Get facebook ids for a list of email addresses.

getfacebookidsfromemail.py

#!/usr/bin/env python
'''A simple script for finding facebook profile ids for a given list of email
addresses.

Requires eventlet to be installed.
    $ pip install eventlet

Also, currently requires a list of contacts exported from Google contacts.
Choose the outlook format and name the file 'contacts.csv' in this dir.'''

from urllib import urlencode, unquote_plus
from urlparse import urlparse, parse_qs
from eventlet.green import urllib2
from os.path import isfile
import csv
import eventlet
import json

# APP_ID and APP_URL are needed if you are requesting a new auth token.
APP_URL = ''
APP_ID = ''

# Get token from .token on disk. 
TOKEN = ''
if not TOKEN and isfile('.token'):
    with open('.token') as f:
        TOKEN = f.read()

def get_new_oauth_token():
    '''Will generate a new oauth token and save to ".token".'''
    params = {
        'type': 'user_agent',
        'client_id': APP_ID,
        'redirect_uri': APP_URL,
        'scope': 'user_photos,email,user_birthday,user_online_presence',
    }
    base_url = 'https://graph.facebook.com/oauth/authorize'
    url = '?'.join([base_url, urlencode(params)])
    # Open a browser so the user can authorize the application.
    print 'Please visit %s' % url
    print 'After authorizing the app, you will be redirected to a new url.'
    print 'Please paste that new url below.'
    url = raw_input('> ')
    _, fragment = url.split('#')
    args = parse_qs(fragment)
    access_token = args['access_token'][0]

    with open('.token', 'w') as f:
        f.write(access_token)
    raw_input('Token has been saved to disk. Continue? >')
    return access_token

def emails_from_google_csv(fn):
    '''Export your google contacts as an outlook CSV file. This function will
    return a list of email addresses. Currently only parses the first email
    field.'''
    with open(fn, 'rb') as f:
        contacts = csv.DictReader(f)
        emails = [c['E-mail Address'] for c in contacts if c['E-mail Address']]
    return emails

def fetch(url):
    '''Downloads a response for a given url.'''
    print "Opening", url
    u = urllib2.urlopen(url)
    body = u.read()
    u.close()
    print "Closing", url
    return body

def get_facebook_uuids(token, emails, poolsize=5):
    '''Returns a list of uuids (or None) for a list of emails. Also requires an
    oauth token.'''
    urls = ['https://graph.facebook.com/search?q=%s&type=user&access_token=%s'
        % (email, TOKEN) for email in emails]

    pool = eventlet.GreenPool(poolsize)
    responses = pool.imap(fetch, urls)

    uuids = []
    for resp in responses:
        obj = json.loads(resp)
        try:
            uuids.append(obj['data'][0]['id'])
        except IndexError: 
            # No facebook profile found
            uuids.append(None)

    return uuids

if __name__ == '__main__':
    # If you need a new oauth token, uncomment this line.
    #TOKEN = get_new_oauth_token()

    emails = emails_from_google_csv('contacts.csv')
    uuids = get_facebook_uuids(TOKEN, emails)
    
    print zip(emails, uuids)
    print 'Done.'