#!/usr/bin/python
# -*- coding: iso-8859-1 -*-
# vim: set ft=python ts=3 sw=3 expandtab:
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
#              C E D A R
#          S O L U T I O N S       "Software done right."
#           S O F T W A R E
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
# Copyright (c) 2006-2007,2009 Kenneth J. Pronovici.
# All rights reserved.
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License,
# Version 2, as published by the Free Software Foundation.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
#
# Copies of the GNU General Public License are available from
# the Free Software Foundation website, http://www.gnu.org/.
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #
#
# Authors  : Kenneth J. Pronovici <pronovic@ieee.org>
# Language : Python (>= 2.5)
# Revision : $Id: gmailwhitelist 1369 2014-04-02 13:47:35Z pronovic $
# Purpose  : Generates a whitelist using gmail as a source
#
# # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # # #

########################################################################
# Notes
########################################################################

"""
This program generates an email address whitelist based on gmail data, suitable
for use with the Cedar Solutions whitelist script.

Currently, the gmail whitelist will contain all of your "trusted" gmail
contacts (the ones shown in "My Contacts" in the UI).  

Previously, this script also used to grab information from the sent mail
folder, adding recent sent addresses to the whitelist.  Unfortunately, this
behavior broke when libgmail stopped working, and I haven't found the time to
rewrite it using some other mechanism.

To make this script work, you will need to install Google's Python gdata
library.  On a Debian system, install the python-gdata package.  Otherwise,
download the source package from http://code.google.com/p/gdata-python-client
and follow the instructions to install it (there are no outside dependencies,
so it's very easy).  Most recently, I have tested with version 2.0.18.

@author: Kenneth J. Pronovici <pronovic@ieee.org>
"""

########################################################################
# Imported modules
########################################################################

# System modules
import sys
import os
from datetime import datetime
from stat import S_IMODE
from getpass import getpass
from ConfigParser import ConfigParser
from gdata.contacts.service import ContactsService, ContactsQuery


########################################################################
# Constants and configuration
########################################################################

CONFIG_FILE_PATH = os.path.join(os.environ['HOME'], ".gmailwhitelist")


#######################################################################
# Functions
#######################################################################

def usage():
   """
   Prints out program usage information.
   """
   sys.stderr.write("\n")
   sys.stderr.write("Usage: %s \n" % os.path.basename(sys.argv[0]))
   sys.stderr.write("\n")
   sys.stderr.write("This program generates an email address whitelist (suitable for use\n")
   sys.stderr.write("with the Cedar Solutions whitelist script) based on gmail data.\n")
   sys.stderr.write("\n")
   sys.stderr.write("Currently, the gmail whitelist will contain all of your \"trusted\" gmail\n")
   sys.stderr.write("contacts (the ones shown in \"My Contacts\" in the UI).\n")
   sys.stderr.write("\n")
   sys.stderr.write("Output is always sent to stdout.\n")
   sys.stderr.write("\n")
   sys.stderr.write("Configuration is taken from $HOME/.gmailwhitelist.  Since this file\n")
   sys.stderr.write("may contain sensitive login information, it must have mode 400.\n")
   sys.stderr.write("\n")
   sys.stderr.write("Here is a sample configuration file:\n")
   sys.stderr.write("\n")
   sys.stderr.write("   [GmailLogin]\n")
   sys.stderr.write("   user=user@gmail.com\n")
   sys.stderr.write("   password=somethingsafe\n")
   sys.stderr.write("\n")
   sys.stderr.write("If the login information is not available, you will be prompted for\n")
   sys.stderr.write("it interactively.\n")

def getLoginInfo():
   """
   Gets login information.
      
   If the config file exists, the information is taken from there.  This is a
   simple file INI-style file parsed by Python's ConfigParser class. 

      [GmailLogin]
      user=user@google.com
      password=somethingsafe

   If the config file does not exist, the user is prompted to enter login
   information on stdin.

   If config file does exist, but has a mode other than 400, then a warning
   will be displayed and the file will not be used.
   """
   if os.path.exists(CONFIG_FILE_PATH):
      if S_IMODE(os.stat(CONFIG_FILE_PATH).st_mode) != 0400:
         raise Exception("Config file [%s] exists, but does not have mode 400." % CONFIG_FILE_PATH)
      config = ConfigParser()
      config.read(CONFIG_FILE_PATH)
      try:
         user = config.get("GmailLogin", "user")
         password = config.get("GmailLogin", "password")
      except:
         user = raw_input("Gmail login (xxxx@gmail.com): ")
         password = getpass("Password: ")
   else:
      user = raw_input("Gmail login (xxxx@gmail.com): ")
      password = getpass("Password: ")
   return (user, password)

def getContactAddresses(user, password):
   """
   Gets contact email addresses from gmail.

   This loops through all of the contacts.  For each contact, if the contact
   has an email address, then that email address is added to the list.

   @param user        User to log in as
   @param password    Password to use to log in

   @return Set of unique email addresses among contacts.
   """
   addresses = set()
   client = ContactsService(additional_headers={"GData-Version":"2"})
   client.ssl = True
   client.ClientLogin(user, password)
   groups = client.GetGroupsFeed()
   for group in groups.entry:
      if group.content.text == "System Group: My Contacts":
         query = ContactsQuery()
         query.max_results = 9999   # large enough that we'll get "everything"
         query.group = group.id.text
         contacts = client.GetContactsFeed(query.ToUri())
         for contact in contacts.entry:
            for email in contact.email:
               addresses.add(email.address.lower())
         break
   return addresses

def dumpAddresses(user, addresses):
   """
   Dumps addresses to stdout in a form usable by the whitelist program.
   @param user        User that addresses are for (for reference)
   @param addresses   Addresses to dump
   """
   print "# Whitelist email entries for %s" % user
   print "# Auto-generated by gmailwhitelist at %s" % datetime.today().ctime()
   for address in sorted(addresses):
      print "%s" % address


########################################################################
# Main routine
########################################################################

def main():
   """
   Main routine for program.
   """
   if len(sys.argv) > 1:
      usage()
      sys.exit(1)
   try:
      (user, password) = getLoginInfo()
      addresses = getContactAddresses(user, password)
      dumpAddresses(user, addresses)
   except Exception, e:
      sys.stderr.write("Error generating whitelist: %s" % e)


########################################################################
# Module entry point
########################################################################

# Run the main routine if the module is executed rather than sourced
if __name__ == '__main__':
   main()