Scrapper for the-big-bang-theory.com

from multiprocessing import Pool
import re
from urllib import urlopen
from lxml.html import document_fromstring

def _get_viewers(tuple):#season, episode):
    season, episode= tuple
    print season, '\t', episode
    p=re.compile('Viewers: (?P<n>\d+(\.\d+)?)')
    url= 'http://the-big-bang-theory.com/episodeguide/episode/%d%02d'
    url= url % (season, episode)
    raw= urlopen(url).read()
    dom= document_fromstring(raw)
    ps= dom.cssselect('p')
    ps= [e.text_content() for e in ps]
    ps= [e for e in ps if 'Viewers' in e]
    if len(ps) == 0: return
    m= p.search(ps[0])
    return float(m.groupdict()['n'])

def get_viewers():
    res= {}
    l= [(s,e) for s in xrange(1,7) for e in xrange(1,30)]
    pool= Pool(20)
    v= pool.map(_get_viewers, l)
    return zip(l,v)