ansaurus

Question

Answer 1

+3 A:

What you are trying to do is not possible. You are going to have to use a browser to get an access token one way or another. You cannot collect username and passwords (a big violation of Facebook's TOS). If you need a script that runs without user interaction you will still need to use a browser to authenticate, but once you have the user's token you can use it without their direct interaction. You must request the "offline_access" permission to gain an access token that does not expire. You can save this token and then use it for however long you need.

Nathan Totten 2010-10-22 21:05:27

Thanks for the information. I'm not collecting username/passwords to be clear though. It's our companies username/password. Just putting together a script that'll run nightly via cron or something. Looks like I'll be going a different route. Thanks again :)

f4nt 2010-10-22 21:10:59

So if that is the case, just do the authentication one time with offline_access and just save the access token and use it in your script. The token will never expire unless the account removes the application.

Nathan Totten 2010-10-22 21:18:02

@f4nt I think your going to need a browser. That said, you can automate a browser without being present. Check out Watir (ruby lib) for an excellent browser automation. Watir has versions for n windows, mac, and linux and works with safari, chrome, IE, and firefox. I do everything in python except browser automation, which I do with ruby/Watir.

Steven Rumbalski 2010-10-22 21:23:39

@f4nt Nevermind. Looks like Nathan really knows how to do it.

Steven Rumbalski 2010-10-22 21:24:20

what is the browser doing that can't be done with something like mechanize?

Corey Goldberg 2010-10-22 22:27:18

You could do it with the browser, but that isn't really the right way to do it. For him, he is just doing this with his own username and password, but if it were other people's u/p that would be a violation of Facebook's TOS.

Nathan Totten 2010-10-23 23:06:50

@Corey You get an access token via their javascript OAuth nonsense I believe. The purpose is to grab that access token, and then leverage their API. I can get in with httplib2 and stuff, but that access token never becomes obtainable for me. Then again, I haven't fought it *that* hard yet :)

f4nt 2010-10-25 20:31:48

Answer 2

A:

I've done this with my own account before using mechanize. You can log in to Facebook using something like the below and then just follow the links to where you want to go. This will print out the contents of you news feed.

#!/usr/bin/env python

import mechanize

browser = mechanize.Browser()
browser.set_handle_robots(false)
cookies = mechanize.CookieJar()
browser.set_cookiejar()
browser.addheaders = [('User-agent', 'Mozilla/5.0 (X11; U; Linux i686; en-US) AppleWebKit/534.7 (KHTML, like Gecko) Chrome/7.0.517.41 Safari/534.7')]

browser.select_form(nr=0)
browser.form['email'] = 'YOUR_EMAIL_ADDRESS'
browser.form['pass'] = 'YOUR_PASSWORD'
response = browser.submit()
print response.read()

EverythingZen 2010-10-23 13:23:28

ansaurus

tags:

views:

answers:

Logging into Facebook without a Browser

related questions