Bolha parser/pajek - prejemanje obvestil o novih oglasih

Prehitro boš blokiran s strani bolhe, ker je takšno početje trivialno identificirati.

amacar je 12. mar 2018 ob 18:20 izjavil:

Kakšen argument je pa to Ko najdeš oglas ti skripta pošlje mail, sms in prav tako ne rabiš biti 24/7 pred računalnikom.

xardas je 12. mar 2018 ob 18:40 izjavil:

Zakaj pa ne bi, recimo, laufali tole na piju? Ali pa uporabili kakšno storitev, ki v takšni ali drugačni obliki to funkcionalnost že ponuja?

import bs4
import unicodedata
import urllib2
import json

import time
import datetime

from najdisi_sms import SMSSender

def get_timestamp(string):
    date_time = string.replace(" ob", "").strip()
    if ". " in string:
        return time.mktime(datetime.datetime.strptime(date_time, "%d.%m. %Y %H:%M").timetuple())
    else:
        return time.mktime(datetime.datetime.strptime(date_time, "%d.%m.%Y %H:%M").timetuple())


def bolha(sms, buzz):
    for i in xrange(1,10):
        try:
            html = urllib2.urlopen("http://www.bolha.com/rekreacija-sport/kolesarstvo/kolesa/cestna-kolesa/?listingType=list&page={}".format(i))
        except Exception, e:
            print e, "Page", i, "does not exist"
            break

        soup = bs4.BeautifulSoup(html, "lxml")
        contents = [unicodedata.normalize('NFKD', at.contents[2].replace("\t","").strip()).encode('ascii', 'ignore')
                    for at in soup.find_all("div", class_="coloumn content")]
        titles = [unicodedata.normalize('NFKD', at.find("h3").text).strip().encode('ascii', 'ignore')
                    for at in soup.find_all("div", class_="coloumn content")]
        urls = [at.find("a")["href"].split("//")[-1].split("&")[0]
                    for at in soup.find_all("div", class_="coloumn content")]
        prices = [unicodedata.normalize('NFKD', at.text).strip().encode("ascii", "ignore")
                  for at in soup.find_all("div", class_="coloumn prices")]

        zipped = zip(titles, contents, prices, urls)

        for ad in zipped:
            relevant_text = [x.lower() for x in ad[:2]]
            if any(word in " ".join(relevant_text) for word in buzz):
                extended = urllib2.urlopen("http://www.bolha.com/{}".format(ad[-1]))
                #parse ad
                soup_ext = bs4.BeautifulSoup(extended, "lxml")
                ad_info = soup_ext.find("div", class_="adInfo")
                info_paragraphs = ad_info.find_all("p")
                if len(info_paragraphs) > 3:
                    #we've got private auction
                    added = get_timestamp(info_paragraphs[1].contents[1])
                    modified = get_timestamp(info_paragraphs[2].contents[1])
        #                         print title, "\n", content, price_
                    diff = (time.time() - modified) / 3600
                    text_message = "B:{title} za {cena}\n{vsebina}".format(cena = ad[2],
                                                                         title = ad[0],
                                                                         vsebina = ad[1])
                    if diff <= 1:
                        sms.send("040******",text_message[:160])

if __name__ == "__main__":

    sms = SMSSender('username', 'password')
    buzz1 = ["iščem", "vse", "te", "besede"]
    bolha(sms, buzz1)