google.py

#!/usr/bin/python
# -*- coding: utf-8 -*-

import re
import random
import urllib2
from urllib import urlencode

import irssi

agents = (
    'Mozilla/5.0',
    'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.0.1) Gecko/2008072612 GranParadiso/3.0.1',
    'Mozilla/5.0 (Windows; U; Windows NT 6.0; pl; rv:1.9) Gecko/2008052906 Firefox/3.0',
    'Mozilla/5.0 (Windows; U; Windows NT 5.1; pl; rv:1.9) Gecko/2008052906 Firefox/3.0',
    'Mozilla/5.0 (X11; U; Linux i686; pl-PL; rv:1.9) Gecko/2008061015 Firefox/3.0',
    'Mozilla/5.0 (Windows; U; Windows NT 6.0; pl; rv:1.8.1.15) Gecko/20080623 Firefox/2.0.0.15',
    'Opera/9.51 (Windows NT 6.0; U; pl)',
    'Opera/9.51 (X11; Linux i686; U; pl)',
    'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.9.1a1) Gecko/2008072306 Shiretoko/3.1a1',
    'Mozilla/5.0 (Windows; U; Windows NT 5.2; pl; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1',
    'Mozilla/5.0 (Windows; U; Windows NT 5.1; pl; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1',
    'Opera/9.52 (Windows NT 5.1; U; pl)',
    'Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.16) Gecko/20080703 Mandriva/2.0.0.16-1.1mdv2008.1 (2008.1) Firefox/2.0.0.16',
)
replacements = (('<b>', '*'), ('</b>', '*'), ('<em>', '*'), ('</em>', '*'), ('&#39;', '\''), ('&amp;', '&'), ('&quot;', '"'))

pattern = re.compile('<h3 class=r><a href="(.*?)" class=l.*?>(.*?)</a></h3>', re.DOTALL)

def search(data, server, witem):
    global formated
    query = data
    if not query: 
        return
    params = urlencode({'hl' : 'pl', 'q' : query})
    url = 'http://www.google.pl/search?%s'
    
    request = urllib2.Request(url % params)
    request.add_header('User-Agent', random.choice(agents))
    try:
        opener = urllib2.build_opener().open(request)
        content = opener.read()
        results = pattern.findall(content)
        opener.close()
    except:
        print "Brak wynikow"
        return

    if not results:
        print "Brak wynikow"
        return
    
    def r(txt): 
        for what, to in replacements:
            txt = txt.replace(what, to)
        return txt
    
    i = 1
    for url, title in results:
        formated[i] = "(%i) %s <%s>" % (i, r(title), url)
        i += 1

    print " | ".join(formated.values())

def send(data, server, witem):
    global formated
    if not formated:
        return
    if not data.isdigit():
        return
    if not data:
        return
    if not witem:
        return

    witem.command('msg %s %s' % (witem.name, formated.get(int(data))))
    

formated = {}
irssi.command_bind('sr', search)
irssi.command_bind('g', send)