chg: [Crawler] crawler accept all kind of domains

This commit is contained in:
Terrtia 2018-08-24 10:13:56 +02:00
parent e9580d6775
commit 7e24943537
No known key found for this signature in database
GPG key ID: 1E1B1F50D84613D0
6 changed files with 112 additions and 69 deletions

View file

@ -34,21 +34,21 @@ def crawl_onion(url, domain, date, date_month):
exit(0)
if r.status_code == 200:
process = subprocess.Popen(["python", './torcrawler/tor_crawler.py', url, domain, paste, super_father],
process = subprocess.Popen(["python", './torcrawler/tor_crawler.py', splash_url, http_proxy, type_hidden_service, url, domain, paste, super_father],
stdout=subprocess.PIPE)
while process.poll() is None:
time.sleep(1)
if process.returncode == 0:
if r_serv_metadata.exists('paste_children:'+paste):
msg = 'infoleak:automatic-detection="onion";{}'.format(paste)
msg = 'infoleak:automatic-detection="{}";{}'.format(type_hidden_service, paste)
p.populate_set_out(msg, 'Tags')
print(process.stdout.read())
else:
r_onion.sadd('onion_down:'+date , domain)
r_onion.sadd('onion_down_link:'+date , url)
r_onion.sadd('{}_down:{}'.format(type_hidden_service, date), domain)
r_onion.sadd('{}_down_link:{}'.format(type_hidden_service, date), url)
print(process.stdout.read())
else:
## FIXME: # TODO: relaunch docker
@ -67,8 +67,28 @@ if __name__ == '__main__':
# Setup the I/O queues
p = Process(config_section)
splash_url = p.config.get("Crawler", "splash_url")
http_proxy = p.config.get("Crawler", "http_proxy")
url_onion = "((http|https|ftp)?(?:\://)?([a-zA-Z0-9\.\-]+(\:[a-zA-Z0-9\.&%\$\-]+)*@)*((25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9])\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[0-9])|localhost|([a-zA-Z0-9\-]+\.)*[a-zA-Z0-9\-]+\.onion)(\:[0-9]+)*(/($|[a-zA-Z0-9\.\,\?\'\\\+&%\$#\=~_\-]+))*)"
re.compile(url_onion)
url_i2p = "((http|https|ftp)?(?:\://)?([a-zA-Z0-9\.\-]+(\:[a-zA-Z0-9\.&%\$\-]+)*@)*((25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9])\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[0-9])|localhost|([a-zA-Z0-9\-]+\.)*[a-zA-Z0-9\-]+\.i2p)(\:[0-9]+)*(/($|[a-zA-Z0-9\.\,\?\'\\\+&%\$#\=~_\-]+))*)"
re.compile(url_i2p)
type_hidden_service = 'onion'
if type_hidden_service == 'onion':
regex_hidden_service = url_onion
splash_url = p.config.get("Crawler", "splash_url_onion")
http_proxy = p.config.get("Crawler", "http_proxy_onion")
elif type_hidden_service == 'i2p':
regex_hidden_service = url_i2p
splash_url = p.config.get("Crawler", "splash_url_i2p")
http_proxy = p.config.get("Crawler", "http_proxy_i2p")
elif type_hidden_service == 'regular':
regex_hidden_service = url_i2p
splash_url = p.config.get("Crawler", "splash_url_onion")
http_proxy = p.config.get("Crawler", "http_proxy_onion")
else:
print('incorrect crawler type: {}'.format(type_hidden_service))
exit(0)
crawler_depth_limit = p.config.getint("Crawler", "crawler_depth_limit")
#signal.signal(signal.SIGINT, signal_handler)
@ -91,93 +111,94 @@ if __name__ == '__main__':
db=p.config.getint("ARDB_Onion", "db"),
decode_responses=True)
url_regex = "((http|https|ftp)?(?:\://)?([a-zA-Z0-9\.\-]+(\:[a-zA-Z0-9\.&%\$\-]+)*@)*((25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9])\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[1-9]|0)\.(25[0-5]|2[0-4][0-9]|[0-1]{1}[0-9]{2}|[1-9]{1}[0-9]{1}|[0-9])|localhost|([a-zA-Z0-9\-]+\.)*[a-zA-Z0-9\-]+\.onion)(\:[0-9]+)*(/($|[a-zA-Z0-9\.\,\?\'\\\+&%\$#\=~_\-]+))*)"
re.compile(url_regex)
while True:
message = p.get_from_set()
# Recovering the streamed message informations.
#message = r_onion.spop('mess_onion')
print(message)
message = r_onion.spop('{}_crawler_queue'.format(type_hidden_service))
#message='https://www.myip.com/;/home/aurelien/git/python3/AIL-framework/PASTES/crawled/2018/08/10/onionsnjajzkhm5g.onion49eac19d-d71b-48b5-bc55-9a3c63e5b1e2'
# # FIXME: remove
if message is None:
print('get ardb message')
message = r_onion.spop('mess_onion')
print(message)
if message is not None:
splitted = message.split(';')
if len(splitted) == 2:
url, paste = splitted
if not '.onion' in url:
print('not onion')
continue
url_list = re.findall(url_regex, url)[0]
url_list = re.findall(regex_hidden_service, url)[0]
if url_list[1] == '':
url= 'http://{}'.format(url)
link, s, credential, subdomain, domain, host, port, \
resource_path, query_string, f1, f2, f3, f4 = url_list
domain = url_list[4]
r_onion.srem('onion_domain_crawler_queue', domain)
#domain = 'myip.com'
domain_url = 'http://{}'.format(domain)
print('------------------START ONION CRAWLER------------------')
print('------------------START CRAWLER------------------')
print(type_hidden_service)
print('-------------------------------------------------')
print('url: {}'.format(url))
print('domain: {}'.format(domain))
print('domain_url: {}'.format(domain_url))
'''if not r_onion.sismember('full_onion_up', domain):
r_onion.sadd('mess_onion', message)
print('added ..............')'''
if not r_onion.sismember('banned_onion', domain):
if not r_onion.sismember('banned_{}'.format(type_hidden_service), domain):
date = datetime.datetime.now().strftime("%Y%m%d")
date_month = datetime.datetime.now().strftime("%Y%m")
if not r_onion.sismember('month_onion_up:{}'.format(date_month), domain) and not r_onion.sismember('onion_down:'+date , domain):
if not r_onion.sismember('month_{}_up:{}'.format(type_hidden_service, date_month), domain) and not r_onion.sismember('{}_down:{}'.format(type_hidden_service, date), domain):
crawl_onion(url, domain, date, date_month)
if url != domain_url:
crawl_onion(domain_url, domain, date, date_month)
# save down onion
if not r_onion.sismember('onion_up:'+date , domain):
r_onion.sadd('onion_down:'+date , domain)
r_onion.sadd('onion_down_link:'+date , url)
r_onion.hincrby('onion_link_down', url, 1)
if not r_onion.exists('onion_metadata:{}'.format(domain)):
r_onion.hset('onion_metadata:{}'.format(domain), 'first_seen', date)
r_onion.hset('onion_metadata:{}'.format(domain), 'last_seen', date)
if not r_onion.sismember('{}_up:{}'.format(type_hidden_service, date), domain):
r_onion.sadd('{}_down:{}'.format(type_hidden_service, date), domain)
r_onion.sadd('{}_down_link:{}'.format(type_hidden_service, date), url)
r_onion.hincrby('{}_link_down'.format(type_hidden_service), url, 1)
if not r_onion.exists('{}_metadata:{}'.format(type_hidden_service, domain)):
r_onion.hset('{}_metadata:{}'.format(type_hidden_service, domain), 'first_seen', date)
r_onion.hset('{}_metadata:{}'.format(type_hidden_service,domain), 'last_seen', date)
else:
r_onion.hincrby('onion_link_up', url, 1)
r_onion.hincrby('{}_link_up'.format(type_hidden_service), url, 1)
# last check
r_onion.hset('onion_metadata:{}'.format(domain), 'last_check', date)
r_onion.hset('{}_metadata:{}'.format(type_hidden_service, domain), 'last_check', date)
# check external onions links (full_scrawl)
external_domains = set()
for link in r_onion.smembers('domain_onion_external_links:{}'.format(domain)):
external_domain = re.findall(url_regex, link)
for link in r_onion.smembers('domain_{}_external_links:{}'.format(type_hidden_service, domain)):
external_domain = re.findall(url_onion, link)
external_domain.extend(re.findall(url_i2p, link))
if len(external_domain) > 0:
external_domain = external_domain[0][4]
else:
continue
# # TODO: add i2p
if '.onion' in external_domain and external_domain != domain:
external_domains.add(external_domain)
elif '.i2p' in external_domain and external_domain != domain:
external_domains.add(external_domain)
if len(external_domains) >= 10:
r_onion.sadd('onion_potential_source', domain)
r_onion.delete('domain_onion_external_links:{}'.format(domain))
print(r_onion.smembers('domain_onion_external_links:{}'.format(domain)))
r_onion.sadd('{}_potential_source'.format(type_hidden_service), domain)
r_onion.delete('domain_{}_external_links:{}'.format(type_hidden_service, domain))
print(r_onion.smembers('domain_{}_external_links:{}'.format(type_hidden_service, domain)))
r_onion.lpush('last_onions', domain)
r_onion.ltrim('last_onions', 0, 15)
r_onion.lpush('last_{}'.format(type_hidden_service), domain)
r_onion.ltrim('last_{}'.format(type_hidden_service), 0, 15)
else:
continue

View file

@ -150,9 +150,12 @@ if __name__ == "__main__":
if '.i2p' in url:
print('add i2p')
print(domain)
if not r_onion.sismember('i2p_domain', domain):
if not r_onion.sismember('i2p_domain', domain) and not r_onion.sismember('i2p_domain_crawler_queue', domain):
r_onion.sadd('i2p_domain', domain)
r_onion.sadd('i2p_link', url)
r_onion.sadd('i2p_domain_crawler_queue', domain)
msg = '{};{}'.format(url,PST.p_path)
r_onion.sadd('i2p_crawler_queue', msg)
# Saving the list of extracted onion domains.
PST.__setattr__(channel, domains_list)
@ -193,9 +196,12 @@ if __name__ == "__main__":
continue
if not r_onion.sismember('month_onion_up:{}'.format(date_month), domain) and not r_onion.sismember('onion_down:'+date , domain):
msg = '{};{}'.format(url,PST.p_path)
print('send to crawler')
p.populate_set_out(msg, 'Crawler')
if not r_onion.sismember('onion_domain_crawler_queue', domain):
print('send to onion crawler')
r_onion.sadd('onion_domain_crawler_queue', domain)
msg = '{};{}'.format(url,PST.p_path)
r_onion.sadd('onion_crawler_queue', msg)
#p.populate_set_out(msg, 'Crawler')
else:
publisher.info('{}Onion related;{}'.format(to_print, PST.p_path))

View file

@ -9,6 +9,7 @@ import uuid
import datetime
import base64
import redis
import json
from scrapy.spidermiddlewares.httperror import HttpError
from twisted.internet.error import DNSLookupError
@ -30,7 +31,6 @@ class TorSplashCrawler():
self.crawler = Crawler(self.TorSplashSpider, {
'USER_AGENT': 'Mozilla/5.0 (Windows NT 6.1; rv:24.0) Gecko/20100101 Firefox/24.0',
'SPLASH_URL': splash_url,
'HTTP_PROXY': http_proxy,
'ROBOTSTXT_OBEY': False,
'DOWNLOADER_MIDDLEWARES': {'scrapy_splash.SplashCookiesMiddleware': 723,
'scrapy_splash.SplashMiddleware': 725,
@ -41,14 +41,15 @@ class TorSplashCrawler():
'DEPTH_LIMIT': crawler_depth_limit
})
def crawl(self, url, domain, original_paste, super_father):
self.process.crawl(self.crawler, url=url, domain=domain,original_paste=original_paste, super_father=super_father)
def crawl(self, type, url, domain, original_paste, super_father):
self.process.crawl(self.crawler, type=type, url=url, domain=domain,original_paste=original_paste, super_father=super_father)
self.process.start()
class TorSplashSpider(Spider):
name = 'TorSplashSpider'
def __init__(self, url, domain,original_paste, super_father, *args, **kwargs):
def __init__(self, type, url, domain,original_paste, super_father, *args, **kwargs):
self.type = type
self.original_paste = original_paste
self.super_father = super_father
self.start_urls = url
@ -100,12 +101,13 @@ class TorSplashCrawler():
args={ 'html': 1,
'wait': 10,
'render_all': 1,
'har': 1,
'png': 1}
)
def parse(self,response):
print(response.headers)
print(response.status)
#print(response.headers)
#print(response.status)
# # TODO: # FIXME:
self.r_cache.setbit(response.url, 0, 1)
@ -119,17 +121,18 @@ class TorSplashCrawler():
# save new paste on disk
if self.save_crawled_paste(filename_paste, response.data['html']):
self.r_serv_onion.sadd('onion_up:'+self.full_date , self.domains[0])
self.r_serv_onion.sadd('full_onion_up', self.domains[0])
self.r_serv_onion.sadd('month_onion_up:{}'.format(self.date_month), self.domains[0])
self.r_serv_onion.sadd('{}_up:{}'.format(self.type, self.full_date), self.domains[0])
self.r_serv_onion.sadd('full_{}_up'.format(self.type), self.domains[0])
self.r_serv_onion.sadd('month_{}_up:{}'.format(self.type, self.date_month), self.domains[0])
# create onion metadata
if not self.r_serv_onion.exists('onion_metadata:{}'.format(self.domains[0])):
self.r_serv_onion.hset('onion_metadata:{}'.format(self.domains[0]), 'first_seen', self.full_date)
self.r_serv_onion.hset('onion_metadata:{}'.format(self.domains[0]), 'last_seen', self.full_date)
if not self.r_serv_onion.exists('{}_metadata:{}'.format(self.type, self.domains[0])):
self.r_serv_onion.hset('{}_metadata:{}'.format(self.type, self.domains[0]), 'first_seen', self.full_date)
self.r_serv_onion.hset('{}_metadata:{}'.format(self.type, self.domains[0]), 'last_seen', self.full_date)
self.r_serv_onion.hset('{}_metadata:{}'.format(self.type, self.domains[0]), 'paste_parent', self.original_paste)
# add onion screenshot history
self.r_serv_onion.sadd('onion_history:{}'.format(self.domains[0]), self.full_date)
self.r_serv_onion.sadd('{}_history:{}'.format(self.type, self.domains[0]), self.full_date)
#create paste metadata
self.r_serv_metadata.hset('paste_metadata:'+filename_paste, 'super_father', self.super_father)
@ -144,17 +147,20 @@ class TorSplashCrawler():
os.makedirs(dirname)
size_screenshot = (len(response.data['png'])*3) /4
print(size_screenshot)
if size_screenshot < 5000000: #bytes
with open(filename_screenshot, 'wb') as f:
f.write(base64.standard_b64decode(response.data['png'].encode()))
#interest = response.data['har']['log']['entries'][0]['response']['header'][0]
with open(filename_screenshot+'har.txt', 'wb') as f:
f.write(json.dumps(response.data['har']).encode())
# save external links in set
lext = LinkExtractor(deny_domains=self.domains, unique=True)
for link in lext.extract_links(response):
self.r_serv_onion.sadd('domain_onion_external_links:{}'.format(self.domains[0]), link.url)
self.r_serv_metadata.sadd('paste_onion_external_links:{}'.format(filename_paste), link.url)
self.r_serv_onion.sadd('domain_{}_external_links:{}'.format(self.type, self.domains[0]), link.url)
self.r_serv_metadata.sadd('paste_{}_external_links:{}'.format(self.type, filename_paste), link.url)
#le = LinkExtractor(unique=True)
le = LinkExtractor(allow_domains=self.domains, unique=True)
@ -169,6 +175,7 @@ class TorSplashCrawler():
args={ 'html': 1,
'png': 1,
'render_all': 1,
'har': 1,
'wait': 10}
#errback=self.errback_catcher
)

View file

@ -8,8 +8,8 @@ from TorSplashCrawler import TorSplashCrawler
if __name__ == '__main__':
if len(sys.argv) != 5:
print('usage:', 'tor_crawler.py', 'url', 'domain', 'paste', 'super_father')
if len(sys.argv) != 8:
print('usage:', 'tor_crawler.py', 'splash_url', 'http_proxy', 'type', 'url', 'domain', 'paste', 'super_father')
exit(1)
configfile = os.path.join(os.environ['AIL_BIN'], 'packages/config.cfg')
@ -21,14 +21,15 @@ if __name__ == '__main__':
cfg = configparser.ConfigParser()
cfg.read(configfile)
splash_url = cfg.get("Crawler", "splash_url")
http_proxy = cfg.get("Crawler", "http_proxy")
splash_url = sys.argv[1]
http_proxy = sys.argv[2]
type = sys.argv[3]
crawler_depth_limit = cfg.getint("Crawler", "crawler_depth_limit")
url = sys.argv[1]
domain = sys.argv[2]
paste = sys.argv[3]
super_father = sys.argv[4]
url = sys.argv[4]
domain = sys.argv[5]
paste = sys.argv[6]
super_father = sys.argv[7]
crawler = TorSplashCrawler(splash_url, http_proxy, crawler_depth_limit)
crawler.crawl(url, domain, paste, super_father)
crawler.crawl(type, url, domain, paste, super_father)

View file

@ -43,7 +43,7 @@ def get_onion_status(domain, date):
@hiddenServices.route("/hiddenServices/", methods=['GET'])
def hiddenServices_page():
last_onions = r_serv_onion.lrange('last_onions', 0 ,-1)
last_onions = r_serv_onion.lrange('last_onion', 0 ,-1)
list_onion = []
for onion in last_onions:
@ -72,9 +72,11 @@ def onion_domain():
last_check = r_serv_onion.hget('onion_metadata:{}'.format(onion_domain), 'last_check')
first_seen = r_serv_onion.hget('onion_metadata:{}'.format(onion_domain), 'first_seen')
domain_paste = r_serv_onion.hget('onion_metadata:{}'.format(onion_domain), 'paste_parent')
date_crawled = r_serv_onion.smembers('onion_history:{}'.format(onion_domain))
return render_template("showDomain.html", domain=onion_domain, last_check=last_check, first_seen=first_seen)
return render_template("showDomain.html", domain=onion_domain, last_check=last_check, first_seen=first_seen,
domain_paste=domain_paste)
# ============= JSON ==============
@hiddenServices.route("/hiddenServices/domain_crawled_7days_json", methods=['GET'])

View file

@ -49,6 +49,12 @@
<td>Last Check</td>
<td>{{ last_check }}</td>
</tr>
<tr>
<td>Origin Paste</td>
<td>
<a target="_blank" href="{{ url_for('showsavedpastes.showsavedpaste', paste=domain_paste) }}" />{{ domain_paste }}</a>
</td>
</tr>
</tbody>
</table>
</div>