class HtmlDom:
def __init__(self, url):
try:
f = file(url)
data = f.read()
f.close()
except IOError, e:
try:
result = fetch(url, agent=MOZILLA_AGENT)
data = result['data']
except:
raise IOError, 'invalid URL'
# create parser
parser = tidy.TreeBuilder()
parser.feed(data)
xmlText = _etree.tostring(parser.close())
#create the DOM
reader = PyExpat.Reader()
self.dom = reader.fromString(xmlText)
self.nss = {u'html': XHTML_NAMESPACE}
self.context = xml.xpath.Context.Context(self.dom, processorNss=self.nss)
