Question

I’m trying to use Mechanize login to Google Docs so that I can scrape something (not possible from the API) but I keep seem to keep getting a 404 when trying to follow the meta redirect:

require 'rubygems'
require 'mechanize'

USERNAME = "..."
PASSWORD = "..."

LOGIN_URL = "https://www.google.com/accounts/Login?hl=en&continue=http://docs.google.com/"

agent = Mechanize.new
login_page = agent.get(LOGIN_URL)
login_form = login_page.forms.first
login_form.Email = USERNAME
login_form.Passwd = PASSWORD
login_response_page = agent.submit(login_form)

redirect = login_response_page.meta[0].uri.to_s

puts "redirect: #{redirect}"

followed_page = agent.get(redirect) # throws a HTTPNotFound exception

pp followed_page

Can anyone see why this isn’t working?

Answer 1

Andy you’re awesome!!
Your code helped me to make my script workable and to login into google account. I found your error after couple of hours.It was about html escaping. As I found,Mechanize automatically escapes uri it recieves as a parameter for ‘get’ method. So my solution is:

EMAIL  = ".."
PASSWD = ".."
agent = Mechanize.new{ |a| a.log = Logger.new("mech.log")}
agent.user_agent_alias = 'Linux Mozilla'
agent.open_timeout = 3
agent.read_timeout = 4
agent.keep_alive   = true
agent.redirect_ok  = true
LOGIN_URL = "https://www.google.com/accounts/Login?hl=en"

login_page = agent.get(LOGIN_URL)
login_form = login_page.forms.first
login_form.Email = EMAIL
login_form.Passwd = PASSWD
login_response_page = agent.submit(login_form)

redirect = login_response_page.meta[0].uri.to_s

puts redirect.split('&')[0..-2].join('&') + "&continue=https://www.google.com/"
followed_page = agent.get(redirect.split('&')[0..-2].join('&') + "&continue=https://www.google.com/adplanner")
pp followed_page

This works just fine for me. I have replaced continue parameter from the meta tag (which is already escaped) by new one.