def parse_url_list(file_path, start, stop):
"""Return list of urls from a file."""
try:
with open(file_path) as f:
# read file contents and split into elements
file_contents = f.read()
url_list = file_contents.splitlines()
url_list = [url for url in url_list if url and not url.startswith('#')]
url_list = url_list[start - 1:stop]
# process urls and tokens into valid youtube video strings
processed_list = []
for url in url_list:
parsed = urlparse(url)
# if no hostname, assume string is the video ID
if not parsed.hostname:
if not youtube_id_validation(url):
raise ValueError('Token is not a valid youtube video ID: {}'.format(url))
url = 'https://www.youtube.com/watch?v=' + url
else:
token = youtube_url_validation(url)
if not token:
raise ValueError('Not a valid youtube URL: {}'.format(url))
url = 'https://www.youtube.com/watch?v=' + token
# stop autoplay and related videos
# youtube may ignore these parameters
url += '&autoplay=0&rel=0'
processed_list.append(url)
except Exception as e:
wl_log.error("while parsing URL list: {} \n{}".format(e, traceback.format_exc()))
sys.exit(-1)
return processed_list
def post_visit(self):
guard_ips = set([ip for ip in self.controller.get_all_guard_ips()])
wl_log.debug("Found %s guards in the consensus.", len(guard_ips))
wl_log.info("Filtering packets without a guard IP.")
try:
ut.filter_pcap(self.job.pcap_file, guard_ips)
except Exception as e:
wl_log.error("ERROR: filtering pcap file: %s.", e)
wl_log.error("Check pcap: %s", self.job.pcap_file)
def post_visit(self):
guard_ips = set([ip for ip in self.controller.get_all_guard_ips()])
#hard-coded bridge ips, used when we set up our own bridges
guard_ips.update(cm.My_Bridge_Ips)
wl_log.debug("Found %s guards in the consensus.", len(guard_ips))
wl_log.info("Filtering packets without a guard IP.")
try:
ut.filter_pcap(self.job.pcap_file, guard_ips)
except Exception as e:
wl_log.error("ERROR: filtering pcap file: %s.", e)
wl_log.error("Check pcap: %s", self.job.pcap_file)
def _do_instance(self):
for self.job.visit in range(self.job.visits):
ut.create_dir(self.job.path)
wl_log.info("*** Visit #%s to %s ***", self.job.visit, self.job.url)
#self.job.screen_num = 0
with self.driver.launch():
try:
self.driver.set_page_load_timeout(cm.SOFT_VISIT_TIMEOUT)
except WebDriverException as seto_exc:
wl_log.error("Setting soft timeout %s", seto_exc)
self._do_visit()
sleep(float(self.job.config['pause_between_loads']))
self.post_visit()
def post_visit(self):
if ('usebridges' in self.controller.torrc_dict) and (
'bridge' in self.controller.torrc_dict):
#Bridge=obfs4 52.175.31.228:35000 cert=Dsk4OM5YMIQsLuqZfOVxaCbsgOxgkX1E3xgPQctB2I1Fy9cs+UvJf4CAwfo0DCGOJrOsIQ iat-mode=0
guard_ips = set([
self.controller.torrc_dict['bridge'].split(" ")[1].split(":")
[0]
])
wl_log.info("Found bridge {}".format(guard_ips))
else:
guard_ips = set([ip for ip in self.controller.get_all_guard_ips()])
wl_log.info("Found %s guards in the consensus.", len(guard_ips))
wl_log.info("Filtering packets without a guard IP.")
try:
ut.filter_pcap(self.job.pcap_file, guard_ips)
except Exception as e:
wl_log.error("ERROR: filtering pcap file: %s.", e)
wl_log.error("Check pcap: %s", self.job.pcap_file)
def parse_url_list(file_path, start, stop):
"""Return list of urls from a file."""
try:
with open(file_path) as f:
# read file contents and split into elements
file_contents = f.read()
url_list = file_contents.splitlines()
url_list = [url for url in url_list if url and not url.startswith('#')]
url_list = url_list[start - 1:stop]
processed_list = []
for url in url_list:
parsed = urlparse(url)
if not parsed.hostname:
raise ValueError('URL {} has invalid hostname!'.format(url))
processed_list.append(url)
except Exception as e:
wl_log.error("while parsing URL list: {} \n{}".format(e, traceback.format_exc()))
sys.exit(-1)
return processed_list
def _do_visit(self):
with Sniffer(path=self.job.pcap_file,
filter=cm.DEFAULT_FILTER,
device=self.device,
dumpcap_log=self.job.pcap_log):
sleep(1) # make sure dumpcap is running
try:
screenshot_count = 0
with ut.timeout(cm.HARD_VISIT_TIMEOUT):
# begin loading page
self.driver.get(self.job.url)
sleep(1) # sleep to catch some lingering AJAX-type traffic
# take first screenshot
if self.screenshots:
try:
self.driver.get_screenshot_as_file(
self.job.png_file(screenshot_count))
screenshot_count += 1
except WebDriverException:
wl_log.error("Cannot get screenshot.")
except (cm.HardTimeoutException, TimeoutException):
wl_log.error("Visit to %s reached hard timeout!", self.job.url)
except Exception as exc:
wl_log.error("Unknown exception: %s", exc)
def _do_visit(self):
with Sniffer(path=self.job.pcap_file,
filter=cm.DEFAULT_FILTER,
device=self.device,
dumpcap_log=self.job.pcap_log):
sleep(1) # make sure dumpcap is running
try:
screenshot_count = 0
with ut.timeout(cm.HARD_VISIT_TIMEOUT):
# begin loading page
self.driver.get(self.job.url)
# take first screenshot
if self.screenshots:
try:
self.driver.get_screenshot_as_file(
self.job.png_file(screenshot_count))
screenshot_count += 1
except WebDriverException:
wl_log.error("Cannot get screenshot.")
# check video player status
status_to_string = [
'ended', 'played', 'paused', 'buffered', 'queued',
'unstarted'
]
js = "return document.getElementById('movie_player').getPlayerState()"
player_status = self.driver.execute_script(js)
# continue visit capture until video is has fully played
ts = time()
while player_status != 0:
# attempt to simulate user skipping add
if player_status == -1:
try:
skipAds = self.driver.find_elements(
By.XPATH,
"//button[@class=\"ytp-ad-skip-button ytp-button\"]"
)
wl_log.info(len(skipAds))
for skipAd in skipAds:
skipAd.click()
except WebDriverException as e:
pass
# unpause video if state is unstarted or is for some reason paused
if player_status == -1 or player_status == 2:
self.driver.execute_script(
"return document.getElementById('movie_player').playVideo()"
)
# busy loop delay
sleep(1)
# check video state again
new_ps = self.driver.execute_script(js)
# print progress updates every time the video state changes
# or on the screenshot interval
ts_new = time()
if player_status != new_ps or ts_new - ts > cm.SCREENSHOT_INTERVAL:
wl_log.debug(
'youtube status: {} for {:.2f} seconds'.format(
status_to_string[player_status],
ts_new - ts))
ts = ts_new
# take periodic screenshots
if self.screenshots:
try:
self.driver.get_screenshot_as_file(
self.job.png_file(screenshot_count))
screenshot_count += 1
except WebDriverException:
wl_log.error("Cannot get screenshot.")
player_status = new_ps
except (cm.HardTimeoutException, TimeoutException):
wl_log.error("Visit to %s reached hard timeout!", self.job.url)
except Exception as exc:
wl_log.error("Unknown exception: %s", exc)
