mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-12-16 04:58:56 +00:00
Code Issues Projects Releases Wiki Activity

Merge branch 'yt-dlp:master' into pr/live-sections

Browse Source
This commit is contained in:
bashonly
2024-12-06 17:37:01 -06:00
parent 670bafe148 6fef824025
commit 03eee53919
101 changed files with 2851 additions and 2084 deletions
Download Patch File Download Diff File Expand all files Collapse all files

446
yt_dlp/extractor/youtube.py
View File

@@ -22,7 +22,7 @@ import urllib.parse
from .common import InfoExtractor, SearchInfoExtractor
from .openload import PhantomJSwrapper
from ..jsinterp import JSInterpreter
from ..networking.exceptions import HTTPError, TransportError, network_exceptions
from ..networking.exceptions import HTTPError, network_exceptions
from ..utils import (
NO_DEFAULT,
ExtractorError,
@@ -50,12 +50,12 @@ from ..utils import (
parse_iso8601,
parse_qs,
qualities,
remove_end,
remove_start,
smuggle_url,
str_or_none,
str_to_int,
strftime_or_none,
time_seconds,
traverse_obj,
try_call,
try_get,
@@ -78,60 +78,66 @@ INNERTUBE_CLIENTS = {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB',
'clientVersion': '2.20240726.00.00',
'clientVersion': '2.20241126.01.00',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
'REQUIRE_PO_TOKEN': True,
'SUPPORTS_COOKIES': True,
},
# Safari UA returns pre-merged video+audio 144p/240p/360p/720p/1080p HLS formats
'web_safari': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB',
'clientVersion': '2.20240726.00.00',
'clientVersion': '2.20241126.01.00',
'userAgent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/15.5 Safari/605.1.15,gzip(gfe)',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
'REQUIRE_PO_TOKEN': True,
'SUPPORTS_COOKIES': True,
},
'web_embedded': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_EMBEDDED_PLAYER',
'clientVersion': '1.20240723.01.00',
'clientVersion': '1.20241201.00.00',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 56,
'SUPPORTS_COOKIES': True,
},
'web_music': {
'INNERTUBE_HOST': 'music.youtube.com',
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_REMIX',
'clientVersion': '1.20240724.00.00',
'clientVersion': '1.20241127.01.00',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
'SUPPORTS_COOKIES': True,
},
# This client now requires sign-in for every video
'web_creator': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'WEB_CREATOR',
'clientVersion': '1.20240723.03.00',
'clientVersion': '1.20241203.01.00',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 62,
'REQUIRE_AUTH': True,
'SUPPORTS_COOKIES': True,
},
'android': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID',
'clientVersion': '19.29.37',
'clientVersion': '19.44.38',
'androidSdkVersion': 30,
'userAgent': 'com.google.android.youtube/19.29.37 (Linux; U; Android 11) gzip',
'userAgent': 'com.google.android.youtube/19.44.38 (Linux; U; Android 11) gzip',
'osName': 'Android',
'osVersion': '11',
},
@@ -140,13 +146,14 @@ INNERTUBE_CLIENTS = {
'REQUIRE_JS_PLAYER': False,
'REQUIRE_PO_TOKEN': True,
},
# This client now requires sign-in for every video
'android_music': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_MUSIC',
'clientVersion': '7.11.50',
'clientVersion': '7.27.52',
'androidSdkVersion': 30,
'userAgent': 'com.google.android.apps.youtube.music/7.11.50 (Linux; U; Android 11) gzip',
'userAgent': 'com.google.android.apps.youtube.music/7.27.52 (Linux; U; Android 11) gzip',
'osName': 'Android',
'osVersion': '11',
},
@@ -154,15 +161,17 @@ INNERTUBE_CLIENTS = {
'INNERTUBE_CONTEXT_CLIENT_NAME': 21,
'REQUIRE_JS_PLAYER': False,
'REQUIRE_PO_TOKEN': True,
'REQUIRE_AUTH': True,
'SUPPORTS_COOKIES': True,
},
# This client now requires sign-in for every video
'android_creator': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_CREATOR',
'clientVersion': '24.30.100',
'clientVersion': '24.45.100',
'androidSdkVersion': 30,
'userAgent': 'com.google.android.apps.youtube.creator/24.30.100 (Linux; U; Android 11) gzip',
'userAgent': 'com.google.android.apps.youtube.creator/24.45.100 (Linux; U; Android 11) gzip',
'osName': 'Android',
'osVersion': '11',
},
@@ -170,38 +179,25 @@ INNERTUBE_CLIENTS = {
'INNERTUBE_CONTEXT_CLIENT_NAME': 14,
'REQUIRE_JS_PLAYER': False,
'REQUIRE_PO_TOKEN': True,
'REQUIRE_AUTH': True,
},
# YouTube Kids videos aren't returned on this client for some reason
'android_vr': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_VR',
'clientVersion': '1.57.29',
'clientVersion': '1.60.19',
'deviceMake': 'Oculus',
'deviceModel': 'Quest 3',
'androidSdkVersion': 32,
'userAgent': 'com.google.android.apps.youtube.vr.oculus/1.57.29 (Linux; U; Android 12L; eureka-user Build/SQ3A.220605.009.A1) gzip',
'userAgent': 'com.google.android.apps.youtube.vr.oculus/1.60.19 (Linux; U; Android 12L; eureka-user Build/SQ3A.220605.009.A1) gzip',
'osName': 'Android',
'osVersion': '12L',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 28,
'REQUIRE_JS_PLAYER': False,
},
'android_testsuite': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'ANDROID_TESTSUITE',
'clientVersion': '1.9',
'androidSdkVersion': 30,
'userAgent': 'com.google.android.youtube/1.9 (Linux; U; Android 11) gzip',
'osName': 'Android',
'osVersion': '11',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 30,
'REQUIRE_JS_PLAYER': False,
'PLAYER_PARAMS': '2AMB',
'SUPPORTS_COOKIES': True,
},
# iOS clients have HLS live streams. Setting device model to get 60fps formats.
# See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
@@ -209,47 +205,51 @@ INNERTUBE_CLIENTS = {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS',
'clientVersion': '19.29.1',
'clientVersion': '19.45.4',
'deviceMake': 'Apple',
'deviceModel': 'iPhone16,2',
'userAgent': 'com.google.ios.youtube/19.29.1 (iPhone16,2; U; CPU iOS 17_5_1 like Mac OS X;)',
'userAgent': 'com.google.ios.youtube/19.45.4 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
'osName': 'iPhone',
'osVersion': '17.5.1.21F90',
'osVersion': '18.1.0.22B83',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
'REQUIRE_JS_PLAYER': False,
},
# This client now requires sign-in for every video
'ios_music': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS_MUSIC',
'clientVersion': '7.08.2',
'clientVersion': '7.27.0',
'deviceMake': 'Apple',
'deviceModel': 'iPhone16,2',
'userAgent': 'com.google.ios.youtubemusic/7.08.2 (iPhone16,2; U; CPU iOS 17_5_1 like Mac OS X;)',
'userAgent': 'com.google.ios.youtubemusic/7.27.0 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
'osName': 'iPhone',
'osVersion': '17.5.1.21F90',
'osVersion': '18.1.0.22B83',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
'REQUIRE_JS_PLAYER': False,
'REQUIRE_AUTH': True,
'SUPPORTS_COOKIES': True,
},
# This client now requires sign-in for every video
'ios_creator': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'IOS_CREATOR',
'clientVersion': '24.30.100',
'clientVersion': '24.45.100',
'deviceMake': 'Apple',
'deviceModel': 'iPhone16,2',
'userAgent': 'com.google.ios.ytcreator/24.30.100 (iPhone16,2; U; CPU iOS 17_5_1 like Mac OS X;)',
'userAgent': 'com.google.ios.ytcreator/24.45.100 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
'osName': 'iPhone',
'osVersion': '17.5.1.21F90',
'osVersion': '18.1.0.22B83',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 15,
'REQUIRE_JS_PLAYER': False,
'REQUIRE_AUTH': True,
},
# mweb has 'ultralow' formats
# See: https://github.com/yt-dlp/yt-dlp/pull/557
@@ -257,19 +257,22 @@ INNERTUBE_CLIENTS = {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'MWEB',
'clientVersion': '2.20240726.01.00',
'clientVersion': '2.20241202.07.00',
'userAgent': 'Mozilla/5.0 (iPad; CPU OS 16_7_10 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1,gzip(gfe)',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 2,
'SUPPORTS_COOKIES': True,
},
'tv': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'TVHTML5',
'clientVersion': '7.20240724.13.00',
'clientVersion': '7.20241201.18.00',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 7,
'SUPPORTS_COOKIES': True,
},
# This client now requires sign-in for every video
# It was previously an age-gate workaround for videos that were `playable_in_embed`
@@ -282,17 +285,8 @@ INNERTUBE_CLIENTS = {
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 85,
},
# This client has pre-merged video+audio 720p/1080p streams
'mediaconnect': {
'INNERTUBE_CONTEXT': {
'client': {
'clientName': 'MEDIA_CONNECT_FRONTEND',
'clientVersion': '0.1',
},
},
'INNERTUBE_CONTEXT_CLIENT_NAME': 95,
'REQUIRE_JS_PLAYER': False,
'REQUIRE_AUTH': True,
'SUPPORTS_COOKIES': True,
},
}
@@ -321,6 +315,8 @@ def build_innertube_clients():
ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com')
ytcfg.setdefault('REQUIRE_JS_PLAYER', True)
ytcfg.setdefault('REQUIRE_PO_TOKEN', False)
ytcfg.setdefault('REQUIRE_AUTH', False)
ytcfg.setdefault('SUPPORTS_COOKIES', False)
ytcfg.setdefault('PLAYER_PARAMS', None)
ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en')
@@ -577,208 +573,18 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
self._check_login_required()
def _perform_login(self, username, password):
auth_type, _, user = (username or '').partition('+')
if auth_type != 'oauth':
raise ExtractorError(self._youtube_login_hint, expected=True)
self._initialize_oauth(user, password)
'''
OAuth 2.0 Device Authorization Grant flow, used by the YouTube TV client (youtube.com/tv).
For more information regarding OAuth 2.0 and the Device Authorization Grant flow in general, see:
- https://developers.google.com/identity/protocols/oauth2/limited-input-device
- https://accounts.google.com/.well-known/openid-configuration
- https://www.rfc-editor.org/rfc/rfc8628
- https://www.rfc-editor.org/rfc/rfc6749
Note: The official client appears to use a proxied version of the oauth2 endpoints on youtube.com/o/oauth2,
which applies some modifications to the response (such as returning errors as 200 OK).
Since the client works with the standard API, we will use that as it is well-documented.
'''
_OAUTH_PROFILE = None
_OAUTH_ACCESS_TOKEN_CACHE = {}
_OAUTH_DISPLAY_ID = 'oauth'
# YouTube TV (TVHTML5) client. You can find these at youtube.com/tv
_OAUTH_CLIENT_ID = '861556708454-d6dlm3lh05idd8npek18k6be8ba3oc68.apps.googleusercontent.com'
_OAUTH_CLIENT_SECRET = 'SboVhoG9s0rNafixCSGGKXAT'
_OAUTH_SCOPE = 'http://gdata.youtube.com https://www.googleapis.com/auth/youtube-paid-content'
# From https://accounts.google.com/.well-known/openid-configuration
# Technically, these should be fetched dynamically and not hard-coded.
# However, as these endpoints rarely change, we can risk saving an extra request for every invocation.
_OAUTH_DEVICE_AUTHORIZATION_ENDPOINT = 'https://oauth2.googleapis.com/device/code'
_OAUTH_TOKEN_ENDPOINT = 'https://oauth2.googleapis.com/token'
@property
def _oauth_cache_key(self):
return f'oauth_refresh_token_{self._OAUTH_PROFILE}'
def _read_oauth_error_response(self, response):
return traverse_obj(
self._webpage_read_content(response, self._OAUTH_TOKEN_ENDPOINT, self._OAUTH_DISPLAY_ID, fatal=False),
({json.loads}, 'error', {str}))
def _set_oauth_info(self, token_response):
YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE.setdefault(self._OAUTH_PROFILE, {}).update({
'access_token': token_response['access_token'],
'token_type': token_response['token_type'],
'expiry': time_seconds(
seconds=traverse_obj(token_response, ('expires_in', {float_or_none}), default=300) - 10),
})
refresh_token = traverse_obj(token_response, ('refresh_token', {str}))
if refresh_token:
self.cache.store(self._NETRC_MACHINE, self._oauth_cache_key, refresh_token)
YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE[self._OAUTH_PROFILE]['refresh_token'] = refresh_token
def _initialize_oauth(self, user, refresh_token):
self._OAUTH_PROFILE = user or 'default'
if self._OAUTH_PROFILE in YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE:
self.write_debug(f'{self._OAUTH_DISPLAY_ID}: Using cached access token for profile "{self._OAUTH_PROFILE}"')
return
YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE[self._OAUTH_PROFILE] = {}
if refresh_token:
msg = f'{self._OAUTH_DISPLAY_ID}: Using password input as refresh token'
if self.get_param('cachedir') is not False:
msg += ' and caching token to disk; you should supply an empty password next time'
self.to_screen(msg)
self.cache.store(self._NETRC_MACHINE, self._oauth_cache_key, refresh_token)
else:
refresh_token = self.cache.load(self._NETRC_MACHINE, self._oauth_cache_key)
if refresh_token:
YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE[self._OAUTH_PROFILE]['refresh_token'] = refresh_token
try:
token_response = self._refresh_token(refresh_token)
except ExtractorError as e:
error_msg = str(e.orig_msg).replace('Failed to refresh access token: ', '')
self.report_warning(f'{self._OAUTH_DISPLAY_ID}: Failed to refresh access token: {error_msg}')
token_response = self._oauth_authorize
else:
token_response = self._oauth_authorize
self._set_oauth_info(token_response)
self.write_debug(f'{self._OAUTH_DISPLAY_ID}: Logged in using profile "{self._OAUTH_PROFILE}"')
def _refresh_token(self, refresh_token):
try:
token_response = self._download_json(
self._OAUTH_TOKEN_ENDPOINT,
video_id=self._OAUTH_DISPLAY_ID,
note='Refreshing access token',
data=json.dumps({
'client_id': self._OAUTH_CLIENT_ID,
'client_secret': self._OAUTH_CLIENT_SECRET,
'refresh_token': refresh_token,
'grant_type': 'refresh_token',
}).encode(),
headers={'Content-Type': 'application/json'})
except ExtractorError as e:
if isinstance(e.cause, HTTPError):
error = self._read_oauth_error_response(e.cause.response)
if error == 'invalid_grant':
# RFC6749 § 5.2
raise ExtractorError(
'Failed to refresh access token: Refresh token is invalid, revoked, or expired (invalid_grant)',
expected=True, video_id=self._OAUTH_DISPLAY_ID)
raise ExtractorError(
f'Failed to refresh access token: Authorization server returned error {error}',
video_id=self._OAUTH_DISPLAY_ID)
raise
return token_response
@property
def _oauth_authorize(self):
code_response = self._download_json(
self._OAUTH_DEVICE_AUTHORIZATION_ENDPOINT,
video_id=self._OAUTH_DISPLAY_ID,
note='Initializing authorization flow',
data=json.dumps({
'client_id': self._OAUTH_CLIENT_ID,
'scope': self._OAUTH_SCOPE,
}).encode(),
headers={'Content-Type': 'application/json'})
verification_url = traverse_obj(code_response, ('verification_url', {str}))
user_code = traverse_obj(code_response, ('user_code', {str}))
if not verification_url or not user_code:
if username.startswith('oauth'):
raise ExtractorError(
'Authorization server did not provide verification_url or user_code', video_id=self._OAUTH_DISPLAY_ID)
f'Login with OAuth is no longer supported. {self._youtube_login_hint}', expected=True)
# note: The whitespace is intentional
self.to_screen(
f'{self._OAUTH_DISPLAY_ID}: To give yt-dlp access to your account, '
f'go to {verification_url} and enter code {user_code}')
# RFC8628 § 3.5: default poll interval is 5 seconds if not provided
poll_interval = traverse_obj(code_response, ('interval', {int}), default=5)
for retry in self.RetryManager():
while True:
try:
token_response = self._download_json(
self._OAUTH_TOKEN_ENDPOINT,
video_id=self._OAUTH_DISPLAY_ID,
note=False,
errnote='Failed to request access token',
data=json.dumps({
'client_id': self._OAUTH_CLIENT_ID,
'client_secret': self._OAUTH_CLIENT_SECRET,
'device_code': code_response['device_code'],
'grant_type': 'urn:ietf:params:oauth:grant-type:device_code',
}).encode(),
headers={'Content-Type': 'application/json'})
except ExtractorError as e:
if isinstance(e.cause, TransportError):
retry.error = e
break
elif isinstance(e.cause, HTTPError):
error = self._read_oauth_error_response(e.cause.response)
if not error:
retry.error = e
break
if error == 'authorization_pending':
time.sleep(poll_interval)
continue
elif error == 'expired_token':
raise ExtractorError(
'Authorization timed out', expected=True, video_id=self._OAUTH_DISPLAY_ID)
elif error == 'access_denied':
raise ExtractorError(
'You denied access to an account', expected=True, video_id=self._OAUTH_DISPLAY_ID)
elif error == 'slow_down':
# RFC8628 § 3.5: add 5 seconds to the poll interval
poll_interval += 5
time.sleep(poll_interval)
continue
else:
raise ExtractorError(
f'Authorization server returned an error when fetching access token: {error}',
video_id=self._OAUTH_DISPLAY_ID)
raise
return token_response
def _update_oauth(self):
token = YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE.get(self._OAUTH_PROFILE)
if token is None or token['expiry'] > time.time():
return
self._set_oauth_info(self._refresh_token(token['refresh_token']))
self.report_warning(
f'Login with password is not supported for YouTube. {self._youtube_login_hint}')
@property
def _youtube_login_hint(self):
return ('Use --username=oauth[+PROFILE] --password="" to log in using oauth, '
f'or else u{self._login_hint(method="cookies")[1:]}. '
'See https://github.com/yt-dlp/yt-dlp/wiki/Extractors#logging-in-with-oauth for more on how to use oauth. '
'See https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies for help with cookies')
return (f'{self._login_hint(method="cookies")}. Also see '
'https://github.com/yt-dlp/yt-dlp/wiki/Extractors#exporting-youtube-cookies '
'for tips on effectively exporting YouTube cookies')
def _check_login_required(self):
if self._LOGIN_REQUIRED and not self.is_authenticated:
@@ -928,7 +734,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
@functools.cached_property
def is_authenticated(self):
return self._OAUTH_PROFILE or bool(self._generate_sapisidhash_header())
return bool(self._generate_sapisidhash_header())
def extract_ytcfg(self, video_id, webpage):
if not webpage:
@@ -938,16 +744,6 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg',
default='{}'), video_id, fatal=False) or {}
def _generate_oauth_headers(self):
self._update_oauth()
oauth_token = YoutubeBaseInfoExtractor._OAUTH_ACCESS_TOKEN_CACHE.get(self._OAUTH_PROFILE)
if not oauth_token:
return {}
return {
'Authorization': f'{oauth_token["token_type"]} {oauth_token["access_token"]}',
}
def _generate_cookie_auth_headers(self, *, ytcfg=None, account_syncid=None, session_index=None, origin=None, **kwargs):
headers = {}
account_syncid = account_syncid or self._extract_account_syncid(ytcfg)
@@ -977,14 +773,10 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
'Origin': origin,
'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg),
'User-Agent': self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT']['client']['userAgent'], default_client=default_client),
**self._generate_oauth_headers(),
**self._generate_cookie_auth_headers(ytcfg=ytcfg, account_syncid=account_syncid, session_index=session_index, origin=origin),
}
return filter_dict(headers)
def _generate_webpage_headers(self):
return self._generate_oauth_headers()
def _download_ytcfg(self, client, video_id):
url = {
'web': 'https://www.youtube.com',
@@ -994,8 +786,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
if not url:
return {}
webpage = self._download_webpage(
url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config',
headers=self._generate_webpage_headers())
url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config')
return self.extract_ytcfg(video_id, webpage) or {}
@staticmethod
@@ -1566,6 +1357,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
}
_SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
_DEFAULT_CLIENTS = ('ios', 'mweb')
_DEFAULT_AUTHED_CLIENTS = ('web_creator', 'mweb')
_GEO_BYPASS = False
@@ -3139,7 +2931,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Obtain from MPD's maximum seq value
old_mpd_url = mpd_url
last_error = ctx.pop('last_error', None)
expire_fast = immediate or last_error and isinstance(last_error, HTTPError) and last_error.status == 403
expire_fast = immediate or (last_error and isinstance(last_error, HTTPError) and last_error.status == 403)
mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000)
or (mpd_url, stream_number, False))
if not refresh_sequence:
@@ -3289,8 +3081,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
code = self._download_webpage(
player_url, video_id, fatal=fatal,
note='Downloading player ' + player_id,
errnote=f'Download of {player_url} failed',
headers=self._generate_webpage_headers())
errnote=f'Download of {player_url} failed')
if code:
self._code_cache[player_id] = code
return self._code_cache.get(player_id)
@@ -3357,19 +3148,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
self.to_screen('Extracted signature function:\n' + code)
def _parse_sig_js(self, jscode):
# Examples where `sig` is funcname:
# sig=function(a){a=a.split(""); ... ;return a.join("")};
# ;c&&(c=sig(decodeURIComponent(c)),a.set(b,encodeURIComponent(c)));return a};
# {var l=f,m=h.sp,n=sig(decodeURIComponent(h.s));l.set(m,encodeURIComponent(n))}
# sig=function(J){J=J.split(""); ... ;return J.join("")};
# ;N&&(N=sig(decodeURIComponent(N)),J.set(R,encodeURIComponent(N)));return J};
# {var H=u,k=f.sp,v=sig(decodeURIComponent(f.s));H.set(k,encodeURIComponent(v))}
funcname = self._search_regex(
(r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
(r'\b(?P<var>[a-zA-Z0-9$]+)&&\((?P=var)=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\((?P=var)\)\)',
r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*(?P<arg>[a-zA-Z0-9$]+)\s*\)\s*{\s*(?P=arg)\s*=\s*(?P=arg)\.split\(\s*""\s*\)\s*;\s*[^}]+;\s*return\s+(?P=arg)\.join\(\s*""\s*\)',
r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\))?',
# Old patterns
r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)',
r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)',
r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)(?:;[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\))?',
r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
# Obsolete patterns
r'("|\')signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
jscode, 'Initial JS player signature function name', group='sig')
@@ -3443,6 +3241,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# * a.D&&(b="nn"[+a.D],c=a.get(b))&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
# * a.D&&(PL(a),b=a.j.n||null)&&(b=narray[0](b),a.set("n",b),narray.length||nfunc("")
# * a.D&&(b="nn"[+a.D],vL(a),c=a.j[b]||null)&&(c=narray[idx](c),a.set(b,c),narray.length||nfunc("")
# * J.J="";J.url="";J.Z&&(R="nn"[+J.Z],mW(J),N=J.K[R]||null)&&(N=narray[idx](N),J.set(R,N))}};
funcname, idx = self._search_regex(
r'''(?x)
(?:
@@ -3459,7 +3258,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
)\)&&\(c=|
\b(?P<var>[a-zA-Z0-9_$]+)=
)(?P<nfunc>[a-zA-Z0-9_$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z]\)
(?(var),[a-zA-Z0-9_$]+\.set\("n"\,(?P=var)\),(?P=nfunc)\.length)''',
(?(var),[a-zA-Z0-9_$]+\.set\((?:"n+"|[a-zA-Z0-9_$]+)\,(?P=var)\))''',
jscode, 'n function name', group=('nfunc', 'idx'), default=(None, None))
if not funcname:
self.report_warning(join_nonempty(
@@ -3468,7 +3267,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return self._search_regex(
r'''(?xs)
;\s*(?P<name>[a-zA-Z0-9_$]+)\s*=\s*function\([a-zA-Z0-9_$]+\)
\s*\{(?:(?!};).)+?["']enhanced_except_''',
\s*\{(?:(?!};).)+?return\s*(?P<q>["'])[\w-]+_w8_(?P=q)\s*\+\s*[a-zA-Z0-9_$]+''',
jscode, 'Initial JS player n function name', group='name')
elif not idx:
return funcname
@@ -3477,6 +3276,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])\s*[,;]', jscode,
f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
def _fixup_n_function_code(self, argnames, code):
return argnames, re.sub(
rf';\s*if\s*\(\s*typeof\s+[a-zA-Z0-9_$]+\s*===?\s*(["\'])undefined\1\s*\)\s*return\s+{argnames[0]};',
';', code)
def _extract_n_function_code(self, video_id, player_url):
player_id = self._extract_player_info(player_url)
func_code = self.cache.load('youtube-nsig', player_id, min_ver='2024.07.09')
@@ -3488,7 +3292,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
func_name = self._extract_n_function_name(jscode, player_url=player_url)
func_code = jsi.extract_function_code(func_name)
# XXX: Workaround for the `typeof` gotcha
func_code = self._fixup_n_function_code(*jsi.extract_function_code(func_name))
self.cache.store('youtube-nsig', player_id, func_code)
return jsi, player_id, func_code
@@ -3504,7 +3309,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
except Exception as e:
raise JSInterpreter.Exception(traceback.format_exc(), cause=e)
if ret.startswith('enhanced_except_'):
if ret.startswith('enhanced_except_') or ret.endswith(s):
raise JSInterpreter.Exception('Signature function returned an exception')
return ret
@@ -3573,8 +3378,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
self._download_webpage(
url, video_id, f'Marking {label}watched',
'Unable to mark watched', fatal=False,
headers=self._generate_webpage_headers())
'Unable to mark watched', fatal=False)
@classmethod
def _extract_from_webpage(cls, url, webpage):
@@ -4063,12 +3867,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _get_requested_clients(self, url, smuggled_data):
requested_clients = []
excluded_clients = []
default_clients = self._DEFAULT_AUTHED_CLIENTS if self.is_authenticated else self._DEFAULT_CLIENTS
allowed_clients = sorted(
(client for client in INNERTUBE_CLIENTS if client[:1] != '_'),
key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
for client in self._configuration_arg('player_client'):
if client == 'default':
requested_clients.extend(self._DEFAULT_CLIENTS)
requested_clients.extend(default_clients)
elif client == 'all':
requested_clients.extend(allowed_clients)
elif client.startswith('-'):
@@ -4078,7 +3883,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
else:
requested_clients.append(client)
if not requested_clients:
requested_clients.extend(self._DEFAULT_CLIENTS)
requested_clients.extend(default_clients)
for excluded_client in excluded_clients:
if excluded_client in requested_clients:
requested_clients.remove(excluded_client)
@@ -4088,9 +3893,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if smuggled_data.get('is_music_url') or self.is_music_url(url):
for requested_client in requested_clients:
_, base_client, variant = _split_innertube_client(requested_client)
music_client = f'{base_client}_music'
music_client = f'{base_client}_music' if base_client != 'mweb' else 'web_music'
if variant != 'music' and music_client in INNERTUBE_CLIENTS:
requested_clients.append(music_client)
client_info = INNERTUBE_CLIENTS[music_client]
if not client_info['REQUIRE_AUTH'] or (self.is_authenticated and client_info['SUPPORTS_COOKIES']):
requested_clients.append(music_client)
if self.is_authenticated:
unsupported_clients = [
client for client in requested_clients if not INNERTUBE_CLIENTS[client]['SUPPORTS_COOKIES']
]
for client in unsupported_clients:
self.report_warning(f'Skipping client "{client}" since it does not support cookies', only_once=True)
requested_clients.remove(client)
return orderedSet(requested_clients)
@@ -4197,16 +4012,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
else:
prs.append(pr)
''' This code is pointless while web_creator is in _DEFAULT_AUTHED_CLIENTS
# EU countries require age-verification for accounts to access age-restricted videos
# If account is not age-verified, _is_agegated() will be truthy for non-embedded clients
if self.is_authenticated and self._is_agegated(pr):
self.to_screen(
f'{video_id}: This video is age-restricted and YouTube is requiring '
'account age-verification; some formats may be missing', only_once=True)
# web_creator and mediaconnect can work around the age-verification requirement
# _testsuite & _vr variants can also work around age-verification
# web_creator can work around the age-verification requirement
# android_vr may also be able to work around age-verification
# tv_embedded may(?) still work around age-verification if the video is embeddable
append_client('web_creator', 'mediaconnect')
append_client('web_creator')
'''
prs.extend(deprioritized_prs)
@@ -4222,8 +4039,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return prs, player_url
def _needs_live_processing(self, live_status, duration):
if (live_status == 'is_live' and self.get_param('live_from_start')
or live_status == 'post_live' and (duration or 0) > 2 * 3600):
if ((live_status == 'is_live' and self.get_param('live_from_start'))
or (live_status == 'post_live' and (duration or 0) > 2 * 3600)):
return live_status
def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
@@ -4422,7 +4239,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
skip_manifests = set(self._configuration_arg('skip'))
if (not self.get_param('youtube_include_hls_manifest', True)
or needs_live_processing == 'is_live' # These will be filtered out by YoutubeDL anyway
or needs_live_processing and skip_bad_formats):
or (needs_live_processing and skip_bad_formats)):
skip_manifests.add('hls')
if not self.get_param('youtube_include_dash_manifest', True):
@@ -4558,7 +4375,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if pp:
query['pp'] = pp
webpage = self._download_webpage(
webpage_url, video_id, fatal=False, query=query, headers=self._generate_webpage_headers())
webpage_url, video_id, fatal=False, query=query)
master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg()
@@ -4620,14 +4437,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
expected_type=dict)
translated_title = self._get_text(microformats, (..., 'title'))
video_title = (self._preferred_lang and translated_title
video_title = ((self._preferred_lang and translated_title)
or get_first(video_details, 'title') # primary
or translated_title
or search_meta(['og:title', 'twitter:title', 'title']))
translated_description = self._get_text(microformats, (..., 'description'))
original_description = get_first(video_details, 'shortDescription')
video_description = (
self._preferred_lang and translated_description
(self._preferred_lang and translated_description)
# If original description is blank, it will be an empty string.
# Do not prefer translated description in this case.
or original_description if original_description is not None else translated_description)
@@ -4701,6 +4518,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
self.raise_geo_restricted(subreason, countries, metadata_available=True)
reason += f'. {subreason}'
if reason:
if 'sign in' in reason.lower():
reason = remove_end(reason, 'This helps protect our community. Learn more')
reason = f'{remove_end(reason.strip(), ".")}. {self._youtube_login_hint}'
self.raise_no_formats(reason, expected=True)
keywords = get_first(video_details, 'keywords', expected_type=list) or []
@@ -5225,6 +5045,10 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
for item in grid_renderer['items']:
if not isinstance(item, dict):
continue
if lockup_view_model := traverse_obj(item, ('lockupViewModel', {dict})):
if entry := self._extract_lockup_view_model(lockup_view_model):
yield entry
continue
renderer = self._extract_basic_item_renderer(item)
if not isinstance(renderer, dict):
continue
@@ -5323,7 +5147,27 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
continue
yield self._extract_video(renderer)
def _extract_lockup_view_model(self, view_model):
content_id = view_model.get('contentId')
if not content_id:
return
content_type = view_model.get('contentType')
if content_type not in ('LOCKUP_CONTENT_TYPE_PLAYLIST', 'LOCKUP_CONTENT_TYPE_PODCAST'):
self.report_warning(
f'Unsupported lockup view model content type "{content_type}"{bug_reports_message()}', only_once=True)
return
return self.url_result(
f'https://www.youtube.com/playlist?list={content_id}', ie=YoutubeTabIE, video_id=content_id,
title=traverse_obj(view_model, (
'metadata', 'lockupMetadataViewModel', 'title', 'content', {str})),
thumbnails=self._extract_thumbnails(view_model, (
'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail', 'thumbnailViewModel', 'image'), final_key='sources'))
def _rich_entries(self, rich_grid_renderer):
if lockup_view_model := traverse_obj(rich_grid_renderer, ('content', 'lockupViewModel', {dict})):
if entry := self._extract_lockup_view_model(lockup_view_model):
yield entry
return
renderer = traverse_obj(
rich_grid_renderer,
('content', ('videoRenderer', 'reelItemRenderer', 'playlistRenderer', 'shortsLockupViewModel'), any)) or {}
@@ -5846,7 +5690,7 @@ class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor):
webpage, data = None, None
for retry in self.RetryManager(fatal=fatal):
try:
webpage = self._download_webpage(url, item_id, note='Downloading webpage', headers=self._generate_webpage_headers())
webpage = self._download_webpage(url, item_id, note='Downloading webpage')
data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {}
except ExtractorError as e:
if isinstance(e.cause, network_exceptions):
@@ -6021,7 +5865,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
'info_dict': {
'id': 'UCYO_jab_esuFRV4b17AJtAw',
'title': '3Blue1Brown - Playlists',
'description': 'md5:4d1da95432004b7ba840ebc895b6b4c9',
'description': 'md5:602e3789e6a0cb7d9d352186b720e395',
'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw',
'channel': '3Blue1Brown',
'channel_id': 'UCYO_jab_esuFRV4b17AJtAw',
@@ -6945,22 +6789,22 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
},
'playlist_count': 0,
}, {
# Podcasts tab, with rich entry playlistRenderers
# Podcasts tab, with rich entry lockupViewModel
'url': 'https://www.youtube.com/@99percentinvisiblepodcast/podcasts',
'info_dict': {
'id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
'channel_id': 'UCVMF2HD4ZgC0QHpU9Yq5Xrw',
'uploader_url': 'https://www.youtube.com/@99percentinvisiblepodcast',
'description': 'md5:3a0ed38f1ad42a68ef0428c04a15695c',
'title': '99 Percent Invisible - Podcasts',
'uploader': '99 Percent Invisible',
'title': '99% Invisible - Podcasts',
'uploader': '99% Invisible',
'channel_follower_count': int,
'channel_url': 'https://www.youtube.com/channel/UCVMF2HD4ZgC0QHpU9Yq5Xrw',
'tags': [],
'channel': '99 Percent Invisible',
'channel': '99% Invisible',
'uploader_id': '@99percentinvisiblepodcast',
},
'playlist_count': 0,
'playlist_count': 5,
}, {
# Releases tab, with rich entry playlistRenderers (same as Podcasts tab)
'url': 'https://www.youtube.com/@AHimitsu/releases',
@@ -7040,7 +6884,7 @@ class YoutubeTabIE(YoutubeTabBaseInfoExtractor):
tab_url = urljoin(base_url, traverse_obj(
tab, ('endpoint', 'commandMetadata', 'webCommandMetadata', 'url')))
tab_id = (tab_url and self._get_url_mobj(tab_url)['tab'][1:]
tab_id = ((tab_url and self._get_url_mobj(tab_url)['tab'][1:])
or traverse_obj(tab, 'tabIdentifier', expected_type=str))
if tab_id:
return {