def _discover_page_link_data(self, url, html_body):
#Prepare to store any url parameters present in links on the page
self.url_data[url]['urlparams'] = set()
#We're also storing unique links acessible from the given page
self.url_data[url]['accessible_links'] = set()
all_links = set(
filter(lambda url: validate_url(url, self.source_url),
html_body.xpath("//a/@href")))
for link in all_links:
absolute_link = self._generate_absolute_link(link)
#we want our accessible links to be links without url parameters
self.url_data[url]['accessible_links'].add(absolute_link)
#get the url parameters from the url and store them in the data
#structure
urlparams = get_url_params(absolute_link)
self.url_data[url]['urlparams'].update(urlparams)
all_links = set(
map(lambda url: trim_url_params(self._generate_absolute_link(url)),
all_links))
for link in (all_links - self.discovered_urls):
self.discovered_urls.add(link)
self.urlqueue.append(link)
def remove_arc_map_service_from_uri(self, uri):
if validate_url(uri):
ams = ([a for a in self._arc_map_servers if a.uri == uri] +
[None])[0]
if ams is not None:
self.remove_arc_map_server(ams)
else:
warn("not a valid Url: {}".format(uri))
def shorten_url():
body = request.get_json()
if not isinstance(body, dict) or 'url' not in body:
return {'ok': False, 'reason': 'Invalid body'}, 400
if not validate_url(body['url']):
return {'ok': False, 'reason': 'Malformed URL'}, 400
suffix = rand_str(5)
short_url = '{proto}://{base}/go/{suffix}'.format(proto=PROTO,
base=BASE,
suffix=suffix)
_redis.set(suffix, body['url'])
return {'ok': True, 'url': short_url}
def runTestTwo(self):
self.prt("\n>>> STARTING TEST TWO - Dead links checker")
self.prt(f"[{TARGET_DOMAIN}] Parsing href, src links", end="\r")
self.scraped_elements_for_url[
TARGET_DOMAIN] = self.parse_all_page_elements(
TARGET_DOMAIN, '[href],[src]')
if not self.scraped_elements_for_url[TARGET_DOMAIN]:
self.prt(f"[ERROR] Could not parse sublinks. Moving forward")
links = []
for a in self.scraped_elements_for_url[TARGET_DOMAIN]:
link = {}
try:
link['raw'] = a['attributes']['href']
except:
try:
link['raw'] = a['attributes']['src']
# Check if link comes from Base64 encoded string
if 'data:' in link['raw']:
continue
except:
continue
# Check if link is not an URL. It may be a route!
if parse_domain(link['raw']) != parse_domain(TARGET_DOMAIN):
# They're not the same domain. This link may be a route to the same website, let's check
parsed = validate_url(link['raw'])
# if True is returned, then it's fine. Otherwise, it may be a route, so let's append it to the TARGET DOMAIN
if parsed != True:
if parsed[0:2] == "//":
link['formatted'] = "https:" + parsed
else:
link['formatted'] = parsed.replace('//', '/')
link['formatted'] = TARGET_DOMAIN + \
"/" + link['formatted']
else:
continue
else:
link['formatted'] = link['raw']
links.append(link)
# Ignore links containing pre-configured keywords
ignore_current = False
for i in range(0, len(TEST_TWO_IGNORE_KEYWORDS)):
if TEST_TWO_IGNORE_KEYWORDS[i] in link['raw']:
ignore_current = True
break
if ignore_current:
continue
total_links_tested = 0
errors = []
total_links = len(links)
for i in range(0, total_links):
# Iterate over the found elements and perform loading test
self.prt(f"Testing link {i} / {total_links}", end="\r")
if request_success_url(URL=links[i]['formatted']):
total_links_tested += 1
else:
self.dead_links_found.append(links[i]['raw'])
total_errors = len(self.dead_links_found)
if total_errors > 0:
for i in range(0, total_errors):
self.prt(f"[ERROR] DEAD {self.dead_links_found[i]}")
else:
self.prt(f"[OK] Links tested: {total_links_tested}")
def runTestFour(self):
self.prt("\n>>> STARTING TEST FOUR - Insecure Content Links Checker")
self.prt(f"[{TARGET_DOMAIN}] Parsing Webpage Links", end="\r")
links = []
# this test will run on all subpages plus target_domain
# check if links have already been parsed by previous tasks
if self.scraped_elements_for_url[TARGET_DOMAIN]:
found_elements = get_elements_by_selector(TARGET_DOMAIN,
'[href], [src]')
if found_elements is None or len(found_elements) < 1:
self.prt(
f"[ERROR] Could not parse any page links. Moving forward to next test"
)
return
for i in range(0, len(found_elements)):
# Iterate over the found elements and perform loading test
cur = found_elements[i]
link = None
lt = ""
if 'href' in cur['attributes']:
link = cur['attributes']['href']
elif 'src' in cur['attributes']:
link = cur['attributes']['src']
lt = "src"
else:
continue
links.append(link)
else:
links = self.first_page_alive_hrefs + self.first_page_alive_srcs
total_links_tested = 0
errors = []
warnings = []
total_links = len(links)
for i in range(0, total_links):
link = links[i]
# Ignore links containing pre-configured keywords
ignore_current = False
for i in range(0, len(TEST_FOUR_IGNORE_KEYWORDS)):
if TEST_FOUR_IGNORE_KEYWORDS[i] in link:
ignore_current = True
break
if ignore_current:
continue
if parse_domain(link) != parse_domain(TARGET_DOMAIN):
# They're not the same domain. This link may be a route to the same website, let's check
parsed = validate_url(link)
if parsed:
link = TARGET_DOMAIN + f"/{parsed}"
if not "https" in link:
# HTTPS link was not found. Check if it's HTTP, otherwise throw a warning that none of them was found
if 'http' in link:
errors.append(f"[ERROR] FOUND HTTP LINK ({link})")
else:
warnings.append(f"[WARNING] UNRECOGNIZED LINK ({link})")
errors = []
total_links_tested += 1
total_errors = len(errors)
total_warnings = len(warnings)
if total_errors > 0:
for i in range(0, total_errors):
self.prt(errors[i])
if total_warnings > 0:
for i in range(0, total_warnings):
self.prt(warnings[i])
if not total_errors:
self.prt(f"[OK] Links tested: {total_links_tested}")
def runTestThree(self):
self.prt(">>> STARTING TEST THREE ( Sub-pages loading time test )")
page_elements = []
# Check if there are any DOM elements parsed for current page
if TARGET_DOMAIN not in self.scraped_elements_for_url:
self.prt(f"[{TARGET_DOMAIN}] Parsing href links", end="\r")
page_elements = get_elements_by_selector(TARGET_DOMAIN, '[href]')
if page_elements is None or len(page_elements) < 1:
self.prt(
f"[ERROR] Could not parse any page links. Moving forward to next test"
)
return
self.scraped_elements_for_url[TARGET_DOMAIN] = page_elements
else:
page_elements = self.scraped_elements_for_url[TARGET_DOMAIN]
# Get links from DOM elements
links = []
for i in range(0, len(page_elements)):
try:
cur = page_elements[i]
link = None
if 'href' in cur['attributes']:
link = cur['attributes']['href']
else:
continue
except:
continue
links.append(link)
total_links = len(links)
errors = []
for i in range(0, total_links):
link = links[i]
# Ignore DEAD links
must_skip = True if link in self.dead_links_found else False
if must_skip:
print("Skipping from dead link")
continue
# Skip IGNORE keyords for test three
for ignore in TEST_THREE_IGNORE_KEYWORDS:
if ignore in link:
must_skip = True
if must_skip:
print("Skipping from ignore list")
continue
# Validate the link and check if they're on the same domain
if parse_domain(link) != parse_domain(TARGET_DOMAIN):
# They're not the same domain. This link may be a route to the same website, let's check
parsed = validate_url(link)
# if True is returned, then it's fine. Otherwise, it may be a route, so let's append it to the TARGET DOMAIN
if parsed != True:
if parsed[0:2] == "//":
link = "https:" + parsed
else:
link = parsed.replace('//', '/')
link = TARGET_DOMAIN + "/" + link
else:
continue
pretty_print = f"/{parse_path(link)}"
average_load_time = perform_load_test(
link,
5,
onPerformingRetry=lambda retry: self.prt(
f"[{pretty_print}] Performing loading time test {retry} / {TEST_TWO_MAX_RETRIES}",
end="\r"))
if average_load_time <= TEST_TWO_MAX_LOADING_SECONDS:
self.prt(
f"[OK] ({pretty_print}) Average load time: {average_load_time}s"
)
else:
self.prt(
f"[ERROR] ({pretty_print}) Average load time exceeded with: {average_load_time}s"
)
def remove_arc_service_from_uri(self, uri):
if not validate_url(uri):
raise Exception("Uri for arc service is not valid")
arc_s = ([s for s in self.arc_services if s.uri == uri] + [None])[0]
if arc_s is not None:
self.remove_arc_service(arc_s)
def add_arc_service_from_uri(self, uri):
if not validate_url(uri):
raise Exception("Uri for arc service is not valid")
name = uri.split("/")[-1]
arc_s = arc_service.ArcService(uri, name, self, self.db_client)
self.add_arc_service(arc_s)
