Last active
July 11, 2017 18:39
-
-
Save cristianmiranda/a169e00d75207d2e7bb507a4b653dc02 to your computer and use it in GitHub Desktop.
This file contains 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
import sys | |
import time | |
import csv | |
from sys import stdout | |
from selenium import webdriver | |
from selenium.webdriver import FirefoxProfile | |
''' | |
# SELENIUM - GECKO | |
export PATH=$PATH:'/Applications/Geckodriver' | |
export PATH=$PATH:'/Applications/Firefox.app/Contents/MacOS' | |
''' | |
CSV_FILE = '/Users/cristianmiranda/Desktop/inmobiliarias.csv' | |
BASE_URL = 'https://inmobiliarias.enbuenosaires.com.ar/bienes-raices/capital-federal.html#summaryResults%3AorderBySelector=id~desc¤t_page={0}' | |
driver = None | |
def wait_until_visible_by_xpath(xpath): | |
global driver | |
i = 0 | |
visible = False | |
while not visible: | |
stdout.write("\r- Elapsed: {0} seconds - {1}".format(i, xpath)) | |
stdout.flush() | |
try: | |
driver.find_element_by_xpath(xpath) | |
visible = True | |
return visible | |
except: | |
time.sleep(0.2) | |
if i >= 20: | |
return False | |
i += 1 | |
stdout.write("\r\n") | |
stdout.flush() | |
def wait_until_hidden_by_id(id): | |
global driver | |
i = 0 | |
visible = True | |
while visible: | |
stdout.write("\r- Elapsed: {0} seconds - {1}".format(i, id)) | |
stdout.flush() | |
try: | |
element = driver.find_element_by_id(id) | |
if (not element.is_displayed()) or (i > 15): | |
visible = False | |
i += 1 | |
except: | |
time.sleep(0.5) | |
if i > 15: | |
visible = False | |
i += 1 | |
stdout.write("\r\n") | |
stdout.flush() | |
def main(): | |
global driver | |
profile = FirefoxProfile() | |
profile.set_preference("browser.download.panel.shown", False) | |
profile.set_preference("browser.download.manager.showWhenStarting", False) | |
profile.set_preference("browser.helperApps.neverAsk.openFile", "text/csv") | |
profile.set_preference("browser.helperApps.neverAsk.saveToDisk", "text/csv") | |
profile.set_preference("browser.download.folderList", 2) | |
profile.set_preference("browser.download.lastDir", "/Users/cristianmiranda/Downloads/DKT") | |
driver = webdriver.Firefox(firefox_profile=profile) | |
reload(sys) | |
sys.setdefaultencoding('utf-8') | |
# Access main page | |
driver.get("https://inmobiliarias.enbuenosaires.com.ar/bienes-raices/capital-federal.html") | |
wait_until_visible_by_xpath("*//nav/ul/li[9]/a") | |
pages = driver.find_element_by_xpath("*//nav/ul/li[9]/a").get_attribute("text") | |
print "Pages: {0}".format(pages) | |
with open(CSV_FILE, 'wb') as csvfile: | |
fieldnames = ['page', 'name', 'tel', 'email', 'site'] | |
writer = csv.DictWriter(csvfile, fieldnames=fieldnames) | |
writer.writeheader() | |
# Inmobiliarias | |
for i in xrange(0, int(pages)): | |
driver.get(BASE_URL.format(i)) | |
wait_until_hidden_by_id("loading") | |
wait_until_visible_by_xpath("//*[@class='businessheader']/a") | |
wait_until_visible_by_xpath("//*[@class='businessheader']/a") | |
inmobiliarias = len(driver.find_elements_by_xpath("//*[@class='businessheader']/a")) | |
# Acceder a inmobiliarias | |
for j in xrange(0, inmobiliarias): | |
with open(CSV_FILE, 'a') as csvfile: | |
writer = csv.writer(csvfile, delimiter=';') | |
driver.find_elements_by_xpath("//*[@class='businessheader']/a")[j].click() | |
wait_until_hidden_by_id("loading") | |
contacto = wait_until_visible_by_xpath("*//a[contains(., 'Ver Datos de Contacto')]") | |
if contacto: | |
driver.find_element_by_xpath("*//a[contains(., 'Ver Datos de Contacto')]").click() | |
wait_until_visible_by_xpath("*//dt[contains(., 'Telefono')]") | |
elements = len(driver.find_elements_by_xpath("*//dt[contains(., 'Telefono')]/../dd")) | |
name = 'N/A' | |
if elements >= 1: | |
name = driver.find_element_by_xpath("*//dt[contains(., 'Telefono')]/../dd[1]").get_attribute("innerText") | |
tel = 'N/A' | |
if elements >= 2: | |
tel = driver.find_element_by_xpath("*//dt[contains(., 'Telefono')]/../dd[2]").get_attribute("innerText") | |
email = 'N/A' | |
if elements >= 3: | |
email = driver.find_element_by_xpath("*//dt[contains(., 'Telefono')]/../dd[3]").get_attribute("innerText") | |
site = 'N/A' | |
if elements >= 4: | |
site = driver.find_element_by_xpath("*//dt[contains(., 'Telefono')]/../dd[4]").get_attribute("innerText") | |
writer.writerow([i, name, tel, email, site]) | |
driver.get(BASE_URL.format(i)) | |
wait_until_hidden_by_id("loading") | |
driver.close() | |
if __name__ == "__main__": | |
main() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment