
  </DocSum>
  <DocSum>
   <Id>27</Id>
   <Item Name="Description" Type="String">yada yada something else</It
+em>
  </DocSum>
</eSummaryResult>
};


my $xslt_stylesheet = q{<?xml version="1.0" encoding="UTF-8"?>
<xsl:stylesheet version="1.0"
  xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>
  <xsl:output method="xml" omit-xml-declaration="yes"/>

  <xsl:template match="/eSummaryResult/DocSum"><xsl:apply-templates se
+lect="Id|Item"/></xsl:template>
  <xsl:template match="Id"><xsl:value-of select="."/>, </xsl:template>
  <xsl:template match="Item[@Name='Description']"><xsl:value-of select
+="."/></xsl:template>
</xsl:stylesheet>
};


my $parser      = XML::LibXML->new();
my $xslt        = XML::LibXSLT->new();

my $style_doc   = $parser->parse_string( $xslt_stylesheet );
my $source      = $parser->parse_string( $xml );

my $stylesheet  = $xslt->parse_stylesheet( $style_doc );
my $results     = $stylesheet->transform( $source );

my $output      = $stylesheet->output_string( $results );

print $output;
[download]

:!./t2.pl

  25, Ableson Murine
  26, yada yada
  27, yada yada something else
[download]

Let me see. what would the code be for the very same thing using XML::Rules:

use XML::Rules;

my $xml = q{<eSummaryResult>
  <DocSum>
   <Id>25</Id>
   <Item Name="Description" Type="String">Ableson Murine</Item>
  </DocSum>
  <DocSum>
   <Id>26</Id>
   <Item Name="Description" Type="String">yada yada</Item>
  </DocSum>
  <DocSum>
   <Id>27</Id>
   <Item Name="Description" Type="String">yada yada something else</It
+em>
  </DocSum>
</eSummaryResult>
};

my $parser = XML::Rules->new(
  rules => [
    Item => sub { $_[1]->{Name} => $_[1]->{_content}},
    Id => 'content',
    DocSum => sub {
      print "$_[1]->{Id}, $_[1]->{Description}\n";
      return;
    },
  ]
);

$parser->parse($xml);
[download]

There's less code to get the job done without XSLT then there is to apply it. And I'm sure you'd get the same result using most other modules.

I find XSLT incredibly clumsy, blithering and unreadable.

Jenda


Support Denmark! Defend the free world!