Save data about Facebook posts: "Date and time", 'User', 'Text', 'Reactions', 'Shares', 'Comments', 'URL' using general search by keyword

selenium_facebook_search.py

#!/usr/bin/python3
# -*- coding: utf-8 -*-

from selenium import webdriver
from selenium.webdriver import DesiredCapabilities
from selenium.webdriver.common.keys import Keys
from urllib import parse
import re
from time import sleep
from datetime import datetime
import xlsxwriter


desired_capabilities = DesiredCapabilities.PHANTOMJS.copy()
desired_capabilities['phantomjs.page.customHeaders.User-Agent'] = '(KHTML, like Gecko) Chrome/57.0.2987.133'


dr = webdriver.PhantomJS(executable_path='S:/Path to PhantomJS folder/phantomjs-2.1.1-windows/bin/phantomjs',
                         desired_capabilities=desired_capabilities)
dr.get("https://www.facebook.com/login.php")
login = dr.find_element_by_id("email")
login.clear()
login.send_keys("your email")
password = dr.find_element_by_id("pass")
password.clear()
password.send_keys("your password")
password.send_keys(Keys.RETURN)
query=parse.quote('keyword(s) to search'.encode('utf-8'))
dr.get("https://www.facebook.com/search/str/%s/stories-keyword/stories-public" % query)

deepness=0
while deepness<=2:
    dr.execute_script("window.scrollTo(0, document.body.scrollHeight);")
    sleep(2)
    deepness+=1
# deepness is the quantity of "scroll downs" the results page.
# Here it equals 2 (deepness<=2)
urls=dr.find_elements_by_css_selector("span.fsm.fwn.fcg a._5pcq")
links=[]
for url in urls:
    links.append(url.get_attribute("href"))
text=[]
date=[]
author=[]
reactions=[]
shares=[]
comments=[]
for link in links:
    dr.get(link)
    try:
        reactions.append(int(dr.find_element_by_css_selector("._4arz span").text))
    except:
        try:
            dr.find_element_by_css_selector(".UFILikeSentence").click()
            sleep(1)
            reactions.append(int(dr.find_element_by_css_selector("._4arz span").text))
        except:
            reactions.append(0)
    try:
        share=re.sub("\D", "", dr.find_element_by_css_selector(".UFIShareLink em").text)
        shares.append(int(share))
    except:
        shares.append(0)
    try:
        text.append(dr.find_element_by_css_selector("p").text)
    except:
        try:
            text.append(dr.find_element_by_css_selector(".hasCaption div").text)
        except:
            text.append('')
    try:
        author.append(dr.find_element_by_css_selector(".fwb.fcg").text)
    except:
        try:
            author.append(dr.find_element_by_css_selector(".profileLink").text)
        except:
            try:
                author.append(dr.find_element_by_css_selector(".fbPhotoContributorName a").text)
            except:
                author.append('')
    try:
        dr.execute_script("window.scrollTo(0, document.body.scrollHeight/4);")
        sleep(2)
        comment=dr.find_element_by_css_selector("span.fcg em")
        comment=re.sub("\D", "", comment.text)
        comments.append(int(comment))
    except:
        comments.append('')
    datestamp=int((dr.find_element_by_css_selector("abbr").get_attribute("data-utime")))
    date.append(datetime.fromtimestamp(datestamp))

workbook = xlsxwriter.Workbook('S:/Path to folder where results to be saved/Results.xlsx')
f = workbook.add_worksheet()
f.write_row(0, 0,["Date and time", 'User', 'Text', 'Reactions', 'Shares', 'Comments', 'URL'])
format_date = workbook.add_format({'num_format': 'dd.mm.yyyy hh:mm:ss'})
f.write_column('A2', date, format_date)
f.write_column('B2', author)
f.write_column('C2', text)
f.write_column('D2', reactions)
f.write_column('E2', shares)
f.write_column('F2', comments)
f.write_column('G2', links)
workbook.close()

dr.close()