Mirror/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2026-03-13 21:53:13 +03:00
Code Issues Packages Projects Releases Wiki Activity

[ie/tiktok] Fix challenge solving (#16223)

Browse Source 
Fix e3f0d8b731

Closes #16199
Authored by: bashonly
This commit is contained in:
bashonly
2026-03-13 00:12:36 -05:00
committed by GitHub
parent 3e36cf9cdb
commit db62e438a1
1 changed files with 21 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
yt_dlp/extractor/tiktok.py
View File

@@ -220,7 +220,7 @@ class TikTokBaseIE(InfoExtractor):
raise ExtractorError('Unable to extract aweme detail info', video_id=aweme_id) raise ExtractorError('Unable to extract aweme detail info', video_id=aweme_id)
return self._parse_aweme_video_app(aweme_detail) return self._parse_aweme_video_app(aweme_detail)
def _solve_challenge_and_set_cookie(self, webpage): def _solve_challenge_and_set_cookies(self, webpage):
challenge_data = traverse_obj(webpage, ( challenge_data = traverse_obj(webpage, (
{find_element(id='cs', html=True)}, {extract_attributes}, 'class', {find_element(id='cs', html=True)}, {extract_attributes}, 'class',
filter, {lambda x: f'{x}==='}, {base64.b64decode}, {json.loads})) filter, {lambda x: f'{x}==='}, {base64.b64decode}, {json.loads}))
@@ -250,17 +250,27 @@ class TikTokBaseIE(InfoExtractor):
else: else:
raise ExtractorError('Unable to solve JS challenge') raise ExtractorError('Unable to solve JS challenge')
cookie_value = base64.b64encode( wci_cookie_value = base64.b64encode(
json.dumps(challenge_data, separators=(',', ':')).encode()).decode() json.dumps(challenge_data, separators=(',', ':')).encode()).decode()
# At time of writing, the cookie name was _wafchallengeid # At time of writing, the wci cookie name was `_wafchallengeid`
cookie_name = traverse_obj(webpage, ( wci_cookie_name = traverse_obj(webpage, (
{find_element(id='wci', html=True)}, {extract_attributes}, {find_element(id='wci', html=True)}, {extract_attributes},
'class', {require('challenge cookie name')})) 'class', {require('challenge cookie name')}))
# Actual JS sets Max-Age=1, but we need to adjust for --sleep-requests and Python slowness # At time of writing, the **optional** rci cookie name was `waforiginalreid`
expire_time = int(time.time()) + (self.get_param('sleep_interval_requests') or 0) + 2 rci_cookie_name = traverse_obj(webpage, (
self._set_cookie('.tiktok.com', cookie_name, cookie_value, expire_time=expire_time) {find_element(id='rci', html=True)}, {extract_attributes}, 'class'))
rci_cookie_value = traverse_obj(webpage, (
{find_element(id='rs', html=True)}, {extract_attributes}, 'class'))
# Actual JS sets Max-Age=1 for the cookies, but we'll manually clear them later instead
expire_time = int(time.time()) + (self.get_param('sleep_interval_requests') or 0) + 120
self._set_cookie('.tiktok.com', wci_cookie_name, wci_cookie_value, expire_time=expire_time)
if rci_cookie_name and rci_cookie_value:
self._set_cookie('.tiktok.com', rci_cookie_name, rci_cookie_value, expire_time=expire_time)
return wci_cookie_name, rci_cookie_name
def _extract_web_data_and_status(self, url, video_id, fatal=True): def _extract_web_data_and_status(self, url, video_id, fatal=True):
video_data, status = {}, -1 video_data, status = {}, -1
@@ -287,7 +297,7 @@ class TikTokBaseIE(InfoExtractor):
universal_data = self._get_universal_data(webpage, video_id) universal_data = self._get_universal_data(webpage, video_id)
if not universal_data: if not universal_data:
try: try:
self._solve_challenge_and_set_cookie(webpage) cookie_names = self._solve_challenge_and_set_cookies(webpage)
except ExtractorError as e: except ExtractorError as e:
if fatal: if fatal:
raise raise
@@ -295,6 +305,9 @@ class TikTokBaseIE(InfoExtractor):
return video_data, status return video_data, status
webpage = get_webpage(note='Downloading webpage with challenge cookie') webpage = get_webpage(note='Downloading webpage with challenge cookie')
# Manually clear challenge cookies that should expire immediately after webpage request
for cookie_name in filter(None, cookie_names):
self.cookiejar.clear(domain='.tiktok.com', path='/', name=cookie_name)
if webpage is False: if webpage is False:
return video_data, status return video_data, status
universal_data = self._get_universal_data(webpage, video_id) universal_data = self._get_universal_data(webpage, video_id)