def __init__(self, http_resp):
# Save "_parse" reference
orig_parse = self._parse
# Monkeypatch it!
self._parse = lambda arg: None
# Now call parent's __init__
HTMLParser.__init__(self, http_resp)
# Restore it
self._parse = orig_parse
def __init__(self, http_resp):
# Save "_parse" reference
orig_parse = self._parse
# Monkeypatch it!
self._parse = lambda arg: None
# Now call parent's __init__
HTMLParser.__init__(self, http_resp)
# Restore it
self._parse = orig_parse
def __init__(self, http_resp):
# Create the proper parser instance, please note that
# the order in which we ask for the type is not random,
# first we discard the images which account for a great
# % of the URLs in a site, then we ask for WML which is
# a very specific thing to match, then we try text or HTML
# which is very generic (if we would have exchanged these two
# we would have never got to WML), etc.
if http_resp.is_image():
msg = 'There is no parser for images.'
raise BaseFrameworkException(msg)
elif self._is_wml(http_resp):
parser = WMLParser(http_resp)
elif http_resp.is_text_or_html():
parser = HTMLParser(http_resp)
elif self._is_pdf(http_resp):
parser = PDFParser(http_resp)
elif self._is_swf(http_resp):
parser = SWFParser(http_resp)
else:
msg = 'There is no parser for "%s".' % http_resp.get_url()
raise BaseFrameworkException(msg)
self._parser = parser
def test_parse_html_performance(self):
headers = Headers()
headers['content-type'] = 'text/html'
body = file(self.HTML_FILE).read()
url = URL('http://www.w3af.org/')
response = HTTPResponse(200, body, headers, url, url, charset='utf-8')
#self.measure_memory(1)
parsers = []
for _ in xrange(40):
p = HTMLParser(response)
p.get_dom()
#parsers.append(p)
# Clear any reference to the parser
#del p
#parsers = []
#self.measure_memory(2)
time.sleep(360)