Speedy (lax?) little *generic* XML parser. It takes a funky approach and matches against substrs instead of gathering in the match, this made it over 20x faster in my test case. You can pick your own root node in the document as well..

There are no generic XML parsers. Just as a document either is XML or it isn't; something either parses XML, or it doesn't. Your so-called 'XML parser' doesn't.

There are no compromises.

<?xml version="1.0" encoding="UTF-8" ?>
<?xml-stylesheet type="text/css" href="style.css"?>
<one>
  <two foo="bar &quot;" baz:quux='4'>
    <three:toto wawa=''/>
  </two>
  <![CDATA[
    1 && 2 && 3
    Foo Bar <Baz></Foo><Bar /><Foo>
  ]>
  <!-- one two three -->
</one>
[download]

PS. And yes, I was the one who downvoted your node.