<page id="1">
  <content><![CDATA[
    Ths is <div class="red">some HTML text</div>]]>
  </content>
</page>
[download]

No, it is not a matter of time. I need to keep all XHTML contents of <content> tags in a single place, and do not parse it into perl data structures.

--dda

If you are embedding data that has >'s and <'s, then you prolly wanna use the CDATA directive/option/thing within your xml to denote, "this is data of the XML document, not part of the XML structure".

If that's beyond your control, you can always create a SAX parser that does just what you want.

Or you can write some XSLT that transforms the content nested data into what I described above.

Bart: God, Schmod. I want my monkey-man.

This is not valid XML. You have a tag, "<content>" that contains both a value "This is", and a child tag "<div class="red">some HTML text</div>". Pick one, or hide the <> characters with &lt;&gt;, or do the right thing and use CDATA.

Update: I stand corrected. I just checked the XML spec (http://www.w3.org/TR/2004/REC-xml-20040204) and gellyfish and ktingle are correct. Sorry.

It actually is valid - an node is allowed to have mixed content. This snippet will give rise to a schema like:

<?xml version="1.0"?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema">
  <xs:element name="page">
    <xs:complexType>
      <xs:sequence>
        <xs:element name="content">
          <xs:complexType mixed="true">
            <xs:sequence>
              <xs:element name="div">
                <xs:complexType>
                  <xs:simpleContent>
                    <xs:extension base="xs:string">
                      <xs:attribute use="required" type="xs:string" na
+me="class" />
                    </xs:extension>
                  </xs:simpleContent>
                </xs:complexType>
              </xs:element>
            </xs:sequence>
          </xs:complexType>
        </xs:element>
      </xs:sequence>
      <xs:attribute use="required" type="xs:unsignedByte" name="id" />
    </xs:complexType>
  </xs:element>
</xs:schema>
[download]

However this is certainly not what was intended - if the contents of <content /> are to be taken literally it should be a CDATA section.

/J\

An element can have a value and a child element, check the XML spec. Its awkward, but valid XML.

Did you take a look at XML::Twig? AFAIK it's very flexible in parsing/filtering tags.

use XML::Twig;

my $xml = '<?xml version="1.0" ?>
<page id="1">
  <content>
      This is <div class="red">some HTML text</div>
  </content>
</page>';

my $twig = XML::Twig->new(
  twig_handlers => {
    content => sub { $_->print; print "\n";  },
  },
);

$twig->parse($xml);
$twig->purge;
[download]