CATE results alerter

cate.py

#!/usr/bin/python

from urllib2 import HTTPPasswordMgrWithDefaultRealm, HTTPBasicAuthHandler, build_opener, install_opener, urlopen
from base64 import b64decode
from BeautifulSoup import BeautifulSoup
import re
import smtplib
import pickle

username = 'USERNAME'
password = 'PASSWORD' # base64 encode password (allows editing with giving away password)
theurl = 'https://cate.doc.ic.ac.uk/student.cgi?key=2011:c4:' + username

def email(message):
  try:
     smtpObj = smtplib.SMTP('smtp.cc.ic.ac.uk')
     smtpObj.ehlo()
     smtpObj.starttls()
     smtpObj.login(username, b64decode(password))
     smtpObj.sendmail(username + '@ic.ac.uk', username + '@ic.ac.uk', message)     
     smtpObj.close()    
  except:
     pass

def compare_results(current_results, previous_results):
    for course, coursework_set in current_results.items():
      if previous_results[course]:
        prev_coursework_set = previous_results[course]
        diff = [ k for k in coursework_set if coursework_set[k] != prev_coursework_set[k] ] #lc
        
        if diff:
          message = ''
          for k in diff:
            message += course + '\n'
            message += k + ' changed from ' + prev_coursework_set[k] + ' to ' + coursework_set[k] + '\n'
            
          email(message)

def main():
  passman = HTTPPasswordMgrWithDefaultRealm()
  # because we have put None at the start it will always
  # use this username/password combination for  urls
  # for which `theurl` is a super-url
  passman.add_password(None, theurl, username, b64decode(password))

  # create the AuthHandler
  authhandler = HTTPBasicAuthHandler(passman)

  opener = build_opener(authhandler)
  install_opener(opener)

  # authentication is now handled automatically for us
  pagehandle = urlopen(theurl)

  htmlparser = BeautifulSoup(pagehandle.read())

  rows = htmlparser.findAll('tr')

  results = []
  courses = set([])
  coursework_results = dict()

  for row in rows:
    results.extend(row.findAll('td', align='left', colspan='5'))

  for r in results:
    try:
      course = r
      coursework_row = r.findNext('tr')
      courses.add(coursework_row)
      if not 'No exercise activities recorded for this module' in coursework_row.text:
          coursename = r.text[3:]
          if not coursename in courses:
            courses.add(coursename)
            coursework_results[coursename] = dict()
            
            while coursework_row.text and not 'No student interaction for this exercise' in coursework_row.text:
              courseworkname = coursework_row.findAll(bgcolor='white')[2].text.replace(' ', '')
              grade = coursework_row.findAll(align='center')[-1].text
              
              coursework_results[coursename][courseworkname] = grade
              
              coursework_row = coursework_row.findNext('tr')
    except:
      pass
      
  try: 
    try:
      prev_coursework_results = pickle.load(open('results.pkl', 'rb'))
      compare_results(coursework_results, prev_coursework_results)
    except:
      pass

    pickle.dump(coursework_results, open('results.pkl', 'w+b'))
  except IOError as e:
    pass
 
# script entry point
if __name__ == '__main__':
	main()