def nicehtml(html):
# ensure that the first <li is wrapped in a <ul
oul = re.search('(?i)<[ou]l', html)
li = re.search('(?i)<li', html)
if li:
if not oul or (li.start() < oul.start()):
# add <ul>
st = li.start()
html = html[:st] + '</p><ul>' + html[st:]
bs = BeautifulSoup.ICantBelieveItsBeautifulSoup(html)
return bs.prettify().decode('utf-8')
def asEmail(self):
if self._email:
return self._email
self.prepareMessage()
hostname = socket.getfqdn() # used below in a few places
mpp = self.mp.copy()
def wrapOrZap(key, title):
val = self.mp.get(key, None)
if val:
mpp[key] = "<p class='%s' style='margin: 0;'>%s: %s</p>\n"
mpp[key] %= (key, title, val)
else:
mpp[key] = ''
wrapOrZap('sender', 'Fra')
wrapOrZap('recipient', 'Til')
wrapOrZap('cc', 'Kopi til')
if mpp.get('time', None):
mpp['ttime'] = u' ' + mpp['time']
else:
mpp['ttime'] = u''
# create initial HTML version
html = u'''<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
<title>%(title)s</title>
</head>
<body style='font-family: Verdana,Arial,Helvetica'>
<h1>%(title)s</h1>
<div class='meta' style='background-color: #eaeaea; color: #000; padding: 5px; margin: 0 0 10px 0;'>
%(sender)s%(recipient)s%(cc)s <p class='date' style='margin: 0;'>Dato: %(date)s%(ttime)s</p>
</div>
<div class='text'>
%(nicehtml)s
</div>
</body>
</html>
'''
html %= mpp
html = BeautifulSoup.ICantBelieveItsBeautifulSoup(html)
# first look for inline images (if any)
# iimags: mapping from URL to (cid, binary string contents)
iimgs = {}
for imgtag in html.findAll('img'):
if not imgtag.has_key('src'):
continue # ignore
url = imgtag['src']
if url.lower().startswith('data:'):
# ignore 'inline' images
continue
elif not url:
# ignore empty URLs
continue
if url not in iimgs:
try:
data = surllib.skoleGetURL(url, False)
except urllib2.URLError:
# could not fetch URL for some reason - ignore
continue
# is this actually an image?
if not imghdr.what('', data):
continue # ignore
cid = 'image%d-%f@%s' % (len(iimgs) + 1, time.time(), hostname)
iimgs[url] = (cid, data)
cid, _ = iimgs[url]
imgtag['src'] = 'cid:' + cid
# next, handle attachments
# attachments: email attachments ready for attachment :)
attachments = []
for atag in html.findAll('a'):
try:
url = atag['href']
except KeyError:
atag.replaceWithChildren() # kill the "broken" link
continue
url = atag['href']
if 'Tilmelding/Oversigt.asp' in url:
atag.replaceWithChildren() # kill link
continue
if url.startswith('/') or config.HOSTNAME in url: # onsite!
data = None
try:
data = surllib.skoleGetURL(url, False)
except:
# unable to fetch URL
config.log(
u'%s: Kan ikke hente flg. URL: %s' %
(self.mp['title'] if self.mp['title'] else self, url))
if data:
if atag.has_key('usefilename'):
usefilename = atag['usefilename']
else:
usefilename = None
eatt = generateMIMEAttachment(url, data, usefilename)
attachments.append(eatt)
atag.replaceWithChildren() # kill the actual link
# now, put the pieces together
html = html.prettify().decode('utf-8')
msgHtml = MIMEText(html, 'html', 'utf-8')
if not iimgs and not attachments:
# pure HTML version
msg = msgHtml
else:
# inline images but no attachments
# multipart/related
# text/html with html text
# image/xxx with inline images
# OR
# email with inline images + attachment
# multipart/mixed
# text/html med html udgave
# image/gif med billede
# application/xxx with word document
if attachments:
msg = MIMEMultipart('mixed', type='text/html')
else:
msg = MIMEMultipart('related', type='text/html')
del msgHtml['MIME-Version']
msg.attach(msgHtml)
# attach images if any
for (url, (cid, data)) in iimgs.items():
m = MIMEImage(data)
m.add_header('Content-ID', '<%s>' % cid)
fn = os.path.basename(url).encode('utf-8')
m.add_header('Content-Disposition',
'inline',
filename=('utf-8', '', fn))
del m['MIME-Version']
msg.attach(m)
# attach attachments if any
for attachment in attachments:
del attachment['MIME-Version']
msg.attach(attachment)
# now for the general headers
dt = self.mp['date']
if self.mp.get('time', None):
dt += ' ' + self.mp['time']
else:
if dt == time.strftime('%d-%m-%Y'): # today
ts = time.strftime('%H:%M:%S')
if ts > '12:00:00':
ts = '12:00:00'
dt += ' ' + ts
else:
dt += ' 12:00:00'
dt = time.strptime(dt, '%d-%m-%Y %H:%M:%S')
dt = email.utils.formatdate(time.mktime(dt), True)
msg['Received'] = ('from %s ([127.0.0.1] helo=%s) '
'by %s with smtp (fskintra) for %s; %s') % (
hostname, hostname, hostname, config.EMAIL, dt)
msg['Date'] = dt
title = self.mp['title']
if self.mp['childname']:
title = u'[%s] %s' % (self.mp['childname'], title)
msg['Subject'] = Header(title, 'utf-8', 60)
if 'sender' in self.mp and self.mp['sender']:
sender = u'Skoleintra - %s' % self.mp['sender']
else:
sender = u'Skoleintra'
sender = headerEncodeField(sender) + u' <%s>' % config.SENDER
msg['From'] = sender
msg['To'] = config.EMAIL
# other tags just for ourselves
keys = 'mid,md5'.split(',')
for key in keys:
if self.mp.get(key, None):
kkey = 'X-skoleintra-%s' % key
msg[kkey] = Header(self.mp[key], 'utf-8', header_name=kkey)
self._email = msg
return msg
def asEmail(self):
if self._email:
return self._email
self.prepareMessage()
hostname = socket.getfqdn() # used below in a few places
mpp = self.mp.copy()
def wrapOrZap(key, title):
val = self.mp.get(key, None)
if val:
mpp[key] = "<p class='%s' style='margin: 0;'>%s: %s</p>\n"
mpp[key] %= (key, title, val)
else:
mpp[key] = ''
wrapOrZap('sender', 'Fra')
wrapOrZap('recipient', 'Til')
if mpp.get('time', None):
mpp['ttime'] = u' ' + mpp['time']
else:
mpp['ttime'] = u''
# create initial HTML version
html = u'''<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
<title>%(title)s</title>
</head>
<body style='font-family: Verdana,Arial,Helvetica'>
<h1>%(title)s</h1>
<div class='meta' style='background-color: #eaeaea; color: #000; padding: 5px; margin: 0 0 10px 0;'>
%(sender)s%(recipient)s <p class='date' style='margin: 0;'>Dato: %(date)s%(ttime)s</p>
</div>
<div class='text'>
%(nicehtml)s
</div>
</body>
</html>
'''
html %= mpp
html = BeautifulSoup.ICantBelieveItsBeautifulSoup(html)
# first look for inline images (if any)
# iimags: mapping from URL to (cid, binary string contents)
iimgs = {}
for imgtag in html.findAll('img'):
url = imgtag['src']
if url.lower().startswith('data:'):
# ignore 'inline' images
continue
elif not url:
# ignore empty URLs
continue
if url not in iimgs:
try:
data = surllib.skoleGetURL(url, False)
except urllib2.URLError, e:
# could not fetch URL for some reason - ignore
continue
# is this actually an image?
if not imghdr.what('', data):
continue # ignore
cid = 'image%d-%f@%s' % (len(iimgs) + 1, time.time(), hostname)
iimgs[url] = (cid, data)
cid, _ = iimgs[url]
imgtag['src'] = 'cid:' + cid