Facebook data without Opengraph

[Gabriela Campedelli]

Hello:

As you may know, Facebook has Opengraph API with JSON output and token access, but some useful data are not available using opengraph and nor it is in Json. 

I have search data that I would like to get using Scraperwiki, now Quickcode. 

The snippet I´m using is bellow but it only gets the page network, which means others pages that the page I´m scraping liked. 

But I want to get users IDs that liked the page and that is deprecated in Opengraph. So this information is only available in search.

The url is something like this: https://www.facebook.com/search/199567963416843/likers?ref=about

The major problem is that account has to be logged in Facebook to access this page as a regular user. I can,, of course, scrape the code after login using a js script in Chrome console, but it would be a great idea to do it automatically using Scraperwiki. Does anyone have an idea on how to do that? I promise to share the js I´m using in Chrome console. 

Here is the snippet for Opengraph:

#!/usr/bin/env python

import scraperwiki

import json

html = scraperwiki.scrape("https://graph.facebook.com/v2.7/{page-id}/likes?access_token={value - you can replace with your access token here}")

print  html

# Saving data:

for i in html:

    html={'data'}

    data=['i']

    print i

#scraperwiki.sql.save(unique_keys, data)

Thank you so much for your help

Gabriela

[Gabriela Campedelli]

I found a code that is able to login but I have to understand more about cookies because it is giving me an error. I´ve commented the line, error has gone away, I´ve logged and Facebook notified me that I have a login from Dublin after blocking my account for security reasons. So, if anyone can help me to improve this code, I would be very happy for that. Is it better to use Mechanize? How do you feel about it? 

#!/usr/bin/env python

import requests

USERNAME = 'your email'

PASSWORD = 'your password'

PROTECTED_URL = 'https://www.facebook.com/search/199567963416843/likers?ref=about'

def login(session, email, password):

  

    # Attempt to login to Facebook

    response = session.post('https://m.facebook.com/login.php', data={

        'email': email,

        'pass': password

    }, allow_redirects=False)

    assert response.status_code == 302

    assert 'c_user' in response.cookies

    return response.cookies

if __name__ == "__main__":

    session = requests.session()

    #cookies = login(session, USERNAME, PASSWORD)

    response = session.get(PROTECTED_URL, 

allow_redirects=False)

    #assert response.text.find('Home') != -1