bot politico

botpolitico.py

#!/usr/bin/env python
# coding=utf-8

import urllib2
import argparse
import facebook
import time
import random
import re
from lxml import html

URL = "http://www1.folha.uol.com.br/poder/?cmpid=menutopo"

# caputamos algumas noticias
def news_political(timeout=10):
    try:
        resp = urllib2.urlopen(URL, timeout=timeout)
    except urllib2.HTTPError as e:
        if e.code == 404:
            print u"Xii url não existe mais"
            return None
    except socket.timeout:
        print "Excedeu o tempo limite."
        return False
    except urllib2.URLError as e:
        print e
        print "Error na captura do processo:"
        return False
    else:
        return resp.read()

# filtramos o conteudo das noticias encontradas
def news_parser(content, expressoes=['temer', 'psdb', 'russomano']):
    news_filter = []
    noticias = html.fromstring(content).xpath("//a[contains(@href, 'http://www1.folha.uol.com.br/poder/2016')]")

    if noticias:
        for num, noticia in enumerate(noticias):
            for expressao in expressoes:
                if expressao in noticia.xpath("./@href")[0]:
                    news_filter.append((noticia.xpath("./@href")[0], re.sub("[\r\n\t]", "", "".join(noticia.xpath(".//text()")))))
        return news_filter
    else:
        print "Não foram encontrado notícias"
        return None

# publicamos no facebook
def public_facebook(token, link, description, timesleep=300):
    attachment =  {
        'name': 'Robo Político',
        'link': link,
        'caption': description[:20]+' ...',
        'description': description,
        'picture': ''
    }
    # print attachment
    graph = facebook.GraphAPI(token)
    status = graph.put_wall_post(message='Veja isso!!!', attachment=attachment)
    print "TROLANDO AMIGOS!!!"
    time.sleep(timesleep)
    return True

def main(token, timeout, expressoes):
    content = news_political()
    if content:
        news = news_parser(content, expressoes)
        print news
        if news:
            for n in range(len(news)):
                public_facebook(token, n[0], n[1], timeout)

if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('-t', '--token', type=str,
                        help="token facebook https://developers.facebook.com/apps",
                        required=True)
    parser.add_argument('-ti', '--timeout', type=int, default=10,
                        help="time wait to new post in minutes")
    parser.add_argument('-f', '--filter', nargs='+', type=str,
                        help="words for filter ex: 'pt e psdb' temer cunha bolsonaro",
                        required=True)

    args = vars(parser.parse_args())
    main(args['token'], args['timeout']*60, args['filter'])