#!/usr/bin/env python

import urllib2
from urllib2 import urlopen
from urllib import urlencode
from cookielib import CookieJar
from BeautifulSoup import BeautifulSoup
import re
import os

# Uncomment this line if you use Google Apps for your domain, and insert your domain
#loginl = 'https://mail.google.com/a/example.com/'
# Uncomment this line if you use normal Gmail
#loginl = 'https://mail.google.com/mail/'

# Enter your username and password
username = 'username'
password = 'password'

clean = re.compile('^ *\n')

# Setup cookie handling
cookies = CookieJar()
opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(cookies))
urllib2.install_opener(opener)

# Load the login page
print 'Loading login page'
loginf = urlopen(loginl)
login = BeautifulSoup(loginf.read())
loginf.close()

# Fill in the POST data
data = {}
for input in login.form.findAll('input'):
	if input.has_key('value'):
		data[input['name']] = input['value']

data['Email'] = username
data['Passwd'] = password

# Submit the login information and get the "loading" page
print 'Logging in'
loadf = urlopen(login.form['action'], urlencode(data))
load = BeautifulSoup(loadf.read())
loadf.close()

# Follow the redirect (only for normal Gmail)
redirect = load.findAll('meta', attrs={'http-equiv': 'refresh'})
if len(redirect) > 0:
	url = redirect[0]['content'].split('=',1)[1].strip("'")

	loadf = urlopen(url)
	load = BeautifulSoup(loadf.read())
	loadf.close()

# Load the HTML view
print 'Loading main page'
mainf = urlopen(loadf.url + '?ui=html&zy=c')
main = BeautifulSoup(mainf.read())
mainf.close()

# Search for chats
nextl = main.base['href'] + '?s=q&q=label%3Achat&nvp_site_mail=Search%2DMail'

# Loop through all pages of the search
while nextl != None:
	# Load the search result
	print 'Loading search results: ' + nextl
	searchf = urlopen(nextl)
	search = BeautifulSoup(searchf.read())
	searchf.close()

	base = search.html.base['href']

	# Process each entry
	for chat in search.findAll('a', href=re.compile('^\?v=c&.*')):
		
		id = dict([part.split('=') for part in chat['href'].split('&')])['th']

		# Check if we've already saved this chat
		if not os.path.isfile(id + '.eml'):

			print 'Saving chat ' + id
			
			origf = urlopen(base + '?v=om&th=' + id)
			orig = origf.read()
			origf.close()

			# Make sure that it's a valid chat
			if orig.find('Message-ID') < 1:
				print id + ' is not a valid chat'
			else:
				# Clean up and save
				orig = orig.replace('\r', '')
				orig = clean.split(orig, 1)[1]
				save = open(id + '.eml', 'w')
				save.write(orig)
				save.close()

	# Find the link to the next results page
	older = search.findAll('b', text=re.compile('Older'))
	if len(older) > 0:
		nextl = base + older[0].parent.parent['href']
	else:
		nextl = None

print 'Done'