def __init__(self, site_url, html):
language_table = stats_lib.set_up_character_ranges_table()
tags = gt.get_tags(html)
self.url = site_url
self.title = gt.get_title( tags )
self.outgoing_links = gt.get_links(site_url, tags)
self.outgoing_link_count = len( self.outgoing_links )
self.scripts = gt.get_scripts(tags)
self.number_of_scripts = len(self.scripts)
self.langs = dict()
self.hash = make_hash(site_url)
self.alphabets_on_site = []
for script in tags(['script', 'style']):
script.extract()
self.body = strip_non_space_whitespace( tags.getText() )
text = self.body
#print( text)
for ch in text:
lang = stats_lib.check_ranges_from_table( ch, language_table )
#print( ch , lang )
if lang in self.langs:
self.langs[lang] += 1
else:
self.langs[lang] = 1
for key in self.langs:
if self.langs[key] > len(text)/70 and key not in self.alphabets_on_site:
self.alphabets_on_site.append(key)
self.n_grams = stats_lib.count_n_grams( self.body, 5 )
self.symbol_freq = stats_lib.count_symbol_frequency( self.body )
self.symbol_entropy = stats_lib.calculate_symbol_entropy( self.symbol_freq )
self.raw_html = html
def increment_link_counter(self, table_name, url):
#print(url)
#url = url.replace("'", '') #shouldn't need this right?
request = "UPDATE " + table_name + " SET n_incoming_links = n_incoming_links + 1 WHERE hash='" + make_hash.make_hash(url) + "';"
self.cursor.execute(request)
return validate_func(url)
if __name__ == "__main__":
argparser = argparse.ArgumentParser(description="get the raw html content of a site.")
argparser.add_argument('-u', dest='url', default='None', help="The URL to try.")
argparser.add_argument('--output', dest='output', default='stdout', help="the type of ouptut. stdout or file")
argparser.add_argument("-f", dest='filename', help='the filename to write to')
args= argparser.parse_args()
#set output type, file or stdout
if args.output == 'file':
filename = ""
if args.filename:
filename = args.filename
else:
filename = make_hash( args.url )
args.output = open( DATADIR+filename+'.pu', 'w' )
else:
args.output = sys.stdout
# visit then write out
if validate_with(is_tip_top_level, args.url):
uni_version = get_html(args.url)
uni_version = unicodedata.normalize('NFC', uni_version).encode("ascii", 'xmlcharrefreplace')
args.output.write( args.url + '\n' )
args.output.write( uni_version )
else:
print "Sorry, requires URLS of the form 'http://www.<somesite>.<domain>' for now."
exit(-2)
##close that shit out.
