Convert Mail.app messages into HTML

parse.py

#!/usr/bin/env python

import re, glob, email.parser

def message_body(message): 
   maintype = message.get_content_maintype()

   if maintype == 'text': 
      return message.get_payload(decode=True)

   if maintype == 'multipart': 
      for part in message.walk(): 
         if part.get_content_maintype() == 'text': 
            return part.get_payload(decode=True)
      raise ValueError('No text part found')

   raise ValueError(maintype)

r_label = re.compile(r'\[[^]]+\][ \t]+')
r_email = re.compile(r'<(.*?)>')

def encode(text): 
   text = text.replace('&', '&amp;')
   return text.replace('<', '&lt;')

def neat(subject): 
   subject = r_label.sub('', subject)
   subject = subject.replace('Re: Re:', 'Re:')
   return encode(subject)

def name_and_email(sender): 
   address = r_email.search(sender).group(1)
   other = r_email.sub('', sender)
   return other.strip('" '), address

print '<link rel="stylesheet" href="http://goo.gl/NVkyD">'
print '<link rel="stylesheet" href="style.css">'

import dateutil.parser

counter = 1
messages = {}
unordered = []

for name in glob.glob('*.emlx'): 
   parser = email.parser.Parser()
   with open(name) as f: 
      next(f)
      bytes = ''.join(line for line in f)
      message = parser.parsestr(bytes)
   date = message['date']
   t = dateutil.parser.parse(date)
   unordered.append((t, message))

for t, message in sorted(unordered): 
   date = message['date']
   sender = message['from']
   person, address = name_and_email(sender)

   subject = message['subject']
   body = message_body(message)

   messageid = message['message-id']
   mid = r_email.search(messageid).group(1)
   messages[mid] = counter

   replied = 0
   reply = message['in-reply-to']
   if reply: 
      replymid = r_email.search(reply).group(1)
      replied = messages.get(replymid, 0)

   if replied: 
      subj = '<a href="#%s">%s</a>' % (replied, neat(subject))
   else: subj = neat(subject)
 
   print '<h2 id="%s">%s</h2>' % (counter, subj)
   print 

   print '<p class="about">(<a href="#%s">#%s</a>) ' % (counter, counter)
   print 'by <strong>%s</strong>' % encode(person)
   address = encode(address).replace('@', '&#8857;')
   print '&lt;%s><br>' % address
   print encode(date)
   print 

   print '<p class="body">'
   parsed = []
   for line in body.splitlines(): 
      if '?xml' in line: break
      if '-- Yahoo!' in line: break
      parsed.append(line)

   body = '\n'.join(parsed)
   body = body.rstrip(' \t\r\n') + '\n'
   for line in body.splitlines(): 
      print encode(line) + '<br>'
   print 

   counter += 1