Question

I’m trying to parse a Tomboy note that has a link to another note inside it. The XML comes out looking like this:

<?xml version="1.0" encoding="utf-8"?>
<note version="0.3" xmlns:link="http://beatniksoftware.com/tomboy/link" xmlns:size="http://beatniksoftware.com/tomboy/size" xmlns="http://beatniksoftware.com/tomboy">
  <title>Our IP Blocks</title>
  <text xml:space="preserve"><note-content version="0.1">Our IP Blocks

What's <link:internal>in use</link:internal>?</note-content></text>
  <last-change-date>2009-03-10T10:24:36.3730770-04:00</last-change-date>
  <last-metadata-change-date>2009-03-10T10:24:36.3730770-04:00</last-metadata-change-date>
  <create-date>2009-03-10T10:23:14.2936280-04:00</create-date>
  <cursor-position>92</cursor-position>
  <width>450</width>
  <height>289</height>
  <x>0</x>
  <y>27</y>
  <open-on-startup>False</open-on-startup>
</note>

I’m parsing this with XML::Simple, and it’s pulling out the <link:internal /> node in to a separate object within perl.

EDIT: The resulting object (for the <text /> node looks like this. Note that ‘link:internal’ is a separate entity from ‘content’.

'text' => {
  'xml:space' => 'preserve',
  'note-content' => {
    'version' => '0.1',
    'link:internal' => 'in use',
    'content' => [
        'Our IP Blocks
        What\'s ',
        '?'
    ]
  }
}

Is this a bug, or am I crazy? All of the validators suggest that this is valid XML, but I’ve never seen it with a tag nested inside text like this before.

If it is a bug, does anyone know of another XML module that will get this right?

Answer 1

XML::Simple is not a proper parsing module for general XML parsing. It was designed with one purpose in mind, that is: parsing config files in XML format. So it simplifies the data somewhat — or a lot, depending on what you feed it. If you care about order of text and tags, you shouldn’t be using this module.