def get_raw_body_text(self, msg: email.message.Message) -> typing.List[typing.Tuple[typing.Any, typing.Any, typing.Any]]:
"""This method recursively retrieves all e-mail body parts and returns them as a list.
Args:
msg (email.message.Message): The actual e-mail message or sub-message.
Returns:
list: Returns a list of sets which are in the form of "set(encoding, raw_body_string, message field headers)"
"""
raw_body: typing.List[typing.Tuple[typing.Any, typing.Any, typing.Any]] = []
if msg.is_multipart():
for part in msg.get_payload():
raw_body.extend(self.get_raw_body_text(part))
else:
# Treat text document attachments as belonging to the body of the mail.
# Attachments with a file-extension of .htm/.html are implicitly treated
# as text as well in order not to escape later checks (e.g. URL scan).
try:
filename = msg.get_filename('').lower()
except (binascii.Error, AssertionError):
logger.exception(
'Exception occurred while trying to parse the content-disposition header. Collected data will not be complete.')
filename = ''
# pylint: disable=too-many-boolean-expressions
if ('content-disposition' not in msg and msg.get_content_maintype() == 'text') \
or (filename.endswith('.html') or filename.endswith('.htm')) \
or ('content-disposition' in msg and msg.get_content_disposition() == 'inline'
and msg.get_content_maintype() == 'text'):
encoding = msg.get('content-transfer-encoding', '').lower()
charset = msg.get_content_charset()
if charset is None:
raw_body_str = msg.get_payload(decode=True)
raw_body_str = eml_parser.decode.decode_string(raw_body_str, None)
else:
try:
raw_body_str = msg.get_payload(decode=True).decode(charset, 'ignore')
except (LookupError, ValueError):
logger.debug('An exception occurred while decoding the payload!', exc_info=True)
raw_body_str = msg.get_payload(decode=True).decode('ascii', 'ignore')
# In case we hit bug 27257, try to downgrade the used policy
try:
raw_body.append((encoding, raw_body_str, msg.items()))
except AttributeError:
former_policy: email.policy.Policy = msg.policy # type: ignore
msg.policy = email.policy.compat32 # type: ignore
raw_body.append((encoding, raw_body_str, msg.items()))
msg.policy = former_policy # type: ignore
return raw_body
def prepare_multipart_part_attachment(
msg: email.message.Message,
counter: int = 0,
include_attachment_data: bool = False) -> typing.Dict[str, typing.Any]:
"""Extract meta-information from a multipart-part.
Args:
msg (email.message.Message): An e-mail message object.
counter (int, optional): A counter which is used for generating attachments
file-names in case there are none found in the header. Default = 0.
include_attachment_data (bool, optional): If true, method includes the raw attachment data when
returning. Default = False.
Returns:
dict: Returns a dict with original multi-part headers as well as generated hash check-sums,
date size, file extension, real mime-type.
"""
attachment = {}
# In case we hit bug 27257, try to downgrade the used policy
try:
lower_keys = dict((k.lower(), v) for k, v in msg.items())
except AttributeError:
former_policy = msg.policy
msg.policy = email.policy.compat32
lower_keys = dict((k.lower(), v) for k, v in msg.items())
msg.policy = former_policy
if ('content-disposition' in lower_keys and msg.get_content_disposition() != 'inline') \
or msg.get_content_maintype() != 'text':
# if it's an attachment-type, pull out the filename
# and calculate the size in bytes
if msg.get_content_type() == 'message/rfc822':
payload = msg.get_payload()
if len(payload) > 1:
logger.warning(
'More than one payload for "message/rfc822" part detected. This is not supported, please report!'
)
data = bytes(payload[0])
file_size = len(data)
else:
data = msg.get_payload(
decode=True) # type: bytes # type is always bytes here
file_size = len(data)
filename = msg.get_filename('')
if filename == '':
filename = 'part-{0:03d}'.format(counter)
else:
filename = eml_parser.decode.decode_field(filename)
file_id = str(uuid.uuid1())
attachment[file_id] = {}
attachment[file_id]['filename'] = filename
attachment[file_id]['size'] = file_size
# os.path always returns the extension as second element
# in case there is no extension it returns an empty string
extension = os.path.splitext(filename)[1].lower()
if extension:
# strip leading dot
attachment[file_id]['extension'] = extension[1:]
attachment[file_id]['hash'] = get_file_hash(data)
if not (magic_mime is None or magic_none is None):
mime_type = magic_none.buffer(data)
mime_type_short = magic_mime.buffer(data)
if not (mime_type is None or mime_type_short is None):
attachment[file_id]['mime_type'] = mime_type
# attachments[file_id]['mime_type_short'] = attachments[file_id]['mime_type'].split(",")[0]
attachment[file_id]['mime_type_short'] = mime_type_short
else:
logger.warning(
'Error determining attachment mime-type - "{}"'.format(
file_id))
if include_attachment_data:
attachment[file_id]['raw'] = base64.b64encode(data)
ch = {} # type: typing.Dict[str, typing.List[str]]
for k, v in msg.items():
k = k.lower()
v = str(v)
if k in ch:
ch[k].append(v)
else:
ch[k] = [v]
attachment[file_id]['content_header'] = ch
counter += 1
return attachment
