def get_urls(url, depth=1):
if depth > 5:
print('Too many redirects.')
return
fex = Faup()
def meta_redirect(content):
c = content.lower()
soup = BeautifulSoup(c)
for result in soup.find_all(attrs={'http-equiv': 'refresh'}):
if result:
out = result["content"].split(";")
if len(out) == 2:
wait, text = out
a, url = text.split('=', 1)
return url.strip()
return None
resolve, reason = try_resolve(fex, url)
if not resolve:
# FIXME: inform that the domain does not resolve
yield url
return
logging.debug("Making HTTP connection to " + url)
headers = {'User-agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0) Gecko/20100101 Firefox/8.0'}
try:
response = requests.get(url, allow_redirects=True, headers=headers,
timeout=15, verify=False)
except:
# That one can fail (DNS for example)
# FIXME: inform that the get failed
yield url
return
if response.history is not None:
for h in response.history:
# Yeld the urls in the order we find them
yield h.url
yield response.url
meta_redir_url = meta_redirect(response.content)
if meta_redir_url is not None:
depth += 1
if not meta_redir_url.startswith('http'):
fex.decode(url)
base = '{}://{}'.format(fex.get_scheme(), fex.get_host())
port = fex.get_port()
if port is not None:
base += ':{}'.format(port)
if not meta_redir_url.startswith('/'):
# relative redirect. resource_path has the initial '/'
if fex.get_resource_path() is not None:
base += fex.get_resource_path()
if not base.endswith('/'):
base += '/'
meta_redir_url = base + meta_redir_url
for url in get_urls(meta_redir_url, depth):
yield url
def get_urls(url, depth=1):
if depth > 5:
print('Too many redirects.')
return
fex = Faup()
def meta_redirect(content):
c = content.lower()
soup = BeautifulSoup(c, "html.parser")
for result in soup.find_all(attrs={'http-equiv': 'refresh'}):
if result:
out = result["content"].split(";")
if len(out) == 2:
wait, text = out
a, url = text.split('=', 1)
return url.strip()
return None
resolve, reason = try_resolve(fex, url)
if not resolve:
# FIXME: inform that the domain does not resolve
yield url
return
logging.debug("Making HTTP connection to " + url)
headers = {'User-agent': 'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0) Gecko/20100101 Firefox/8.0'}
try:
response = requests.get(url, allow_redirects=True, headers=headers,
timeout=15, verify=False)
except:
# That one can fail (DNS for example)
# FIXME: inform that the get failed
yield url
return
if response.history is not None:
for h in response.history:
# Yeld the urls in the order we find them
yield h.url
yield response.url
meta_redir_url = meta_redirect(response.content)
if meta_redir_url is not None:
depth += 1
if not meta_redir_url.startswith('http'):
fex.decode(url)
base = '{}://{}'.format(fex.get_scheme(), fex.get_host())
port = fex.get_port()
if port is not None:
base += ':{}'.format(port)
if not meta_redir_url.startswith('/'):
# relative redirect. resource_path has the initial '/'
if fex.get_resource_path() is not None:
base += fex.get_resource_path()
if not base.endswith('/'):
base += '/'
meta_redir_url = base + meta_redir_url
for url in get_urls(meta_redir_url, depth):
yield url
def __post_init__(self):
f = Faup(
) # Example code at https://programtalk.com/python-examples-amp/pyfaup.faup.Faup/
f.decode(self.url)
self.scheme = f.get_scheme()
self.top_level_domain = f.get_tld()
self.domain = f.get_domain()
self.subdomain = f.get_subdomain()
self.path = f.get_resource_path()
def is_valid_url(url):
cached = _cache_get(url, 'valid')
key = date.today().isoformat() + '_submissions'
r_cache.zincrby(key, url)
if cached is not None:
return cached
fex = Faup()
if url.startswith('hxxp'):
url = 'http' + url[4:]
elif not url.startswith('http'):
url = 'http://' + url
logging.debug("Checking validity of URL: " + url)
fex.decode(url)
scheme = fex.get_scheme()
host = fex.get_host()
if scheme is None or host is None:
reason = "Not a valid http/https URL/URI"
return False, url, reason
_cache_set(url, (True, url, None), 'valid')
return True, url, None
def is_valid_url(url):
cached = _cache_get(url, 'valid')
key = date.today().isoformat() + '_submissions'
r_cache.zincrby(key, url)
if cached is not None:
return cached
fex = Faup()
if url.startswith('hxxp'):
url = 'http' + url[4:]
elif not url.startswith('http'):
url = 'http://' + url
logging.debug("Checking validity of URL: " + url)
fex.decode(url)
scheme = fex.get_scheme()
host = fex.get_host()
if scheme is None or host is None:
reason = "Not a valid http/https URL/URI"
return False, url, reason
_cache_set(url, (True, url, None), 'valid')
return True, url, None
def sort(self, elem_links, url):
fex = Faup()
f = Filters()
f.load()
self.r.switchDB(1)
extend = True
domainfilter = True
schemefilter = True
try:
for link in elem_links:
new_url = link
self.r.switchDB(2)
if not self.r.get(new_url) and new_url:
self.r.switchDB(1)
if not self.r.get(new_url):
fex.decode(new_url)
domain = fex.get_host()
if f.isfilteredscheme(fex.get_scheme()):
self.r.switchDB(2)
self.r.put(new_url, new_url)
schemefilter = False
if f.isfiltereddomains(domain):
self.r.switchDB(2)
self.r.put(new_url, new_url)
domainfilter = False
if f.isfilteredextention(fex.get_resource_path()):
extend = False
self.r.switchDB(2)
self.r.put(new_url, new_url)
if extend and domainfilter and schemefilter:
self.r.switchDB(1)
self.r.rpush('crawl', new_url)
self.queue.append(new_url)
except TypeError as e:
print "TypeError"
def sort(self, elem_links, url):
fex = Faup()
f = Filters()
f.load()
self.r.switchDB(1)
extend = True
domainfilter = True
schemefilter = True
try:
for link in elem_links:
new_url = link
self.r.switchDB(2)
if not self.r.get(new_url) and new_url:
self.r.switchDB(1)
if not self.r.get(new_url):
fex.decode(new_url)
domain = fex.get_host()
if f.isfilteredscheme(fex.get_scheme()):
self.r.switchDB(2)
self.r.put(new_url, new_url)
schemefilter = False
if f.isfiltereddomains(domain):
self.r.switchDB(2)
self.r.put(new_url, new_url)
domainfilter = False
if f.isfilteredextention(fex.get_resource_path()):
extend = False
self.r.switchDB(2)
self.r.put(new_url, new_url)
if extend and domainfilter and schemefilter:
self.r.switchDB(1)
self.r.rpush('crawl', new_url)
self.queue.append(new_url)
except TypeError as e:
print "TypeError"
for h in hashlist_sha256:
misp.add_hashes(new_event, sha256=h)
if (len(hashlist_md5) > 0) or (len(hashlist_sha1) > 0) or (len(hashlist_sha256)
> 0):
for tag in hash_only_tags:
misp.add_tag(new_event, tag)
# Add IOCs and expanded information to MISP
for entry in urllist:
ids_flag = True
f.decode(entry)
domainname = f.get_domain().decode('utf-8', 'ignore')
hostname = f.get_host().decode('utf-8', 'ignore')
try:
schema = f.get_scheme().decode('utf-8', 'ignore')
except:
schema = False
if debug:
syslog.syslog(domainname)
if domainname not in excludelist:
if domainname in internallist:
misp.add_named_attribute(new_event,
'link',
entry,
category='Internal reference',
to_ids=False,
distribution=0)
elif domainname in externallist:
misp.add_named_attribute(new_event,
'link',
class Mail2MISP():
def __init__(self,
misp_url,
misp_key,
verifycert,
config,
offline=False,
urlsonly=False):
self.offline = offline
if not self.offline:
self.misp = ExpandedPyMISP(misp_url,
misp_key,
verifycert,
debug=config.debug)
self.config = config
self.urlsonly = urlsonly
if not hasattr(self.config, 'enable_dns'):
setattr(self.config, 'enable_dns', True)
if self.urlsonly is False:
setattr(self.config, 'enable_dns', False)
self.debug = self.config.debug
self.config_from_email_body = {}
# Init Faup
self.f = Faup()
self.sightings_to_add = []
def load_email(self, pseudofile):
self.pseudofile = pseudofile
self.original_mail = message_from_bytes(self.pseudofile.getvalue(),
policy=policy.default)
self.subject = self.original_mail.get('Subject')
try:
self.sender = self.original_mail.get('From')
except:
self.sender = "<unknown sender>"
# Remove words from subject
for removeword in self.config.removelist:
self.subject = re.sub(removeword, "", self.subject).strip()
# Initialize the MISP event
self.misp_event = MISPEvent()
self.misp_event.info = f'{self.config.email_subject_prefix} - {self.subject}'
self.misp_event.distribution = self.config.default_distribution
self.misp_event.threat_level_id = self.config.default_threat_level
self.misp_event.analysis = self.config.default_analysis
def sighting(self, value, source):
if self.offline:
raise Exception('The script is running in offline mode, ')
'''Add a sighting'''
s = MISPSighting()
s.from_dict(value=value, source=source)
self.misp.add_sighting(s)
def _find_inline_forward(self):
'''Does the body contains a forwarded email?'''
for identifier in self.config.forward_identifiers:
if identifier in self.clean_email_body:
self.clean_email_body, fw_email = self.clean_email_body.split(
identifier)
return self.forwarded_email(
pseudofile=BytesIO(fw_email.encode()))
def _find_attached_forward(self):
forwarded_emails = []
for attachment in self.original_mail.iter_attachments():
attachment_content = attachment.get_content()
# Search for email forwarded as attachment
# I could have more than one, attaching everything.
if isinstance(attachment_content, message.EmailMessage):
forwarded_emails.append(
self.forwarded_email(
pseudofile=BytesIO(attachment_content.as_bytes())))
else:
if isinstance(attachment_content, str):
attachment_content = attachment_content.encode()
filename = attachment.get_filename()
if not filename:
filename = 'missing_filename'
if self.config_from_email_body.get(
'attachment'
) == self.config.m2m_benign_attachment_keyword:
# Attach sane file
self.misp_event.add_attribute(
'attachment',
value=filename,
data=BytesIO(attachment_content))
else:
f_object, main_object, sections = make_binary_objects(
pseudofile=BytesIO(attachment_content),
filename=filename,
standalone=False)
self.misp_event.add_object(f_object)
if main_object:
self.misp_event.add_object(main_object)
[
self.misp_event.add_object(section)
for section in sections
]
return forwarded_emails
def email_from_spamtrap(self):
'''The email comes from a spamtrap and should be attached as-is.'''
raw_body = self.original_mail.get_body(preferencelist=('html',
'plain'))
if raw_body:
self.clean_email_body = html.unescape(
raw_body.get_payload(decode=True).decode(
'utf8', 'surrogateescape'))
else:
self.clean_email_body = ''
return self.forwarded_email(self.pseudofile)
def forwarded_email(self, pseudofile: BytesIO):
'''Extracts all possible indicators out of an email and create a MISP event out of it.
* Gets all relevant Headers
* Attach the body
* Create MISP file objects (uses lief if possible)
* Set all references
'''
email_object = EMailObject(pseudofile=pseudofile,
attach_original_mail=True,
standalone=False)
if email_object.attachments:
# Create file objects for the attachments
for attachment_name, attachment in email_object.attachments:
if not attachment_name:
attachment_name = 'NameMissing.txt'
if self.config_from_email_body.get(
'attachment'
) == self.config.m2m_benign_attachment_keyword:
a = self.misp_event.add_attribute('attachment',
value=attachment_name,
data=attachment)
email_object.add_reference(a.uuid, 'related-to',
'Email attachment')
else:
f_object, main_object, sections = make_binary_objects(
pseudofile=attachment,
filename=attachment_name,
standalone=False)
if self.config.vt_key:
try:
vt_object = VTReportObject(
self.config.vt_key,
f_object.get_attributes_by_relation(
'sha256')[0].value,
standalone=False)
self.misp_event.add_object(vt_object)
f_object.add_reference(vt_object.uuid,
'analysed-with')
except InvalidMISPObject as e:
print(e)
pass
self.misp_event.add_object(f_object)
if main_object:
self.misp_event.add_object(main_object)
for section in sections:
self.misp_event.add_object(section)
email_object.add_reference(f_object.uuid, 'related-to',
'Email attachment')
self.process_body_iocs(email_object)
if self.config.spamtrap or self.config.attach_original_mail or self.config_from_email_body.get(
'attach_original_mail'):
self.misp_event.add_object(email_object)
return email_object
def process_email_body(self):
mail_as_bytes = self.original_mail.get_body(
preferencelist=('html', 'plain')).get_payload(decode=True)
if mail_as_bytes:
self.clean_email_body = html.unescape(
mail_as_bytes.decode('utf8', 'surrogateescape'))
# Check if there are config lines in the body & convert them to a python dictionary:
# <config.body_config_prefix>:<key>:<value> => {<key>: <value>}
self.config_from_email_body = {
k.strip(): v.strip()
for k, v in re.findall(
f'{self.config.body_config_prefix}:(.*):(.*)',
self.clean_email_body)
}
if self.config_from_email_body:
# ... remove the config lines from the body
self.clean_email_body = re.sub(
rf'^{self.config.body_config_prefix}.*\n?',
'',
html.unescape(
self.original_mail.get_body(
preferencelist=('html', 'plain')).get_payload(
decode=True).decode('utf8',
'surrogateescape')),
flags=re.MULTILINE)
# Check if autopublish key is present and valid
if self.config_from_email_body.get(
'm2mkey') == self.config.m2m_key:
if self.config_from_email_body.get('distribution') is not None:
self.misp_event.distribution = self.config_from_email_body.get(
'distribution')
if self.config_from_email_body.get('threat_level') is not None:
self.misp_event.threat_level_id = self.config_from_email_body.get(
'threat_level')
if self.config_from_email_body.get('analysis') is not None:
self.misp_event.analysis = self.config_from_email_body.get(
'analysis')
if self.config_from_email_body.get('publish'):
self.misp_event.publish()
self._find_inline_forward()
else:
self.clean_email_body = ''
self._find_attached_forward()
def process_body_iocs(self, email_object=None):
if email_object:
body = html.unescape(
email_object.email.get_body(
preferencelist=('html',
'plain')).get_payload(decode=True).decode(
'utf8', 'surrogateescape'))
else:
body = self.clean_email_body
# Cleanup body content
# Depending on the source of the mail, there is some cleanup to do. Ignore lines in body of message
for ignoreline in self.config.ignorelist:
body = re.sub(rf'^{ignoreline}.*\n?', '', body, flags=re.MULTILINE)
# Remove everything after the stopword from the body
body = body.split(self.config.stopword, 1)[0]
# Add tags to the event if keywords are found in the mail
for tag in self.config.tlptags:
for alternativetag in self.config.tlptags[tag]:
if alternativetag in body.lower():
self.misp_event.add_tag(tag)
# Prepare extraction of IOCs
# Refang email data
body = refang(body)
# Extract and add hashes
contains_hash = False
for h in set(re.findall(hashmarker.MD5_REGEX, body)):
contains_hash = True
attribute = self.misp_event.add_attribute(
'md5', h, enforceWarninglist=self.config.enforcewarninglist)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
if self.config.sighting:
self.sightings_to_add.append((h, self.config.sighting_source))
for h in set(re.findall(hashmarker.SHA1_REGEX, body)):
contains_hash = True
attribute = self.misp_event.add_attribute(
'sha1', h, enforceWarninglist=self.config.enforcewarninglist)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
if self.config.sighting:
self.sightings_to_add.append((h, self.config.sighting_source))
for h in set(re.findall(hashmarker.SHA256_REGEX, body)):
contains_hash = True
attribute = self.misp_event.add_attribute(
'sha256', h, enforceWarninglist=self.config.enforcewarninglist)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
if self.config.sighting:
self.sightings_to_add.append((h, self.config.sighting_source))
if contains_hash:
[
self.misp_event.add_tag(tag)
for tag in self.config.hash_only_tags
]
# # Extract network IOCs
urllist = []
urllist += re.findall(urlmarker.WEB_URL_REGEX, body)
urllist += re.findall(urlmarker.IP_REGEX, body)
if self.debug:
syslog.syslog(str(urllist))
hostname_processed = []
# Add IOCs and expanded information to MISP
for entry in set(urllist):
ids_flag = True
self.f.decode(entry)
domainname = self.f.get_domain()
if domainname in self.config.excludelist:
# Ignore the entry
continue
hostname = self.f.get_host()
scheme = self.f.get_scheme()
if scheme:
scheme = scheme
resource_path = self.f.get_resource_path()
if resource_path:
resource_path = resource_path
if self.debug:
syslog.syslog(domainname)
if domainname in self.config.internallist and self.urlsonly is False: # Add link to internal reference unless in urlsonly mode
attribute = self.misp_event.add_attribute(
'link',
entry,
category='Internal reference',
to_ids=False,
enforceWarninglist=False)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
elif domainname in self.config.externallist or self.urlsonly is False: # External analysis
attribute = self.misp_event.add_attribute(
'link',
entry,
category='External analysis',
to_ids=False,
enforceWarninglist=False)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
elif domainname in self.config.externallist or self.urlsonly: # External analysis
if self.urlsonly:
comment = self.subject + " (from: " + self.sender + ")"
else:
comment = ""
attribute = self.misp.add_attribute(
self.urlsonly, {
"type": 'link',
"value": entry,
"category": 'External analysis',
"to_ids": False,
"comment": comment
})
for tag in self.config.tlptags:
for alternativetag in self.config.tlptags[tag]:
if alternativetag in self.subject.lower():
self.misp.tag(attribute["uuid"], tag)
new_subject = comment.replace(alternativetag, '')
self.misp.change_comment(attribute["uuid"],
new_subject)
else: # The URL is probably an indicator.
comment = ""
if (domainname in self.config.noidsflaglist) or (
hostname in self.config.noidsflaglist):
ids_flag = False
comment = "Known host (mostly for connectivity test or IP lookup)"
if self.debug:
syslog.syslog(str(entry))
if scheme:
if is_ip(hostname):
attribute = self.misp_event.add_attribute(
'url',
entry,
to_ids=False,
enforceWarninglist=self.config.enforcewarninglist)
if email_object:
email_object.add_reference(attribute.uuid,
'contains')
else:
if resource_path: # URL has path, ignore warning list
attribute = self.misp_event.add_attribute(
'url',
entry,
to_ids=ids_flag,
enforceWarninglist=False,
comment=comment)
if email_object:
email_object.add_reference(
attribute.uuid, 'contains')
else: # URL has no path
attribute = self.misp_event.add_attribute(
'url',
entry,
to_ids=ids_flag,
enforceWarninglist=self.config.
enforcewarninglist,
comment=comment)
if email_object:
email_object.add_reference(
attribute.uuid, 'contains')
if self.config.sighting:
self.sightings_to_add.append(
(entry, self.config.sighting_source))
if hostname in hostname_processed:
# Hostname already processed.
continue
hostname_processed.append(hostname)
if self.config.sighting:
self.sightings_to_add.append(
(hostname, self.config.sighting_source))
if self.debug:
syslog.syslog(hostname)
comment = ''
port = self.f.get_port()
if port:
port = port
comment = f'on port: {port}'
if is_ip(hostname):
attribute = self.misp_event.add_attribute(
'ip-dst',
hostname,
to_ids=ids_flag,
enforceWarninglist=self.config.enforcewarninglist,
comment=comment)
if email_object:
email_object.add_reference(attribute.uuid, 'contains')
else:
related_ips = []
if HAS_DNS and self.config.enable_dns:
try:
syslog.syslog(hostname)
for rdata in dns.resolver.query(hostname, 'A'):
if self.debug:
syslog.syslog(str(rdata))
related_ips.append(rdata.to_text())
except Exception as e:
if self.debug:
syslog.syslog(str(e))
if related_ips:
hip = MISPObject(name='ip-port')
hip.add_attribute(
'hostname',
value=hostname,
to_ids=ids_flag,
enforceWarninglist=self.config.enforcewarninglist,
comment=comment)
for ip in set(related_ips):
hip.add_attribute('ip',
type='ip-dst',
value=ip,
to_ids=False,
enforceWarninglist=self.config.
enforcewarninglist)
self.misp_event.add_object(hip)
if email_object:
email_object.add_reference(hip.uuid, 'contains')
else:
if self.urlsonly is False:
attribute = self.misp_event.add_attribute(
'hostname',
value=hostname,
to_ids=ids_flag,
enforceWarninglist=self.config.
enforcewarninglist,
comment=comment)
if email_object:
email_object.add_reference(attribute.uuid,
'contains')
def add_event(self):
'''Add event on the remote MISP instance.'''
# Add additional tags depending on others
tags = []
for tag in [t.name for t in self.misp_event.tags]:
if self.config.dependingtags.get(tag):
tags += self.config.dependingtags.get(tag)
# Add additional tags according to configuration
for malware in self.config.malwaretags:
if malware.lower() in self.subject.lower():
tags += self.config.malwaretags.get(malware)
if tags:
[self.misp_event.add_tag(tag) for tag in tags]
has_tlp_tag = False
for tag in [t.name for t in self.misp_event.tags]:
if tag.lower().startswith('tlp'):
has_tlp_tag = True
if not has_tlp_tag:
self.misp_event.add_tag(self.config.tlptag_default)
if self.offline:
return self.misp_event.to_json()
event = self.misp.add_event(self.misp_event, pythonify=True)
if self.config.sighting:
for value, source in self.sightings_to_add:
self.sighting(value, source)
return event
class Query():
def __init__(self, loglevel: int = logging.DEBUG):
self.__init_logger(loglevel)
self.fex = Faup()
self.cache = Redis(unix_socket_path=get_socket_path('cache'),
db=1,
decode_responses=True)
def __init_logger(self, loglevel) -> None:
self.logger = logging.getLogger(f'{self.__class__.__name__}')
self.logger.setLevel(loglevel)
def _cache_set(self, key, value, field=None):
if field is None:
self.cache.setex(key, json.dumps(value), 3600)
else:
self.cache.hset(key, field, json.dumps(value))
self.cache.expire(key, 3600)
def _cache_get(self, key, field=None):
if field is None:
value_json = self.cache.get(key)
else:
value_json = self.cache.hget(key, field)
if value_json is not None:
return json.loads(value_json)
return None
def to_bool(self, s):
"""
Converts the given string to a boolean.
"""
return s.lower() in ('1', 'true', 'yes', 'on')
def get_submissions(self, url, day=None):
if day is None:
day = date.today().isoformat()
else:
day = day.isoformat()
return self.cache.zscore(f'{day}_submissions', url)
def get_mail_sent(self, url, day=None):
if day is None:
day = date.today().isoformat()
else:
day = day.isoformat()
self.fex.decode(url)
host = self.fex.get_host()
return self.cache.sismember(f'{day}_mails', host)
def set_mail_sent(self, url, day=None):
if day is None:
day = date.today().isoformat()
else:
day = day.isoformat()
self.fex.decode(url)
host = self.fex.get_host()
return self.cache.sadd(f'{day}_mails', host)
def is_valid_url(self, url):
cached = self._cache_get(url, 'valid')
key = f'{date.today().isoformat()}_submissions'
self.cache.zincrby(key, 1, url)
if cached is not None:
return cached
if url.startswith('hxxp'):
url = 'http' + url[4:]
elif not url.startswith('http'):
url = 'http://' + url
logging.debug("Checking validity of URL: " + url)
self.fex.decode(url)
scheme = self.fex.get_scheme()
host = self.fex.get_host()
if scheme is None or host is None:
reason = "Not a valid http/https URL/URI"
return False, url, reason
self._cache_set(url, (True, url, None), 'valid')
return True, url, None
def is_ip(self, host):
try:
ipaddress.ip_address(host)
return True
except ValueError:
return False
def try_resolve(self, url):
self.fex.decode(url)
host = self.fex.get_host().lower()
if self.is_ip(host):
return True, None
try:
ipaddr = dns.resolver.query(host, 'A')
except Exception:
reason = "DNS server problem. Check resolver settings."
return False, reason
if not ipaddr:
reason = "Host " + host + " does not exist."
return False, reason
return True, None
def get_urls(self, url, depth=1):
if depth > 5:
print('Too many redirects.')
return
def meta_redirect(content):
c = content.lower()
soup = BeautifulSoup(c, "html.parser")
for result in soup.find_all(attrs={'http-equiv': 'refresh'}):
if result:
out = result["content"].split(";")
if len(out) == 2:
wait, text = out
try:
a, url = text.split('=', 1)
return url.strip()
except Exception:
print(text)
return None
resolve, reason = self.try_resolve(url)
if not resolve:
# FIXME: inform that the domain does not resolve
yield url
return
logging.debug(f"Making HTTP connection to {url}")
headers = {
'User-agent':
'Mozilla/5.0 (Windows NT 6.1; WOW64; rv:8.0) Gecko/20100101 Firefox/8.0'
}
try:
response = requests.get(url,
allow_redirects=True,
headers=headers,
timeout=15,
verify=False)
except Exception:
# That one can fail (DNS for example)
# FIXME: inform that the get failed
yield url
return
if response.history is not None:
for h in response.history:
# Yeld the urls in the order we find them
yield h.url
yield response.url
meta_redir_url = meta_redirect(response.content)
if meta_redir_url is not None:
depth += 1
if not meta_redir_url.startswith('http'):
self.fex.decode(url)
base = '{}://{}'.format(self.fex.get_scheme(),
self.fex.get_host())
port = self.fex.get_port()
if port is not None:
base += f':{port}'
if not meta_redir_url.startswith('/'):
# relative redirect. resource_path has the initial '/'
if self.fex.get_resource_path() is not None:
base += self.fex.get_resource_path()
if not base.endswith('/'):
base += '/'
meta_redir_url = base + meta_redir_url
for url in self.get_urls(meta_redir_url, depth):
yield url
def url_list(self, url):
cached = self._cache_get(url, 'list')
if cached is not None:
return cached
list_urls = []
for u in self.get_urls(url):
if u is None or u in list_urls:
continue
list_urls.append(u)
self._cache_set(url, list_urls, 'list')
return list_urls
def dns_resolve(self, url):
cached = self._cache_get(url, 'dns')
if cached is not None:
return cached
self.fex.decode(url)
host = self.fex.get_host().lower()
ipv4 = None
ipv6 = None
if self.is_ip(host):
if ':' in host:
try:
socket.inet_pton(socket.AF_INET6, host)
ipv6 = [host]
except Exception:
pass
else:
try:
socket.inet_aton(host)
ipv4 = [host]
except Exception:
pass
else:
try:
ipv4 = [str(ip) for ip in dns.resolver.query(host, 'A')]
except Exception:
logging.debug("No IPv4 address assigned to: " + host)
try:
ipv6 = [str(ip) for ip in dns.resolver.query(host, 'AAAA')]
except Exception:
logging.debug("No IPv6 address assigned to: " + host)
self._cache_set(url, (ipv4, ipv6), 'dns')
return ipv4, ipv6
def phish_query(self, url, key, query):
cached = self._cache_get(query, 'phishtank')
if cached is not None:
return cached
postfields = {'url': quote(query), 'format': 'json', 'app_key': key}
response = requests.post(url, data=postfields)
res = response.json()
if res["meta"]["status"] == "success":
if res["results"]["in_database"]:
self._cache_set(query, res["results"]["phish_detail_page"],
'phishtank')
return res["results"]["phish_detail_page"]
else:
# no information
pass
elif res["meta"]["status"] == 'error':
# Inform the user?
# errormsg = res["errortext"]
pass
return None
def sphinxsearch(server, port, url, query):
# WARNING: too dangerous to have on the public interface
return ''
"""
if not sphinx:
return None
cached = _cache_get(query, 'sphinx')
if cached is not None:
return cached
client = sphinxapi.SphinxClient()
client.SetServer(server, port)
client.SetMatchMode(2)
client.SetConnectTimeout(5.0)
result = []
res = client.Query(query)
if res.get("matches") is not None:
for ticket in res["matches"]:
ticket_id = ticket["id"]
ticket_link = url + str(ticket_id)
result.append(ticket_link)
_cache_set(query, result, 'sphinx')
return result
"""
def vt_query_url(self, url, url_up, key, query, upload=True):
cached = self._cache_get(query, 'vt')
if cached is not None and cached[2] is not None:
return cached
parameters = {"resource": query, "apikey": key}
if upload:
parameters['scan'] = 1
response = requests.post(url, data=parameters)
if response.text is None or len(response.text) == 0:
return None
res = response.json()
msg = res["verbose_msg"]
link = res.get("permalink")
positives = res.get("positives")
total = res.get("total")
self._cache_set(query, (msg, link, positives, total), 'vt')
return msg, link, positives, total
def gsb_query(self, url, query):
cached = self._cache_get(query, 'gsb')
if cached is not None:
return cached
param = '1\n' + query
response = requests.post(url, data=param)
status = response.status_code
if status == 200:
self._cache_set(query, response.text, 'gsb')
return response.text
'''
def urlquery_query(url, key, query):
return None
cached = _cache_get(query, 'urlquery')
if cached is not None:
return cached
try:
urlquery.url = url
urlquery.key = key
response = urlquery.search(query)
except Exception:
return None
if response['_response_']['status'] == 'ok':
if response.get('reports') is not None:
total_alert_count = 0
for r in response['reports']:
total_alert_count += r['urlquery_alert_count']
total_alert_count += r['ids_alert_count']
total_alert_count += r['blacklist_alert_count']
_cache_set(query, total_alert_count, 'urlquery')
return total_alert_count
else:
return None
'''
def process_emails(self, emails, ignorelist, replacelist):
to_return = list(set(emails))
for mail in reversed(to_return):
for ignorelist_entry in ignorelist:
if re.search(ignorelist_entry, mail, re.I):
if mail in to_return:
to_return.remove(mail)
for k, v in list(replacelist.items()):
if re.search(k, mail, re.I):
if k in to_return:
to_return.remove(k)
to_return += v
return to_return
def whois(self, server, port, domain, ignorelist, replacelist):
cached = self._cache_get(domain, 'whois')
if cached is not None:
return cached
s = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
s.settimeout(15)
try:
s.connect((server, port))
except Exception:
print("Connection problems - check WHOIS server")
print(("WHOIS request while problem occurred: ", domain))
print(("WHOIS server: {}:{}".format(server, port)))
return None
if domain.startswith('http'):
self.fex.decode(domain)
d = self.fex.get_domain().lower()
else:
d = domain
s.send(("{}\r\n".format(d)).encode())
response = b''
while True:
d = s.recv(4096)
response += d
if d == b'':
break
s.close()
match = re.findall(r'[\w\.-]+@[\w\.-]+', response.decode())
emails = self.process_emails(match, ignorelist, replacelist)
if len(emails) == 0:
return None
list_mail = list(set(emails))
self._cache_set(domain, list_mail, 'whois')
return list_mail
def pdnscircl(self, url, user, passwd, q):
cached = self._cache_get(q, 'pdns')
if cached is not None:
return cached
pdns = PyPDNS(url, basic_auth=(user, passwd))
response = pdns.query(q)
all_uniq = []
for e in reversed(response):
host = e['rrname'].lower()
if host in all_uniq:
continue
else:
all_uniq.append(host)
response = (len(all_uniq), all_uniq[:5])
self._cache_set(q, response, 'pdns')
return response
def psslcircl(self, url, user, passwd, q):
cached = self._cache_get(q, 'pssl')
if cached is not None:
return cached
pssl = PyPSSL(url, basic_auth=(user, passwd))
response = pssl.query(q)
if response.get(q) is not None:
certinfo = response.get(q)
entries = {}
for sha1 in certinfo['certificates']:
entries[sha1] = []
if certinfo['subjects'].get(sha1):
for value in certinfo['subjects'][sha1]['values']:
entries[sha1].append(value)
self._cache_set(q, entries, 'pssl')
return entries
return None
def eupi(self, url, key, q):
cached = self._cache_get(q, 'eupi')
if cached is not None:
return cached
eu = PyEUPI(key, url)
response = eu.search_url(url=q)
if response.get('results'):
r = response.get('results')[0]['tag_label']
self._cache_set(q, r, 'eupi')
return r
eu.post_submission(q)
return None
def bgpranking(self, ip):
cached = self._cache_get(ip, 'ipasn')
if cached is not None:
asn = cached['asn']
prefix = cached['prefix']
else:
ipasn = IPASNHistory()
response = ipasn.query(ip)
if 'response' not in response:
asn = None
prefix = None
entry = response['response'][list(response['response'].keys())[0]]
if entry:
self._cache_set(ip, entry, 'ipasn')
asn = entry['asn']
prefix = entry['prefix']
else:
asn = None
prefix = None
if not asn or not prefix:
# asn, prefix, asn_descr, rank, position, known_asns
return None, None, None, None, None, None
cached = self._cache_get(ip, 'bgpranking')
if cached is not None:
return cached
bgpranking = BGPRanking()
response = bgpranking.query(asn,
date=(date.today() -
timedelta(1)).isoformat())
if 'response' not in response or not response['response']:
return None, None, None, None, None, None
to_return = (asn, prefix, response['response']['asn_description'],
response['response']['ranking']['rank'],
response['response']['ranking']['position'],
response['response']['ranking']['total_known_asns'])
self._cache_set(ip, to_return, 'bgpranking')
return to_return
def lookyloo(self, url):
cached = self._cache_get(url, 'lookyloo')
if cached is not None:
return cached
lookyloo = Lookyloo()
lookyloo_perma_url = lookyloo.enqueue(url)
if lookyloo_perma_url:
self._cache_set(url, lookyloo_perma_url, 'lookyloo')
return lookyloo_perma_url
return None
def _deserialize_cached(self, entry):
to_return = {}
redirects = []
h = self.cache.hgetall(entry)
for key, value in h.items():
v = json.loads(value)
if key == 'list':
redirects = v
continue
to_return[key] = v
return to_return, redirects
def get_url_data(self, url):
data, redirects = self._deserialize_cached(url)
if data.get('dns') is not None:
ipv4, ipv6 = data['dns']
ip_data = {}
if ipv4 is not None:
for ip in ipv4:
info, _ = self._deserialize_cached(ip)
ip_data[ip] = info
if ipv6 is not None:
for ip in ipv6:
info, _ = self._deserialize_cached(ip)
ip_data[ip] = info
if len(ip_data) > 0:
data.update(ip_data)
return {url: data}, redirects
def cached(self, url, digest=False):
url_data, redirects = self.get_url_data(url)
to_return = [url_data]
for u in redirects:
if u == url:
continue
data, redir = self.get_url_data(u)
to_return.append(data)
if digest:
return {'result': to_return, 'digest': self.digest(to_return)}
return {'result': to_return}
def ip_details_digest(self, ips, all_info, all_asns, all_mails):
to_return = ''
for ip in ips:
to_return += '\t' + ip + '\n'
data = all_info[ip]
if data.get('bgpranking'):
to_return += '\t\tis announced by {} ({}). Position {}/{}.\n'.format(
data['bgpranking'][2], data['bgpranking'][0],
data['bgpranking'][4], data['bgpranking'][5])
all_asns.add('{} ({})'.format(data['bgpranking'][2],
data['bgpranking'][0]))
if data.get('whois'):
all_mails.update(data.get('whois'))
return to_return
def digest(self, data):
to_return = ''
all_mails = set()
all_asns = set()
for entry in data:
# Each URL we're redirected to
for url, info in entry.items():
# info contains the information we got for the URL.
to_return += '\n{}\n'.format(url)
if 'whois' in info:
all_mails.update(info['whois'])
if 'lookyloo' in info:
to_return += '\tLookyloo permanent URL: {}\n'.format(
info['lookyloo'])
if 'vt' in info and len(info['vt']) == 4:
if info['vt'][2] is not None:
to_return += '\t{} out of {} positive detections in VT - {}\n'.format(
info['vt'][2], info['vt'][3], info['vt'][1])
else:
to_return += '\t{} - {}\n'.format(
info['vt'][0], info['vt'][1])
if 'gsb' in info:
to_return += '\tKnown as malicious on Google Safe Browsing: {}\n'.format(
info['gsb'])
if 'phishtank' in info:
to_return += '\tKnown on PhishTank: {}\n'.format(
info['phishtank'])
if 'dns' in info:
ipv4, ipv6 = info['dns']
if ipv4 is not None:
to_return += self.ip_details_digest(
ipv4, info, all_asns, all_mails)
if ipv6 is not None:
to_return += self.ip_details_digest(
ipv6, info, all_asns, all_mails)
return to_return, list(all_mails), list(all_asns)
