def _get_attachment_data(self, pos):
"""Return a tuple: file-data, content-type and filename extracted from
the attachment at position `pos`.
"""
# get attachment at position pos
attachment = None
for i, part in enumerate(self.msg.walk()):
if i == pos:
attachment = part
continue
if not attachment:
return None, '', ''
# decode when it's necessary
filename = get_filename(attachment)
if not isinstance(filename, unicode):
filename = filename.decode('utf-8')
# remove line breaks from the filename
filename = re.sub('\s{1,}', ' ', filename)
content_type = attachment.get_content_type()
if content_type == 'message/rfc822':
nested_messages = attachment.get_payload()
assert len(nested_messages) == 1, (
'we expect that attachments with messages only contain one '
'message per attachment.')
data = nested_messages[0].as_string()
else:
data = attachment.get_payload(decode=1)
return data, content_type, filename
def _get_attachment_data(self, pos):
"""Return a tuple: file-data, content-type and filename extracted from
the attachment at position `pos`.
"""
# get attachment at position pos
attachment = None
for i, part in enumerate(self.msg.walk()):
if i == pos:
attachment = part
continue
if not attachment:
return None, '', ''
# decode when it's necessary
filename = get_filename(attachment)
if not isinstance(filename, unicode):
filename = filename.decode('utf-8')
# remove line breaks from the filename
filename = re.sub('\s{1,}', ' ', filename)
content_type = attachment.get_content_type()
if content_type == 'message/rfc822':
nested_messages = attachment.get_payload()
assert len(nested_messages) == 1, (
'we expect that attachments with messages only contain one '
'message per attachment.')
data = nested_messages[0].as_string()
else:
data = attachment.get_payload(decode=1)
return data, content_type, filename
def get_attachment_as_namedfile(self, pos):
"""Return a namedfile extracted from the attachment in
position `pos`.
"""
# get attachment at position pos
attachment = None
for i, part in enumerate(self.context.msg.walk()):
if i == pos:
attachment = part
continue
if not attachment:
return None
# decode when it's necessary
filename = get_filename(attachment)
if not isinstance(filename, unicode):
filename = filename.decode('utf-8')
# remove line breaks from the filename
filename = re.sub('\s{1,}', ' ', filename)
return NamedFile(data=attachment.get_payload(decode=1),
contentType=attachment.get_content_type(),
filename=filename)
def get_attachment_as_namedfile(self, pos):
"""Return a namedfile extracted from the attachment in
position `pos`.
"""
# get attachment at position pos
attachment = None
for i, part in enumerate(self.context.msg.walk()):
if i == pos:
attachment = part
continue
if not attachment:
return None
# decode when it's necessary
filename = get_filename(attachment)
if not isinstance(filename, unicode):
filename = filename.decode('utf-8')
# remove line breaks from the filename
filename = re.sub('\s{1,}', ' ', filename)
return NamedFile(data=attachment.get_payload(decode=1),
contentType=attachment.get_content_type(),
filename=filename)
def render(self):
""" Copied from ftw.mail.attachment:
Just add the contenttype-check (see l.51)
"""
if self.message is None:
raise NotFound
message = email.message_from_string(self.message.data)
# we need an int value for the position
pos = 0
try:
pos = int(self.position)
except ValueError:
raise NotFound
# get attachment at position pos
attachment = None
for i, part in enumerate(message.walk()):
if i == pos:
attachment = part
continue
if attachment is not None:
content_type = attachment.get_content_type()
filename = utils.get_filename(attachment)
if filename is None:
raise NotFound
# the the konsul migrated mails, often has a wrong
# content-type for word documents, therefore we check the
# content-type, in this case we guess it over the filename
if content_type == 'application/octet-stream':
mtr = getToolByName(self.context, 'mimetypes_registry')
content_type = mtr.globFilename(filename)
# make sure we have a unicode string
if not isinstance(filename, unicode):
filename = filename.decode('utf-8', 'ignore')
# internet explorer and safari don't like rfc encoding of filenames
# and they don't like utf encoding too.
# therefore we first try to encode the filename in iso-8859-1
try:
filename = filename.encode('iso-8859-1')
except ConflictError:
raise
except Exception:
filename = filename.encode('utf-8', 'ignore')
self.request.response.setHeader('Content-Type', content_type)
set_attachment_content_disposition(self.request, filename)
if content_type == 'message/rfc822':
return attachment.as_string()
return attachment.get_payload(decode=1)
raise NotFound
def test_get_filename(self):
msg_txt = \
"""Content-Type: application/octet-stream;
name="=?iso-8859-1?Q?Aperovorschl=E4ge_2010=2Epdf?="
Content-Transfer-Encoding: base64
Content-Description: =?iso-8859-1?Q?Aperovorschl=E4ge_2010=2Epdf?=
Content-Disposition: attachment;
filename="=?iso-8859-1?Q?Aperovorschl=E4ge_2010=2Epdf?="
"""
msg = email.message_from_string(msg_txt)
# !!! seems to be a bug in email package
self.assertEquals('Aperovorschläge 2010.pdf', utils.get_filename(msg))
msg_txt = \
"""Content-Disposition: attachment;
filename*=iso-8859-1''f%F6rmularz%FCgriffsber%E4chtigungen.doc
Content-Type: application/msword;
name*=iso-8859-1''f%F6rmularz%FCgriffsber%E4chtigungen.doc
Content-Transfer-Encoding: base64
"""
msg = email.message_from_string(msg_txt)
self.assertEquals('f\xc3\xb6rmularz\xc3\xbcgriffsber\xc3\xa4chtigungen.doc', utils.get_filename(msg))
def render(self):
if self.message is None:
raise NotFound
message = email.message_from_string(self.message.data)
# we need an int value for the position
pos = 0
try:
pos = int(self.position)
except ValueError:
raise NotFound
# get attachment at position pos
attachment = None
for i, part in enumerate(message.walk()):
if i == pos:
attachment = part
continue
if attachment is not None:
content_type = attachment.get_content_type()
filename = utils.get_filename(attachment)
if filename is None:
raise NotFound
# make sure we have a unicode string
if not isinstance(filename, unicode):
filename = filename.decode('utf-8', 'ignore')
# internet explorer and safari don't like rfc encoding of filenames
# and they don't like utf encoding too.
# therefore we first try to encode the filename in iso-8859-1
try:
filename = filename.encode('iso-8859-1')
except:
filename = filename.encode('utf-8', 'ignore')
self.request.response.setHeader('Content-Type', content_type)
self.request.response.setHeader('Content-Disposition', 'inline; filename=%s' % filename)
if content_type == 'message/rfc822':
return attachment.as_string()
return attachment.get_payload(decode=1)
raise NotFound
