def urlscan_screenshot(url):
#First, let's see if it's been indexed already.
domain = tldextract.extract(url).fqdn
print("\t\t[-] Looking up domain screenshot {}".format(domain))
if lookup_screenshot(url):
print ("\t\t[-] Already screenshotted...")
return
else:
print ("\t\t[-] No screenshot yet, sending to urlscan...")
urlscanio = "https://urlscan.io/api/v1/scan/"
key = urlscanapikey
headers = {
"Content-Type" : "application/json",
"API-Key" : key
}
data = {
"url" : url,
"public" : "on"
}
try:
r = requests.post(urlscanio, headers=headers, json=data)
save_screenshot(url, r.json()['result'])
except Exception as e:
print (e)
pass
time.sleep(2)
def get_title_for_classifier(self):
title = self.soup.find("title")
domain = tldextract.extract(self.current_link)
return_string = domain.domain + ", " + re.sub(
r'[^\w]', " ",
title.get_text().strip().lower().replace(domain.domain, ""))
return return_string
def __init__(self, url, config=None, **kwargs):
"""The config object for this source will be passed into all of this
source's children articles unless specified otherwise or re-set.
"""
if (url is None) or ('://' not in url) or (url[:4] != 'http'):
raise Exception('Input url is bad!')
self.config = config or Configuration()
self.config = utils.extend_config(self.config, kwargs)
self.extractor = self.config.extractor(self.config)
self.url = url
self.url = urls.prepare_url(url)
self.domain = urls.get_domain(self.url)
self.scheme = urls.get_scheme(self.url)
self.categories = []
self.feeds = []
self.articles = []
self.html = ''
self.doc = None
self.logo_url = ''
self.favicon = ''
self.brand = tldextract.extract(self.url).domain
self.description = ''
self.is_parsed = False
self.is_downloaded = False
def enviar_noticias(resultado, id_chat, Nombre_Grupo, provincias, tema):
try:
url_api = "bot1477154971:AAHz2Ok9QD8bwzkAxIqqZc64GNPeqGjuRTI/sendMessage"
temas = ''.join(tema)
for m in resultado:
titulo = m[1]
linkNoticia = m[0]
descripcion = m[2]
contenido = m[3]
extracted = tldextract.extract(linkNoticia)
medio = "{}.{}".format(extracted.domain, extracted.suffix)
medio = medio.replace(".com", "").replace(".ar", "")
mensaje = "Medio: " + medio + "\n\n" + " Última Noticia: " + titulo + "\n\n" + " Ver más en ->" + linkNoticia
requests.post('https://api.telegram.org/' + url_api,
data={
'chat_id': id_chat,
'text': mensaje
})
print(requests.status_codes)
try:
mycursor = mydb.cursor()
sql = "INSERT INTO noticias_enviadas (link,tema,id_grupo) " \
"VALUES (%s, %s, %s) "
val = (linkNoticia, temas, id_chat)
mycursor.execute(sql, val)
mydb.commit()
print("insertó correctamente el link: " + linkNoticia + "")
except Exception as e:
print("El Link ya fue guardado: " + linkNoticia + "")
except Exception as e:
print(" 279 - enviar ", e)
def compare_to(self, compared_url):
""" Get other website url and a word
Finds the number of occurrences in the compared website
Prints data in a table
"""
# find number of letters & most common word in current website
num_of_letters, most_common_word = self.get_text_info()
# find number of occurrences in the compared website
text = self.parse(compared_url)
compared_url_occur = text.count(most_common_word[0])
# create a table and print data
table = PrettyTable(['Description', 'Value'])
table.add_row(['Number of Letters in ' + self.name, num_of_letters])
table.add_row([
'Most Common Word in ' + self.name,
"The word '" + str(most_common_word[0]) + "'" + " occurs " +
str(most_common_word[1]) + " times"
])
table.add_row([
'Number of Occurrences in ' +
tldextract.extract(compared_url).domain,
str(compared_url_occur)
])
print(table)
async def select_proxy(self, url: str) -> str:
if not self._redis_client:
await self.setup()
await self.cleanup()
domain = tldextract.extract(url).domain
while True:
free_slots = await asyncio.gather(*[
proxy.get_number_of_free_slots(domain, self._redis_client)
for proxy in self._proxies
])
slots_and_proxies = list(zip(free_slots, self._proxies))
shuffle(slots_and_proxies)
proxies = sorted(slots_and_proxies,
key=lambda d: d[0],
reverse=True)
if proxies[0][0] > 0:
return await proxies[0][1].get_url(domain, self._redis_client)
else:
# No proxy available right now. Wait.
await asyncio.sleep(5)
async def resolve(self, domain):
if not self.bootstraped:
await self.bootstrap()
tld = tldextract.extract(domain).suffix
apex_domain = tldextract.extract(domain).registered_domain
try:
authorative_server = self.server_list[tld]
async with aiohttp.ClientSession() as session:
res = await self._fetch_json(session, "%sdomain/%s" %(authorative_server, apex_domain ))
if res:
entry = RdapDomainEntry.from_rdap_response(res)
return entry
else:
raise LookupError('no valid rdap data received')
except KeyError:
raise LookupError("No rdap server found for domain: %s" % domain)
def update_json_information(self, site):
time.sleep(3)
print("Updating json with recording information")
with open(site["json_path"], "r") as f:
json_data = json.loads(f.read())
# Generate the eTLD+1 list
# Bug 1585598 - Validate list of sites used for testing Fission
etdl = []
for item in json_data["http_protocol"].keys():
base_url = ".".join(item.split(".")[-2:])
if base_url not in etdl:
etdl.append(base_url)
self.information["etdl"] = etdl
self.information["proxy"] = self.proxy
self.information["url"] = site["url"]
self.information["domain"] = tldextract.extract(site["url"]).domain
self.information["label"] = site.get("label")
json_data["info"] = self.information
with open(site["json_path"], "w") as f:
f.write(json.dumps(json_data, sort_keys=True, indent=2))
def getTargetValidityProblems(self):
"""Verify that each target has a valid TLD in order to prevent problematic rewrite
as stated in EFForg/https-everywhere/issues/10877. In particular,
right-wildcard target are ignored from this test.
Returns an array of strings reporting any coverage problems if they exist,
or empty list if coverage is sufficient.
"""
problems = self._determineTestApplication()
# Next, make sure each target has a valid TLD and doesn't overlap with others
for target in self.targets:
# If it's a wildcard, check which other targets it covers
if '*' in target:
target_re = regex.escape(target)
if target_re.startswith(r'\*'):
target_re = target_re[2:]
else:
target_re = r'\A' + target_re
target_re = regex.compile(
target_re.replace(r'\*', r'[^.]*') + r'\Z')
others = [other for other in self.targets if other !=
target and target_re.search(other)]
if others:
problems.append("%s: Target '%s' also covers %s" %
(self.filename, target, others))
# Ignore right-wildcard targets for TLD checks
if target.endswith(".*"):
continue
# Ignore if target is an ipv4 address
try:
socket.inet_aton(target)
continue
except:
pass
# Ignore if target is an ipv6 address
try:
socket.inet_pton(socket.AF_INET6, target)
continue
except:
pass
# Extract TLD from target if possible
res = tldextract.extract(target)
if res.suffix == "":
problems.append("%s: Target '%s' missing eTLD" %
(self.filename, target))
elif res.domain == "":
problems.append("%s: Target '%s' containing entire eTLD" % (
self.filename, target))
return problems
def is_polite(self, url):
ext = tldextract.extract(url)
URL = ext.subdomain + "." + ext.domain + "." + ext.suffix
rp = urllib.robotparser.RobotFileParser()
URL = "https://" + URL + "/robots.txt"
rp.set_url(URL)
rp.read()
flag = rp.can_fetch("*", url)
return flag
async def register_status_code(self, url: str, status_code: int,
proxy_url: str):
domain = tldextract.extract(url).domain
for proxy in self._proxies:
if proxy._url == proxy_url:
await proxy.register_status_code(status_code, domain,
self._redis_client)
break
def get_hostname(hostname):
h_data = tldextract.extract(hostname)
if h_data.subdomain:
hostname = h_data.subdomain
else:
if h_data.registered_domain:
hostname = h_data.registered_domain
return hostname
def lookup_screenshot(url):
domain = tldextract.extract(url).fqdn
if os.path.exists("screenshots/mapping.json"):
f = open("screenshots/mapping.json", 'r')
current = f.read()
f.close()
current = json.loads(current)
return current.get(domain, False)
else:
return False
def get_domain(uri):
"""
Given an uri returns it's domain ("without http://")
:param uri:
:return:
"""
parsed_uri = urlparse(uri)
netloc = '{uri.netloc}'.format(uri=parsed_uri)
extract_result = tldextract.extract(netloc)
domain = extract_result.domain + DOT + extract_result.suffix
return domain.strip().replace("\"\"", "")
def get_hosts_set():
for num in range(len(URLS)):
with open(f'ads_{num}.txt') as f:
for line in f:
if line.startswith('#') or ('0.0.0.0' not in line
and '127.0.0.1' not in line):
continue
domain = line.strip().split(' ')[1]
extracted = tldextract.extract(domain)
if extracted.suffix:
yield extracted
def classify_links(self):
self.same_domain_links = []
self.sub_directory_links = []
self.same_directory_links = []
self.super_directory_links = []
self.sub_domain_links = []
self.super_domain_links = []
self.sister_domain_links = []
self.external_links = []
dir_url = '/'.join(self.url.split('://')[1].split('/')[:-1])
url_tld = extract(self.url)
for link in self.all_links:
link_tld = extract(link['href'])
if link_tld.tld == url_tld.tld and link_tld.domain == url_tld.domain:
# internal link within the domain name e.g. bbc.co.uk
if link_tld.subdomain == url_tld.subdomain:
# we are at the same subdomain
self.same_domain_links.append(link)
if link['href'].startswith(dir_url):
remainder = link['href'].replace(dir_url, '')
r_segs = remainder.split('/')
if len(r_segs) == 2:
self.same_directory_links.append(link)
elif len(r_segs) > 2:
self.sub_directory_links.append(link)
else:
self.super_directory_links.append(link)
elif link_tld.subdomain.endswith(url_tld.subdomain):
# this is a subdomain of the current_domain
self.sub_domain_links.append(link)
elif url_tld.subdomain.endswith(link_tld.subdomain):
# this is a superdomain of the current domain
self.super_domain_links.append(link)
else:
# this must be a sister domain of some kind
self.sister_domain_links.append(link)
else:
# external link
self.external_links.append(link)
def identify_short_url_by_suffix(self):
"""
检查URL的后缀是否为常见的短链后缀
:param url_to_tld_dict: 从文本内容中提取的URL:TLD字典
:return: 经过后缀检查后的URL:TLD字典
"""
# 可能属于短链的URL:TLD的字典
short_url_to_tld_dict = {}
# 项目根目录的路径
rootPath = os.path.split(curPath)[0]
# 短链服务提供商域名列表的绝对路径，由此来获取常用的短链后缀
short_url_service_domain_list_path_absolute = rootPath + "/data/short_url_services_list.txt"
short_url_service_domain_list = loads_file_from_txt_to_list(
fpath=short_url_service_domain_list_path_absolute)
# 常见短链服务提供商的域名后缀集合
suffix_set = set()
for url in short_url_service_domain_list:
suffix = tldextract.extract(url)[2]
suffix_set.add(suffix)
# 通用域名，不列为可疑的短链域名后缀
common_suffix_list = ['com', 'net', 'info', 'org', '']
for common_suffix in common_suffix_list:
suffix_set.remove(common_suffix)
# print("短链服务提供商的常见域名后缀的集合：{0}".format(suffix_set))
for url in self.url_to_tld_dict.keys():
flag = False
input_url_suffix = tldextract.extract(self.url_to_tld_dict[url])[2]
# 传入URL的TLD
tld = self.url_to_tld_dict[url]
for suffix in suffix_set:
if input_url_suffix == suffix:
flag = True
short_url_to_tld_dict[url] = tld
if flag:
self.judge_dict[url].append(True)
else:
self.judge_dict[url].append(False)
return short_url_to_tld_dict
def get_base_url(url):
"""
Takes as input a url, returns the protocol,domain and suffix concatenated
to form the base url of the website. Uses the tldextract library.
"""
tld = tldextract.extract(url)
print(tld.subdomain, ' - ', tld.domain, ' - ', tld.suffix)
if tld.subdomain != "":
base_url = '.'.join([tld.subdomain, tld.domain, tld.suffix])
else:
base_url = '.'.join([tld.domain, tld.suffix])
return base_url
def get_domain(url):
"""
Return the domain without any subdomain
>>> get_domain("http://blog.example.com")
'example.com'
>>> get_domain("http://www.example.com")
'example.com'
>>> get_domain("http://deeper.blog.example.co.uk")
'example.co.uk'
"""
return ".".join(tldextract.extract(url)[-2:])
def get_open_page_rank(self, url):
try:
o = tldextract.extract(url)
domain = ('%s.%s' % (o.domain, o.suffix))
try:
pginfo = self.page_rank.pg[domain]
except KeyError:
config.logger.warn('page rank information for domain [' +
domain + '] not found')
return MISSING_FEATURE * 2, True
return [pginfo['page_rank_decimal'], pginfo['rank']], False
except Exception as e:
config.logger.error(repr(e))
return MISSING_FEATURE * 2, True
def post_reformat(post):
return [
post.id, # 0 post id
Markup((post.title[:120] + '...') if len(post.title) > 120 else post.title), # 1 post title
Markup(post.url), # 2 full url
tldextract.extract(post.url).registered_domain, # 3 domain only url
Markup(post.username), # 4 user name
datetime.datetime.fromtimestamp(post.datetime).strftime('%c'), # 5 tooltip time
human_datetime(post.datetime), # 6 time since post
post.tags, # 7 tags
len(post.votes.split(';')), # 8 votes
len(post.reports.split(';')), # 9 reports
int(time.time() - post.datetime) # 10 epoch seconds
]
def save_screenshot(url, urlscanurl):
domain = tldextract.extract(url).fqdn
if os.path.exists("screenshots/mapping.json"):
f = open("screenshots/mapping.json", 'r')
current = f.read()
f.close()
current = json.loads(current)
current[domain] = urlscanurl
f = open("screenshots/mapping.json", 'w')
current = f.write(json.dumps(current))
f.close()
else:
current = {domain: urlscanurl}
f = open("screenshots/mapping.json", 'w')
current = f.write(json.dumps(current))
f.close()
def get_url_info(url, guess_and_check, max_depth):
# Create a new Link instance
link = Link()
# Save the extracted URL
link.url = url
# Remove markdown and other artifacts from the URL
link.url_clean = remove_markdown(url)
# Check if the clean URL is valid, if so continue with the next steps
link.url_clean_is_valid = check_if_valid_url(link.url_clean)
if link.url_clean_is_valid:
link.is_amp = check_if_amp(link.url_clean)
if link.is_amp:
link.is_cached = check_if_cached(link.url_clean)
link.domain = tldextract.extract(link.url_clean).domain
link = get_canonical(link, guess_and_check, max_depth)
return link
def is_valid_url(url):
extract = tldextract.extract(url)
if not extract.suffix:
return False
# Validators catches 'most' invalid urls, but there are some issues and exceptions that are not really likely
# to cause any major issues in our software. The other alternative is another library with other quircks.
# see: https://github.com/kvesteri/validators/
# Note that this library does not account for 'idna' / punycode encoded domains, so you have to convert
# them yourself. luckily:
# 'аренда.орг' -> 'xn--80aald4bq.xn--c1avg'
# 'google.com' -> 'google.com'
valid_domain = domain(url.encode('idna').decode())
if valid_domain is not True:
return False
return True
def handle(self, *args, **options):
imported_domains = 0
already_existing_domains = 0
file_path = options.get('file', None)
with open(file_path, newline='') as file:
line = file.readline()
while line:
url = line
line = file.readline()
if not urlparse(url).scheme:
url = 'http://' + line
# This uses a list of public suffixes
tld_extract_result = tldextract.extract(url)
if tld_extract_result.subdomain:
url_full_domain = '.'.join([
tld_extract_result.subdomain,
tld_extract_result.domain, tld_extract_result.suffix
])
elif tld_extract_result.suffix:
url_full_domain = '.'.join(
[tld_extract_result.domain, tld_extract_result.suffix])
else:
url_full_domain = tld_extract_result.domain
if ProxyBlacklistedDomain.objects.filter(
domain=url_full_domain).exists():
logger.info('Domain %s already exists, not importing.' %
url_full_domain)
already_existing_domains += 1
continue
ProxyBlacklistedDomain.objects.create(domain=url_full_domain)
logger.info('Imported domain %s successfully.' %
url_full_domain)
imported_domains += 1
logger.info(
'Finished. Imported %d domains and skipped %d as they already existed'
% (imported_domains, already_existing_domains))
def get_url_info(url, use_gac, max_depth) -> Link:
link = Link(canonicals=[])
origin = UrlMeta(url=remove_markdown(url))
origin.is_valid = check_if_valid_url(origin.url)
origin.is_amp = check_if_amp(origin.url) and not any(
map(origin.url.__contains__, static.DENYLISTED_DOMAINS))
if origin.is_valid:
if origin.is_amp:
origin.is_cached = check_if_cached(origin.url)
origin.domain = tldextract.extract(origin.url).domain
link.origin = origin
link = get_canonicals(link=link,
max_depth=max_depth,
use_gac=use_gac)
return link
def valid_image_url(url):
url = clean_url(url)
if not url.startswith(('http://', 'https://')):
return False
path = urlparse(url).path
# input url is not in valid form (scheme, netloc, tld)
if not path.startswith('/'):
return False
# the '/' which may exist at the end of the url provides us no information
if path.endswith('/'):
path = path[:-1]
# '/story/cnn/blahblah/index.html' --> ['story', 'cnn', 'blahblah', 'index.html']
path_chunks = [x for x in path.split('/') if len(x) > 0]
# siphon out the file type. eg: jpeg, png
if len(path_chunks) > 0:
file_type = urls.url_to_filetype(url)
# if the file type is a media type, reject instantly
if file_type and file_type not in ALLOWED_TYPES:
return False
last_chunk = path_chunks[-1].split('.')
# the file type is not of use to use anymore, remove from url
if len(last_chunk) > 1:
path_chunks[-1] = last_chunk[-2]
# extract the tld (top level domain)
tld_dat = tldextract.extract(url)
tld = tld_dat.domain.lower()
if tld in urls.BAD_DOMAINS:
return False
for d in BAD_URLS:
if d in url:
return False
return True
def create_tenant(domain='', **extra_fields):
extra_fields.setdefault('name', '')
extra_fields.setdefault('country', '')
extra_fields.setdefault('currency', '')
if domain and not freemail.is_free(domain):
# We can guess some field values based on the domain.
tld = tldextract.extract(domain)
geo_ip = GeoIP2()
if not extra_fields['name']:
# Use the domain of the email address as tenant name.
extra_fields['name'] = tld.domain.title()
if not extra_fields['country']:
try:
country_code = geo_ip.country(
tld.registered_domain).get('country_code')
except (gaierror, AddressNotFoundError):
pass
else:
if country_code in [c[0] for c in COUNTRIES]:
extra_fields['country'] = country_code
if extra_fields['country'] and not extra_fields['currency']:
currency = get_territory_currencies(
extra_fields['country'])[-1]
if currency in [c[0] for c in CURRENCIES]:
extra_fields['currency'] = currency
if settings.BILLING_ENABLED:
# Chargebee needs extra info on who to bill, so for now only create the plans without activating the trial.
plan, created = Plan.objects.get_or_create(
name=settings.CHARGEBEE_PRO_TRIAL_PLAN_NAME)
billing = Billing.objects.create(plan=plan)
else:
billing = Billing.objects.create()
tenant = Tenant.objects.create(billing=billing, **extra_fields)
create_defaults_for_tenant(tenant)
return tenant
def enviar_noticias(arr, id_chat, Nombre_Grupo, provincias, tema):
try:
url_api = "bot1477154971:AAHz2Ok9QD8bwzkAxIqqZc64GNPeqGjuRTI/sendMessage"
temas = ''.join(tema)
for m in arr:
linkPortal = m[1]
linkNoticia = m[0]
mycursor = mydb.cursor()
#sql = "SELECT * FROM noticias_enviadas WHERE link = '"+str(linkNoticia)+"'" +""+""
#mycursor.execute(sql)
#records = cursor.fetchall()
#if records == []:
extracted = tldextract.extract(linkPortal)
medio = "{}.{}".format(extracted.domain, extracted.suffix)
medio = medio.replace(".com", "").replace(".ar", "")
linktitulo = requests.get(m[0]).text
titulo = funcion_BuscaTitulo(linktitulo)
if isinstance(titulo, str):
mensaje = "Medio: " + medio + "\n\n" + " Última Noticia: " + titulo + "\n\n" + " Ver más en ->" + linkNoticia
requests.post('https://api.telegram.org/' + url_api,
data={
'chat_id': id_chat,
'text': mensaje
})
print(requests.status_codes)
try:
mycursor = mydb.cursor()
sql = "INSERT INTO noticias_enviadas (link,tema,id_grupo) " \
"VALUES (%s, %s, %s) "
val = (linkNoticia, temas, id_chat)
mycursor.execute(sql, val)
mydb.commit()
print("insertó correctamente el link: " + linkNoticia + "")
except Exception as e:
print("El Link ya fue guardado: " + linkNoticia + "")
except Exception as e:
print(" 279 - enviar ", e)
def set_url(self, url):
url = url.strip()
res = extract(url)
segs = []
for seg in [res.subdomain, res.domain, res.tld]:
if seg:
segs.append(seg)
domain_str = '.'.join(segs)
domain = self.get_domain_entry(domain_str)
if not domain:
domain = Domain(
url=domain_str) # create the domain guessing the values
domain.save()
setattr(self, 'domain_name', domain_str)
setattr(self, 'domain_data', domain)
setattr(self, 'link', url.split(domain_str, 1)[1])
self.get_soup()
self.set_title()
self.set_date()
self.set_content()
