Created
September 7, 2016 17:49
-
-
Save jaksi/046d8ee348d770b374ab510f33c5b8b6 to your computer and use it in GitHub Desktop.
Scrape ingatlan.com for new apartments and send notifications via Pushbullet
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python3 | |
import traceback | |
from time import sleep | |
from lxml import html | |
from pushbullet import PushBullet | |
def get_apartments(url): | |
print("GET {}".format(url)) | |
parser = html.HTMLParser(encoding='UTF-8') | |
tree = html.parse(url, parser=parser) | |
apartments = tree.findall('//table[@class=\'search-results\']/tbody/tr[@data-id]') | |
result = {} | |
for apartment_element in apartments: | |
elements = apartment_element.findall('td') | |
apartment = {} | |
address_element = elements[1].find('a') | |
address_element.make_links_absolute() | |
apartment['address'] = ', '.join([part for part in address_element.itertext() if not part.isspace()]) | |
apartment['url'] = address_element.get('href') | |
price_element = elements[2].find('.//span[@class=\'price-huf\']') | |
if price_element is None: | |
price_element = elements[2].find('.//span[@class=\'price-eur\']') | |
apartment['price'] = price_element.text_content() | |
apartment['size'] = elements[3].text_content() | |
apartment['rooms'] = "{} szoba".format(elements[4].text_content()) | |
result[apartment_element.get('data-id')] = apartment | |
button_next = tree.find('//a[@class=\'button next\']') | |
if button_next is not None: | |
button_next.make_links_absolute() | |
sleep(5) | |
result.update(get_apartments(button_next.get('href'))) | |
return result | |
def main(): | |
pushbullet = PushBullet('<pushbullet token>') | |
emails = [] | |
url = ('http://ingatlan.com/listasz/kiado+lakas+butorozott+csak-gepesitett+i-ker+ii-ker+ix-ker+v-ker+vi-ker+vii-ker+viii-ker+xi-ker+xii-ker+havi-60-90-ezer-Ft+25-m2-felett+i-ii-ix-v-vi-vii-viii-xi-xii-ker') | |
apartments = get_apartments(url) | |
print("{} apartments".format(len(apartments))) | |
cur = list(apartments.values())[0] | |
for email in emails: | |
pushbullet.push_link(cur['address'], cur['url'], | |
"{0[price]}, {0[rooms]}, {0[size]}".format(cur), | |
email=email) | |
while True: | |
sleep(5 * 60) | |
try: | |
new_apartments = get_apartments(url) | |
print("{} new apartments".format(len(new_apartments))) | |
if not new_apartments: | |
continue | |
for apartment in new_apartments: | |
if apartment not in apartments: | |
cur = new_apartments[apartment] | |
print("New apartment: {}".format(cur)) | |
for email in emails: | |
pushbullet.push_link(cur['address'], cur['url'], | |
"{0[price]}, {0[rooms]}, {0[size]}".format(cur), | |
email=email) | |
apartments.update(new_apartments) | |
print("{} apartments".format(len(apartments))) | |
except: | |
traceback.print_exc() | |
if __name__ == '__main__': | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Structure of ingatlan.com has changed. The original structure doesn't work anymore.