Programme recherche_voila_popularite.py

#!/usr/bin/python
# -*- coding: iso-8859-15 -*-
import urllib
import re
 

def combien_voila (requete) :
    requete = requete.replace (" ", "%20")
    requete = requete.replace ("\"", "%22")
    #url = "http://search.ke.voila.fr/S/voila?dt=*&image.x=0&bhv=web_mondial&image.y=0&kw=" + requete 
    url = "http://search.ke.voila.fr/S/voila?rtype=kw&rdata=" + requete
    url += "&profil=voila&bhv=web_fr"
    whole = ""
    try :
        f       = urllib.urlopen (url)
        d       = f.read ()
        whole  += d
        f.close ()
    except Exception, exc:
        print "erreur"
        
    str = "réponses"
    pos = whole.find (str)
    if pos == -1 : return 0
        
    tex = whole [pos-30: pos + 5]
        
    pos = tex.find ("sur")
    if pos >= 0 : tex = tex [pos+3:]
        
    res = ""
    for t in tex :
        if "0" <= t <= "9" : res += t
    tex = res
    
    try :
        nb = int (tex)
        return nb
    except :
        return 0
    
def combien_bing (requete) :
    requete = requete.replace (" ", "+")
    requete = requete.replace ("\"", "%22")
    url = "http://www.bing.com/search?q=" + requete + "&go=&form=QBLH&filt=all"
            
    whole = ""
    try :
        f       = urllib.urlopen (url)
        d       = f.read ()
        whole  += d
        f.close ()
    except Exception, exc:
        print "erreur"
    
    exp = re.compile ("sur ([Â0-9  ]+) r")#??sultats")
    nb  = exp.search (whole)
    if nb == None : return 0
    else : return int (nb.groups () [0].replace (" ", "").replace ("Â", "").replace (" ",""))
    
        
def combien_ebay (requete) :
    requete = requete.replace (" ", "+")
    requete = requete.replace ("\"", "%22")
    url = "http://annonces.ebay.fr/France?kw=" + requete + "&go=&form=QBLH&filt=all"
    whole = ""
    try :
        f       = urllib.urlopen (url)
        d       = f.read ()
        whole  += d
        f.close ()
    except Exception, exc:
        print "erreur"
    
    exp = re.compile ("\"regionCntDiv\">[(]([0-9 ]+)[)]</span>")
    nb  = exp.search (whole)
    if nb == None : return 0
    else : return int (nb.groups () [0].replace (" ", ""))
    



def combien_acteur (acteur) :
    r1 = combien_voila ("\"" + acteur + "\"")
    r2 = combien_bing ("\"" + acteur + "\"")
    r3 = combien_ebay ("\"" + acteur + "\"")
    return r1,r2,r3
   
if __name__ == "__main__" :
    l = ["harry potter"]   
    l += ["orlando bloom", "bruce willis", "harrison ford", "johnny depp", "sean connery", "daniel auteuil", "gerard depardieu", "juliette binoche", "catherine deneuve"]
    l += ["jodie foster", "michel blanc", "nathalie portman", "ewan mcgregor", "scarlett johansson", "mick jagger", "tim robbins", "robbin williams"]
    l += ["susan sarandon", "zooey deschanel", "evangeline lilly", "robert downey jr", "katie holmes", "tom cruise", "nicole kidman"]
    l += ["anne heche", "pierce brosnan", "calista flockhart", "portia de rossi", "lucy liu", "matt damon", "ben affleck", "reese witherspoon"]
    l += ["geoffrey rush", "hugh jackman", "robert redford", "clint eastwood", "woody allen", "kenneth branagh", "emma thompson"]
    l += ["maggie smith", "kristen dunst", "chloe sevigny", "jennifer beals", "mia kirschner", "karina lombard", "pam grier"]
    l += ["katherine moenning"]
    l += ["patrick timsit", "patrick bruel", "richard berry", "thierry lhermite", "sandrine kiberlain", "pascal gregory"]
    l += ["cecile de france", "julia roberts", "marion cotillard", "audrey tautou", "romain duris"]
     
    for i in l :
        print i, "\t", combien_acteur (i)