How to parse broken HTML with LXML
Don't just construct that parser, use it (as per the example you link to):
>>> tree = etree.parse(StringIO.StringIO(broken_html), parser=parser)
>>> tree
<lxml.etree._ElementTree object at 0x2fd8e60>
Or use lxml.html as a shortcut:
>>> from lxml import html
>>> broken_html = "<html><head><title>test<body><h1>page title</h3>"
>>> html.fromstring(broken_html)
<Element html at 0x2dde650>
lxml allows you load a broken xml by creating a parser instance with recover=True
etree.HTMLParser(recover=True)
You could use the same technique when creating the parser.