facebook.py

import facebookscroller
import requests
import json
import time
import os 
cwd = os.getcwd()


facebook_email    = "apjanco@uchicago.edu"
facebook_password = "password"
target_profile    = "maria.tselovatova"


# login to Facebook account
facebookscroller.login(facebook_email, facebook_password)

# scroll the account so we can access all friends
facebookscroller.scroll_user_account(target_profile,friends=True)

# grab the Selenium driver variable so we can use it
driver = facebookscroller.driver

# this grabs a content block
friends = driver.find_elements_by_class_name("_698") 

friends_list = []
for friend in friends:
    row = {}
    name = friend.find_elements_by_class_name("fsl")[0]
    row['name'] = name.text
    row['url']  = name.find_element_by_tag_name("a").get_attribute("href")
    profile_image = friend.find_elements_by_class_name("img")[0].get_attribute("src")
    row['profile_image'] = profile_image
    r = requests.get(profile_image, allow_redirects=True)

    #check if user profiles directory, add if not exists and save photos to own directory
    if not os.path.isdir(cwd + '/' + target_profile +'__images'):
        os.mkdir(cwd + '/' + target_profile +'__images')
    open(cwd + '/' + target_profile +'__images/' + name.text +'.jpg', 'wb').write(r.content)
    #print(name.text,row['url'], row['profile_image'])
    friends_list.append(row)

print("[*] Discovered %d friends for %s" % (len(friends),target_profile))

facebookscroller.save_friends_list(target_profile,friends_list)
        
    
# close our browser
driver.quit()

facebookscroller.py

import time
import random
import codecs
import csv
import json

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
from selenium.webdriver.chrome.options import Options
from selenium.webdriver.common.action_chains import ActionChains

import sys
import os 
cwd = os.getcwd()
#reload(sys)
#sys.setdefaultencoding("utf8")

chrome_options = Options()
chrome_options.add_argument("--disable-notifications")

driver = webdriver.Chrome(executable_path="/home/ajanco/Documents/chromedriver",chrome_options=chrome_options)

#
# Login to the Facebook account
#
def login(facebook_email,facebook_password):
    
    driver.get("https://www.facebook.com/")    

    # login to Facebook
    elem = driver.find_element_by_id('email')
    elem.send_keys(facebook_email)
    elem = driver.find_element_by_id('pass')
    elem.send_keys(facebook_password)
    
    elem.send_keys(Keys.RETURN)
    time.sleep(1)

    return

#
# Scroll a user account to load friends, posts, etc.
#
def scroll_user_account(target_profile,friends=False):
    
    # browse to the target's friends list
    if friends == False:
    
        profile_url = "https://www.facebook.com/%s" % target_profile
    
    else:
        
        profile_url = "https://www.facebook.com/%s/friends" % target_profile
    
    driver.get(profile_url)
    time.sleep(5)

    # scroll to the bottom of the page
    driver.execute_script("window.scrollTo(0,document.body.scrollHeight);")
    time.sleep(1)
    
    # check number of elements
    element_size       = len(driver.find_elements_by_xpath("//*"))
    friends_count = len(driver.find_elements_by_class_name("_698")) 
    stop = False
    while stop is False:
        
        element_size_start = element_size
        friends_count_start = friends_count
        
        # scroll to the bottom again
        driver.execute_script("window.scrollTo(0,document.body.scrollHeight);")                   
        
        # count the number of elements
        element_size =  len(driver.find_elements_by_xpath("//*"))
        friends_count = len(driver.find_elements_by_class_name("_698")) 
        time.sleep(1)     
        
        # scroll to the top of the page
        driver.execute_script("window.scrollTo(0,0);") 
        
        # if we have more elements loaded repeat scrolling
        
        if element_size <= element_size_start:
            url = 'https://www.facebook.com/' + target_profile + '/photos'
            try:
                stopper = driver.find_element_by_xpath('//a[@href="'+url+'"]')
                stop = True
            except:
                continue
            spinners = driver.find_elements_by_class_name("_359")
            
            # find the loading animation element
            if len(spinners):
                
                for spinner_element in spinners:
                    
                    scroll_into_view(spinner_element)
                
                time.sleep(1)
                
            else:
                
                # we couldn't find any more loading animations so quit
                break
            
    return

#
# Helper function to scroll an element into view
#
def scroll_into_view(element):
    
    scroll = ActionChains(driver).move_to_element(element)
    scroll.perform()    
    
    return
#
# Saves a friends list to CSV
#
def save_friends_list(target_profile,friends_list):
    
    # open a CSV file for writing and dump out the results
    headers = ["name","url","profile_image"]
    
    with open("%s-friends.csv" % target_profile,"w") as fd:
        
        spreadsheet = csv.DictWriter(fd,fieldnames=headers)
        
        spreadsheet.writeheader()
        
        for friend in friends_list:
            print(friend)
            
            spreadsheet.writerow(friend)
    
    print("[*] Wrote out %d friends for %s" % (len(friends_list),target_profile))    
    
    return