diff options
| author | Jesús <heckyel@hyperbola.info> | 2021-10-18 15:24:21 -0500 |
|---|---|---|
| committer | Jesús <heckyel@hyperbola.info> | 2021-10-18 15:24:21 -0500 |
| commit | 5122028a4bcac4ae577ef7fbd55ccad5cb34ef5e (patch) | |
| tree | 65209bc739db35e31f1c9b5b868eb5df4fe12ae3 /hypervideo_dl/extractor/youtube.py | |
| parent | 27fe903c511691c078942bef5ee9a05a43b15c8f (diff) | |
| download | hypervideo-5122028a4bcac4ae577ef7fbd55ccad5cb34ef5e.tar.lz hypervideo-5122028a4bcac4ae577ef7fbd55ccad5cb34ef5e.tar.xz hypervideo-5122028a4bcac4ae577ef7fbd55ccad5cb34ef5e.zip | |
update from upstream
Diffstat (limited to 'hypervideo_dl/extractor/youtube.py')
| -rw-r--r-- | hypervideo_dl/extractor/youtube.py | 3600 |
1 files changed, 2501 insertions, 1099 deletions
diff --git a/hypervideo_dl/extractor/youtube.py b/hypervideo_dl/extractor/youtube.py index 2272a02..dc5ee63 100644 --- a/hypervideo_dl/extractor/youtube.py +++ b/hypervideo_dl/extractor/youtube.py @@ -2,11 +2,17 @@ from __future__ import unicode_literals +import base64 +import calendar +import copy +import datetime +import hashlib import itertools import json import os.path import random import re +import time import traceback from .common import InfoExtractor, SearchInfoExtractor @@ -22,231 +28,250 @@ from ..compat import ( ) from ..jsinterp import JSInterpreter from ..utils import ( - ExtractorError, + bytes_to_intlist, clean_html, + datetime_from_str, dict_get, + error_to_compat_str, + ExtractorError, float_or_none, + format_field, int_or_none, + intlist_to_bytes, + is_html, mimetype2ext, + network_exceptions, + orderedSet, parse_codecs, + parse_count, parse_duration, + parse_iso8601, + parse_qs, qualities, + remove_end, remove_start, smuggle_url, str_or_none, str_to_int, + traverse_obj, try_get, unescapeHTML, unified_strdate, unsmuggle_url, update_url_query, url_or_none, - urlencode_postdata, urljoin, + variadic, ) -def parse_qs(url): - return compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) +# any clients starting with _ cannot be explicity requested by the user +INNERTUBE_CLIENTS = { + 'web': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'WEB', + 'clientVersion': '2.20210622.10.00', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 1 + }, + 'web_embedded': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'WEB_EMBEDDED_PLAYER', + 'clientVersion': '1.20210620.0.1', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 56 + }, + 'web_music': { + 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30', + 'INNERTUBE_HOST': 'music.youtube.com', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'WEB_REMIX', + 'clientVersion': '1.20210621.00.00', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 67, + }, + 'web_creator': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'WEB_CREATOR', + 'clientVersion': '1.20210621.00.00', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 62, + }, + 'android': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'ANDROID', + 'clientVersion': '16.20', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 3, + 'REQUIRE_JS_PLAYER': False + }, + 'android_embedded': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'ANDROID_EMBEDDED_PLAYER', + 'clientVersion': '16.20', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 55, + 'REQUIRE_JS_PLAYER': False + }, + 'android_music': { + 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30', + 'INNERTUBE_HOST': 'music.youtube.com', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'ANDROID_MUSIC', + 'clientVersion': '4.32', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 21, + 'REQUIRE_JS_PLAYER': False + }, + 'android_creator': { + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'ANDROID_CREATOR', + 'clientVersion': '21.24.100', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 14, + 'REQUIRE_JS_PLAYER': False + }, + # ios has HLS live streams + # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680 + 'ios': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'IOS', + 'clientVersion': '16.20', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 5, + 'REQUIRE_JS_PLAYER': False + }, + 'ios_embedded': { + 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'IOS_MESSAGES_EXTENSION', + 'clientVersion': '16.20', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 66, + 'REQUIRE_JS_PLAYER': False + }, + 'ios_music': { + 'INNERTUBE_API_KEY': 'AIzaSyDK3iBpDP9nHVTk2qL73FLJICfOC3c51Og', + 'INNERTUBE_HOST': 'music.youtube.com', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'IOS_MUSIC', + 'clientVersion': '4.32', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 26, + 'REQUIRE_JS_PLAYER': False + }, + 'ios_creator': { + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'IOS_CREATOR', + 'clientVersion': '21.24.100', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 15, + 'REQUIRE_JS_PLAYER': False + }, + # mweb has 'ultralow' formats + # See: https://github.com/hypervideo/hypervideo/pull/557 + 'mweb': { + 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'MWEB', + 'clientVersion': '2.20210721.07.00', + } + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 2 + }, +} + + +def build_innertube_clients(): + third_party = { + 'embedUrl': 'https://google.com', # Can be any valid URL + } + base_clients = ('android', 'web', 'ios', 'mweb') + priority = qualities(base_clients[::-1]) + + for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()): + ytcfg.setdefault('INNERTUBE_API_KEY', 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8') + ytcfg.setdefault('INNERTUBE_HOST', 'www.youtube.com') + ytcfg.setdefault('REQUIRE_JS_PLAYER', True) + ytcfg['INNERTUBE_CONTEXT']['client'].setdefault('hl', 'en') + ytcfg['priority'] = 10 * priority(client.split('_', 1)[0]) + + if client in base_clients: + INNERTUBE_CLIENTS[f'{client}_agegate'] = agegate_ytcfg = copy.deepcopy(ytcfg) + agegate_ytcfg['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED' + agegate_ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party + agegate_ytcfg['priority'] -= 1 + elif client.endswith('_embedded'): + ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = third_party + ytcfg['priority'] -= 2 + else: + ytcfg['priority'] -= 3 + + +build_innertube_clients() class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" - _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' - _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge' - _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup' - _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge' - _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}' + _RESERVED_NAMES = ( + r'channel|c|user|playlist|watch|w|v|embed|e|watch_popup|clip|' + r'shorts|movies|results|shared|hashtag|trending|feed|feeds|' + r'browse|oembed|get_video_info|iframe_api|s/player|' + r'storefront|oops|index|account|reporthistory|t/terms|about|upload|signin|logout') + + _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM|WL|LL|LM)' _NETRC_MACHINE = 'youtube' + # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False - _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM)' - def _login(self): """ Attempt to log in to YouTube. - True is returned if successful or skipped. - False is returned if login failed. - If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. """ - username, password = self._get_login_info() - # No authentication to be performed - if username is None: - if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None: - raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True) - return True - - login_page = self._download_webpage( - self._LOGIN_URL, None, - note='Downloading login page', - errnote='unable to fetch login page', fatal=False) - if login_page is False: - return - - login_form = self._hidden_inputs(login_page) - - def req(url, f_req, note, errnote): - data = login_form.copy() - data.update({ - 'pstMsg': 1, - 'checkConnection': 'youtube', - 'checkedDomains': 'youtube', - 'hl': 'en', - 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]', - 'f.req': json.dumps(f_req), - 'flowName': 'GlifWebSignIn', - 'flowEntry': 'ServiceLogin', - # TODO: reverse actual botguard identifier generation algo - 'bgRequest': '["identifier",""]', - }) - return self._download_json( - url, None, note=note, errnote=errnote, - transform_source=lambda s: re.sub(r'^[^[]*', '', s), - fatal=False, - data=urlencode_postdata(data), headers={ - 'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8', - 'Google-Accounts-XSRF': 1, - }) - - def warn(message): - self._downloader.report_warning(message) - - lookup_req = [ - username, - None, [], None, 'US', None, None, 2, False, True, - [ - None, None, - [2, 1, None, 1, - 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', - None, [], 4], - 1, [None, None, []], None, None, None, True - ], - username, - ] - - lookup_results = req( - self._LOOKUP_URL, lookup_req, - 'Looking up account info', 'Unable to look up account info') - - if lookup_results is False: - return False - - user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str) - if not user_hash: - warn('Unable to extract user hash') - return False - - challenge_req = [ - user_hash, - None, 1, None, [1, None, None, None, [password, None, True]], - [ - None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4], - 1, [None, None, []], None, None, None, True - ]] - - challenge_results = req( - self._CHALLENGE_URL, challenge_req, - 'Logging in', 'Unable to log in') - if challenge_results is False: - return - - login_res = try_get(challenge_results, lambda x: x[0][5], list) - if login_res: - login_msg = try_get(login_res, lambda x: x[5], compat_str) - warn( - 'Unable to login: %s' % 'Invalid password' - if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg) - return False - - res = try_get(challenge_results, lambda x: x[0][-1], list) - if not res: - warn('Unable to extract result entry') - return False - - login_challenge = try_get(res, lambda x: x[0][0], list) - if login_challenge: - challenge_str = try_get(login_challenge, lambda x: x[2], compat_str) - if challenge_str == 'TWO_STEP_VERIFICATION': - # SEND_SUCCESS - TFA code has been successfully sent to phone - # QUOTA_EXCEEDED - reached the limit of TFA codes - status = try_get(login_challenge, lambda x: x[5], compat_str) - if status == 'QUOTA_EXCEEDED': - warn('Exceeded the limit of TFA codes, try later') - return False - - tl = try_get(challenge_results, lambda x: x[1][2], compat_str) - if not tl: - warn('Unable to extract TL') - return False - - tfa_code = self._get_tfa_info('2-step verification code') - - if not tfa_code: - warn( - 'Two-factor authentication required. Provide it either interactively or with --twofactor <code>' - '(Note that only TOTP (Google Authenticator App) codes work at this time.)') - return False - - tfa_code = remove_start(tfa_code, 'G-') - - tfa_req = [ - user_hash, None, 2, None, - [ - 9, None, None, None, None, None, None, None, - [None, tfa_code, True, 2] - ]] - - tfa_results = req( - self._TFA_URL.format(tl), tfa_req, - 'Submitting TFA code', 'Unable to submit TFA code') - - if tfa_results is False: - return False - - tfa_res = try_get(tfa_results, lambda x: x[0][5], list) - if tfa_res: - tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str) - warn( - 'Unable to finish TFA: %s' % 'Invalid TFA code' - if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg) - return False - - check_cookie_url = try_get( - tfa_results, lambda x: x[0][-1][2], compat_str) - else: - CHALLENGES = { - 'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.", - 'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.', - 'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.", - } - challenge = CHALLENGES.get( - challenge_str, - '%s returned error %s.' % (self.IE_NAME, challenge_str)) - warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge) - return False - else: - check_cookie_url = try_get(res, lambda x: x[2], compat_str) - - if not check_cookie_url: - warn('Unable to extract CheckCookie URL') - return False - - check_cookie_results = self._download_webpage( - check_cookie_url, None, 'Checking cookie', fatal=False) - - if check_cookie_results is False: - return False - - if 'https://myaccount.google.com/' not in check_cookie_results: - warn('Unable to log in') - return False - - return True + if (self._LOGIN_REQUIRED + and self.get_param('cookiefile') is None + and self.get_param('cookiesfrombrowser') is None): + self.raise_login_required( + 'Login details are needed to download this content', method='cookies') + username, password = self._get_login_info() + if username: + self.report_warning(f'Cannot login to YouTube using username and password. {self._LOGIN_HINTS["cookies"]}') def _initialize_consent(self): cookies = self._get_cookies('https://www.youtube.com/') @@ -265,73 +290,402 @@ class YoutubeBaseInfoExtractor(InfoExtractor): def _real_initialize(self): self._initialize_consent() - if self._downloader is None: - return - if not self._login(): - return - - _DEFAULT_API_DATA = { - 'context': { - 'client': { - 'clientName': 'WEB', - 'clientVersion': '2.20201021.03.00', - } - }, - } + self._login() _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;' _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;' _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)' - def _call_api(self, ep, query, video_id, fatal=True): - data = self._DEFAULT_API_DATA.copy() - data.update(query) + def _get_default_ytcfg(self, client='web'): + return copy.deepcopy(INNERTUBE_CLIENTS[client]) + + def _get_innertube_host(self, client='web'): + return INNERTUBE_CLIENTS[client]['INNERTUBE_HOST'] + + def _ytcfg_get_safe(self, ytcfg, getter, expected_type=None, default_client='web'): + # try_get but with fallback to default ytcfg client values when present + _func = lambda y: try_get(y, getter, expected_type) + return _func(ytcfg) or _func(self._get_default_ytcfg(default_client)) + + def _extract_client_name(self, ytcfg, default_client='web'): + return self._ytcfg_get_safe( + ytcfg, (lambda x: x['INNERTUBE_CLIENT_NAME'], + lambda x: x['INNERTUBE_CONTEXT']['client']['clientName']), compat_str, default_client) + def _extract_client_version(self, ytcfg, default_client='web'): + return self._ytcfg_get_safe( + ytcfg, (lambda x: x['INNERTUBE_CLIENT_VERSION'], + lambda x: x['INNERTUBE_CONTEXT']['client']['clientVersion']), compat_str, default_client) + + def _extract_api_key(self, ytcfg=None, default_client='web'): + return self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_API_KEY'], compat_str, default_client) + + def _extract_context(self, ytcfg=None, default_client='web'): + _get_context = lambda y: try_get(y, lambda x: x['INNERTUBE_CONTEXT'], dict) + context = _get_context(ytcfg) + if context: + return context + + context = _get_context(self._get_default_ytcfg(default_client)) + if not ytcfg: + return context + + # Recreate the client context (required) + context['client'].update({ + 'clientVersion': self._extract_client_version(ytcfg, default_client), + 'clientName': self._extract_client_name(ytcfg, default_client), + }) + visitor_data = try_get(ytcfg, lambda x: x['VISITOR_DATA'], compat_str) + if visitor_data: + context['client']['visitorData'] = visitor_data + return context + + _SAPISID = None + + def _generate_sapisidhash_header(self, origin='https://www.youtube.com'): + time_now = round(time.time()) + if self._SAPISID is None: + yt_cookies = self._get_cookies('https://www.youtube.com') + # Sometimes SAPISID cookie isn't present but __Secure-3PAPISID is. + # See: https://github.com/hypervideo/hypervideo/issues/393 + sapisid_cookie = dict_get( + yt_cookies, ('__Secure-3PAPISID', 'SAPISID')) + if sapisid_cookie and sapisid_cookie.value: + self._SAPISID = sapisid_cookie.value + self.write_debug('Extracted SAPISID cookie') + # SAPISID cookie is required if not already present + if not yt_cookies.get('SAPISID'): + self.write_debug('Copying __Secure-3PAPISID cookie to SAPISID cookie') + self._set_cookie( + '.youtube.com', 'SAPISID', self._SAPISID, secure=True, expire_time=time_now + 3600) + else: + self._SAPISID = False + if not self._SAPISID: + return None + # SAPISIDHASH algorithm from https://stackoverflow.com/a/32065323 + sapisidhash = hashlib.sha1( + f'{time_now} {self._SAPISID} {origin}'.encode('utf-8')).hexdigest() + return f'SAPISIDHASH {time_now}_{sapisidhash}' + + def _call_api(self, ep, query, video_id, fatal=True, headers=None, + note='Downloading API JSON', errnote='Unable to download API page', + context=None, api_key=None, api_hostname=None, default_client='web'): + + data = {'context': context} if context else {'context': self._extract_context(default_client=default_client)} + data.update(query) + real_headers = self.generate_api_headers(default_client=default_client) + real_headers.update({'content-type': 'application/json'}) + if headers: + real_headers.update(headers) return self._download_json( - 'https://www.youtube.com/youtubei/v1/%s' % ep, video_id=video_id, - note='Downloading API JSON', errnote='Unable to download API page', - data=json.dumps(data).encode('utf8'), fatal=fatal, - headers={'content-type': 'application/json'}, - query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'}) + 'https://%s/youtubei/v1/%s' % (api_hostname or self._get_innertube_host(default_client), ep), + video_id=video_id, fatal=fatal, note=note, errnote=errnote, + data=json.dumps(data).encode('utf8'), headers=real_headers, + query={'key': api_key or self._extract_api_key()}) + + def extract_yt_initial_data(self, item_id, webpage, fatal=True): + data = self._search_regex( + (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE), + self._YT_INITIAL_DATA_RE), webpage, 'yt initial data', fatal=fatal) + if data: + return self._parse_json(data, item_id, fatal=fatal) - def _extract_yt_initial_data(self, video_id, webpage): - return self._parse_json( - self._search_regex( - (r'%s\s*%s' % (self._YT_INITIAL_DATA_RE, self._YT_INITIAL_BOUNDARY_RE), - self._YT_INITIAL_DATA_RE), webpage, 'yt initial data'), - video_id) + @staticmethod + def _extract_session_index(*data): + """ + Index of current account in account list. + See: https://github.com/hypervideo/hypervideo/pull/519 + """ + for ytcfg in data: + session_index = int_or_none(try_get(ytcfg, lambda x: x['SESSION_INDEX'])) + if session_index is not None: + return session_index + + # Deprecated? + def _extract_identity_token(self, ytcfg=None, webpage=None): + if ytcfg: + token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str) + if token: + return token + if webpage: + return self._search_regex( + r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage, + 'identity token', default=None, fatal=False) + + @staticmethod + def _extract_account_syncid(*args): + """ + Extract syncId required to download private playlists of secondary channels + @params response and/or ytcfg + """ + for data in args: + # ytcfg includes channel_syncid if on secondary channel + delegated_sid = try_get(data, lambda x: x['DELEGATED_SESSION_ID'], compat_str) + if delegated_sid: + return delegated_sid + sync_ids = (try_get( + data, (lambda x: x['responseContext']['mainAppWebResponseContext']['datasyncId'], + lambda x: x['DATASYNC_ID']), compat_str) or '').split('||') + if len(sync_ids) >= 2 and sync_ids[1]: + # datasyncid is of the form "channel_syncid||user_syncid" for secondary channel + # and just "user_syncid||" for primary channel. We only want the channel_syncid + return sync_ids[0] - def _extract_ytcfg(self, video_id, webpage): + @staticmethod + def _extract_visitor_data(*args): + """ + Extracts visitorData from an API response or ytcfg + Appears to be used to track session state + """ + return traverse_obj( + args, (..., ('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))), + expected_type=compat_str, get_all=False) + + @property + def is_authenticated(self): + return bool(self._generate_sapisidhash_header()) + + def extract_ytcfg(self, video_id, webpage): + if not webpage: + return {} return self._parse_json( self._search_regex( r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg', default='{}'), video_id, fatal=False) or {} + def generate_api_headers( + self, *, ytcfg=None, account_syncid=None, session_index=None, + visitor_data=None, identity_token=None, api_hostname=None, default_client='web'): + + origin = 'https://' + (api_hostname if api_hostname else self._get_innertube_host(default_client)) + headers = { + 'X-YouTube-Client-Name': compat_str( + self._ytcfg_get_safe(ytcfg, lambda x: x['INNERTUBE_CONTEXT_CLIENT_NAME'], default_client=default_client)), + 'X-YouTube-Client-Version': self._extract_client_version(ytcfg, default_client), + 'Origin': origin, + 'X-Youtube-Identity-Token': identity_token or self._extract_identity_token(ytcfg), + 'X-Goog-PageId': account_syncid or self._extract_account_syncid(ytcfg), + 'X-Goog-Visitor-Id': visitor_data or self._extract_visitor_data(ytcfg) + } + if session_index is None: + session_index = self._extract_session_index(ytcfg) + if account_syncid or session_index is not None: + headers['X-Goog-AuthUser'] = session_index if session_index is not None else 0 + + auth = self._generate_sapisidhash_header(origin) + if auth is not None: + headers['Authorization'] = auth + headers['X-Origin'] = origin + return {h: v for h, v in headers.items() if v is not None} + + @staticmethod + def _build_api_continuation_query(continuation, ctp=None): + query = { + 'continuation': continuation + } + # TODO: Inconsistency with clickTrackingParams. + # Currently we have a fixed ctp contained within context (from ytcfg) + # and a ctp in root query for continuation. + if ctp: + query['clickTracking'] = {'clickTrackingParams': ctp} + return query + + @classmethod + def _extract_next_continuation_data(cls, renderer): + next_continuation = try_get( + renderer, (lambda x: x['continuations'][0]['nextContinuationData'], + lambda x: x['continuation']['reloadContinuationData']), dict) + if not next_continuation: + return + continuation = next_continuation.get('continuation') + if not continuation: + return + ctp = next_continuation.get('clickTrackingParams') + return cls._build_api_continuation_query(continuation, ctp) + + @classmethod + def _extract_continuation_ep_data(cls, continuation_ep: dict): + if isinstance(continuation_ep, dict): + continuation = try_get( + continuation_ep, lambda x: x['continuationCommand']['token'], compat_str) + if not continuation: + return + ctp = continuation_ep.get('clickTrackingParams') + return cls._build_api_continuation_query(continuation, ctp) + + @classmethod + def _extract_continuation(cls, renderer): + next_continuation = cls._extract_next_continuation_data(renderer) + if next_continuation: + return next_continuation + + contents = [] + for key in ('contents', 'items'): + contents.extend(try_get(renderer, lambda x: x[key], list) or []) + + for content in contents: + if not isinstance(content, dict): + continue + continuation_ep = try_get( + content, (lambda x: x['continuationItemRenderer']['continuationEndpoint'], + lambda x: x['continuationItemRenderer']['button']['buttonRenderer']['command']), + dict) + continuation = cls._extract_continuation_ep_data(continuation_ep) + if continuation: + return continuation + + @classmethod + def _extract_alerts(cls, data): + for alert_dict in try_get(data, lambda x: x['alerts'], list) or []: + if not isinstance(alert_dict, dict): + continue + for alert in alert_dict.values(): + alert_type = alert.get('type') + if not alert_type: + continue + message = cls._get_text(alert, 'text') + if message: + yield alert_type, message + + def _report_alerts(self, alerts, expected=True, fatal=True, only_once=False): + errors = [] + warnings = [] + for alert_type, alert_message in alerts: + if alert_type.lower() == 'error' and fatal: + errors.append([alert_type, alert_message]) + else: + warnings.append([alert_type, alert_message]) + + for alert_type, alert_message in (warnings + errors[:-1]): + self.report_warning('YouTube said: %s - %s' % (alert_type, alert_message), only_once=only_once) + if errors: + raise ExtractorError('YouTube said: %s' % errors[-1][1], expected=expected) + + def _extract_and_report_alerts(self, data, *args, **kwargs): + return self._report_alerts(self._extract_alerts(data), *args, **kwargs) + + def _extract_badges(self, renderer: dict): + badges = set() + for badge in try_get(renderer, lambda x: x['badges'], list) or []: + label = try_get(badge, lambda x: x['metadataBadgeRenderer']['label'], compat_str) + if label: + badges.add(label.lower()) + return badges + + @staticmethod + def _get_text(data, *path_list, max_runs=None): + for path in path_list or [None]: + if path is None: + obj = [data] + else: + obj = traverse_obj(data, path, default=[]) + if not any(key is ... or isinstance(key, (list, tuple)) for key in variadic(path)): + obj = [obj] + for item in obj: + text = try_get(item, lambda x: x['simpleText'], compat_str) + if text: + return text + runs = try_get(item, lambda x: x['runs'], list) or [] + if not runs and isinstance(item, list): + runs = item + + runs = runs[:min(len(runs), max_runs or len(runs))] + text = ''.join(traverse_obj(runs, (..., 'text'), expected_type=str, default=[])) + if text: + return text + + def _extract_response(self, item_id, query, note='Downloading API JSON', headers=None, + ytcfg=None, check_get_keys=None, ep='browse', fatal=True, api_hostname=None, + default_client='web'): + response = None + last_error = None + count = -1 + retries = self.get_param('extractor_retries', 3) + if check_get_keys is None: + check_get_keys = [] + while count < retries: + count += 1 + if last_error: + self.report_warning('%s. Retrying ...' % remove_end(last_error, '.')) + try: + response = self._call_api( + ep=ep, fatal=True, headers=headers, + video_id=item_id, query=query, + context=self._extract_context(ytcfg, default_client), + api_key=self._extract_api_key(ytcfg, default_client), + api_hostname=api_hostname, default_client=default_client, + note='%s%s' % (note, ' (retry #%d)' % count if count else '')) + except ExtractorError as e: + if isinstance(e.cause, network_exceptions): + if isinstance(e.cause, compat_HTTPError) and not is_html(e.cause.read(512)): + e.cause.seek(0) + yt_error = try_get( + self._parse_json(e.cause.read().decode(), item_id, fatal=False), + lambda x: x['error']['message'], compat_str) + if yt_error: + self._report_alerts([('ERROR', yt_error)], fatal=False) + # Downloading page may result in intermittent 5xx HTTP error + # Sometimes a 404 is also recieved. See: https://github.com/ytdl-org/youtube-dl/issues/28289 + # We also want to catch all other network exceptions since errors in later pages can be troublesome + # See https://github.com/hypervideo/hypervideo/issues/507#issuecomment-880188210 + if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429): + last_error = error_to_compat_str(e.cause or e.msg) + if count < retries: + continue + if fatal: + raise + else: + self.report_warning(error_to_compat_str(e)) + return + + else: + try: + self._extract_and_report_alerts(response, only_once=True) + except ExtractorError as e: + # YouTube servers may return errors we want to retry on in a 200 OK response + # See: https://github.com/hypervideo/hypervideo/issues/839 + if 'unknown error' in e.msg.lower(): + last_error = e.msg + continue + if fatal: + raise + self.report_warning(error_to_compat_str(e)) + return + if not check_get_keys or dict_get(response, check_get_keys): + break + # Youtube sometimes sends incomplete data + # See: https://github.com/ytdl-org/youtube-dl/issues/28194 + last_error = 'Incomplete data received' + if count >= retries: + if fatal: + raise ExtractorError(last_error) + else: + self.report_warning(last_error) + return + return response + + @staticmethod + def is_music_url(url): + return re.match(r'https?://music\.youtube\.com/', url) is not None + def _extract_video(self, renderer): - video_id = renderer['videoId'] - title = try_get( - renderer, - (lambda x: x['title']['runs'][0]['text'], - lambda x: x['title']['simpleText']), compat_str) - description = try_get( - renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'], - compat_str) - duration = parse_duration(try_get( - renderer, lambda x: x['lengthText']['simpleText'], compat_str)) - view_count_text = try_get( - renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or '' + video_id = renderer.get('videoId') + title = self._get_text(renderer, 'title') + description = self._get_text(renderer, 'descriptionSnippet') + duration = parse_duration(self._get_text( + renderer, 'lengthText', ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'text'))) + view_count_text = self._get_text(renderer, 'viewCountText') or '' view_count = str_to_int(self._search_regex( r'^([\d,]+)', re.sub(r'\s', '', view_count_text), 'view count', default=None)) - uploader = try_get( - renderer, - (lambda x: x['ownerText']['runs'][0]['text'], - lambda x: x['shortBylineText']['runs'][0]['text']), compat_str) + + uploader = self._get_text(renderer, 'ownerText', 'shortBylineText') + return { '_type': 'url', 'ie_key': YoutubeIE.ie_key(), 'id': video_id, - 'url': video_id, + 'url': f'https://www.youtube.com/watch?v={video_id}', 'title': title, 'description': description, 'duration': duration, @@ -347,13 +701,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'(?:www\.)?redirect\.invidious\.io', r'(?:(?:www|dev)\.)?invidio\.us', # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md + r'(?:www\.)?invidious\.pussthecat\.org', + r'(?:www\.)?invidious\.zee\.li', + r'(?:www\.)?invidious\.ethibox\.fr', + r'(?:www\.)?invidious\.3o7z6yfxhbw7n3za4rss6l434kmv55cgw2vuziwuigpwegswvwzqipyd\.onion', + # youtube-dl invidious instances list r'(?:(?:www|no)\.)?invidiou\.sh', r'(?:(?:www|fi)\.)?invidious\.snopyta\.org', r'(?:www\.)?invidious\.kabi\.tk', - r'(?:www\.)?invidious\.13ad\.de', r'(?:www\.)?invidious\.mastodon\.host', r'(?:www\.)?invidious\.zapashcanon\.fr', - r'(?:www\.)?invidious\.kavin\.rocks', + r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks', r'(?:www\.)?invidious\.tinfoil-hat\.net', r'(?:www\.)?invidious\.himiko\.cloud', r'(?:www\.)?invidious\.reallyancient\.tech', @@ -380,6 +738,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'(?:www\.)?invidious\.toot\.koeln', r'(?:www\.)?invidious\.fdn\.fr', r'(?:www\.)?watch\.nettohikari\.com', + r'(?:www\.)?invidious\.namazso\.eu', + r'(?:www\.)?invidious\.silkky\.cloud', + r'(?:www\.)?invidious\.exonip\.de', + r'(?:www\.)?invidious\.riverside\.rocks', + r'(?:www\.)?invidious\.blamefran\.net', + r'(?:www\.)?invidious\.moomoo\.de', + r'(?:www\.)?ytb\.trom\.tf', + r'(?:www\.)?yt\.cyberhost\.uk', r'(?:www\.)?kgg2m7yk5aybusll\.onion', r'(?:www\.)?qklhadlycap4cnod\.onion', r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion', @@ -388,6 +754,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion', r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p', r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion', + r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion', + r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion', + r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion', + r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion', ) _VALID_URL = r"""(?x)^ ( @@ -402,7 +772,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls (?: # the various things that can precede the ID: - (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/ + (?:(?:v|embed|e|shorts)/(?!videoseries)) # v/ or embed/ or e/ or shorts/ |(?: # or the v= param in all its forms (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)? # preceding watch(_popup|.php) or nothing (like /?v=xxxx) (?:\?|\#!?) # the params delimiter ? or # or #! @@ -421,7 +791,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): )? # all until now is optional -> you can pass the naked ID (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow - $""" % { + (?:\#|$)""" % { 'invidious': '|'.join(_INVIDIOUS_SITES), } _PLAYER_INFO_RE = ( @@ -429,7 +799,116 @@ class YoutubeIE(YoutubeBaseInfoExtractor): r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$', r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$', ) - _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt') + _formats = { + '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, + '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, + '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'}, + '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'}, + '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'}, + '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well + '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'}, + '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + + + # 3D videos + '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20}, + '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + + # Apple HTTP Live Streaming + '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, + + # DASH mp4 video + '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559) + '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, + '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, + '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'}, + + # Dash mp4 audio + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'}, + '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, + '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, + '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'}, + '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'}, + + # Dash webm + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'}, + '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) + '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + + # Dash webm audio + '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128}, + '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256}, + + # Dash webm audio with opus inside + '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50}, + '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70}, + '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160}, + + # RTMP (unnamed) + '_rtmp': {'protocol': 'rtmp'}, + + # av01 video only formats sometimes served with "unknown" codecs + '394': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'}, + '395': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'av01.0.00M.08'}, + '396': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'av01.0.01M.08'}, + '397': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'av01.0.04M.08'}, + '398': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'av01.0.05M.08'}, + '399': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'av01.0.08M.08'}, + '400': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'}, + '401': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'av01.0.12M.08'}, + } + _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt') _GEO_BYPASS = False @@ -440,16 +919,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'BaW_jenozKc', 'ext': 'mp4', - 'title': 'hypervideo test video "\'/\\ä↭𝕐', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', 'uploader': 'Philipp Hagemeister', 'uploader_id': 'phihag', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag', 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q', 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q', 'upload_date': '20121002', - 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for hypervideo.\n\nFor more information, contact phihag@phihag.de .', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', 'categories': ['Science & Technology'], - 'tags': ['hypervideo'], + 'tags': ['youtube-dl'], 'duration': 10, 'view_count': int, 'like_count': int, @@ -480,14 +959,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'info_dict': { 'id': 'BaW_jenozKc', 'ext': 'mp4', - 'title': 'hypervideo test video "\'/\\ä↭𝕐', + 'title': 'youtube-dl test video "\'/\\ä↭𝕐', 'uploader': 'Philipp Hagemeister', 'uploader_id': 'phihag', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag', 'upload_date': '20121002', - 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for hypervideo.\n\nFor more information, contact phihag@phihag.de .', + 'description': 'test chars: "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .', 'categories': ['Science & Technology'], - 'tags': ['hypervideo'], + 'tags': ['youtube-dl'], 'duration': 10, 'view_count': int, 'like_count': int, @@ -535,23 +1014,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'format': '141/bestaudio[ext=m4a]', }, }, - # Controversy video - { - 'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8', - 'info_dict': { - 'id': 'T4XJQO3qol8', - 'ext': 'mp4', - 'duration': 219, - 'upload_date': '20100909', - 'uploader': 'Amazing Atheist', - 'uploader_id': 'TheAmazingAtheist', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist', - 'title': 'Burning Everyone\'s Koran', - 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html', - } - }, - # Normal age-gate video (No vevo, embed allowed), available via embed page + # Age-gate videos. See https://github.com/hypervideo/hypervideo/pull/575#issuecomment-888837000 { + 'note': 'Embed allowed age-gate video', 'url': 'https://youtube.com/watch?v=HtVdAasjOgU', 'info_dict': { 'id': 'HtVdAasjOgU', @@ -567,9 +1032,49 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, }, { - # Age-gated video only available with authentication (unavailable - # via embed page workaround) - 'url': 'XgnwCQzjau8', + 'note': 'Age-gate video with embed allowed in public site', + 'url': 'https://youtube.com/watch?v=HsUATh_Nc2U', + 'info_dict': { + 'id': 'HsUATh_Nc2U', + 'ext': 'mp4', + 'title': 'Godzilla 2 (Official Video)', + 'description': 'md5:bf77e03fcae5529475e500129b05668a', + 'upload_date': '20200408', + 'uploader_id': 'FlyingKitty900', + 'uploader': 'FlyingKitty', + 'age_limit': 18, + }, + }, + { + 'note': 'Age-gate video embedable only with clientScreen=EMBED', + 'url': 'https://youtube.com/watch?v=Tq92D6wQ1mg', + 'info_dict': { + 'id': 'Tq92D6wQ1mg', + 'title': '[MMD] Adios - EVERGLOW [+Motion DL]', + 'ext': 'mp4', + 'upload_date': '20191227', + 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ', + 'uploader': 'Projekt Melody', + 'description': 'md5:17eccca93a786d51bc67646756894066', + 'age_limit': 18, + }, + }, + { + 'note': 'Non-Agegated non-embeddable video', + 'url': 'https://youtube.com/watch?v=MeJVWBSsPAY', + 'info_dict': { + 'id': 'MeJVWBSsPAY', + 'ext': 'mp4', + 'title': 'OOMPH! - Such Mich Find Mich (Lyrics)', + 'uploader': 'Herr Lurik', + 'uploader_id': 'st3in234', + 'description': 'Fan Video. Music & Lyrics by OOMPH!.', + 'upload_date': '20130730', + }, + }, + { + 'note': 'Non-bypassable age-gated video', + 'url': 'https://youtube.com/watch?v=Cr381pDsSsA', 'only_matching': True, }, # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421) @@ -604,7 +1109,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'olympic', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic', 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', - 'uploader': 'Olympic', + 'uploader': 'Olympics', 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', }, 'params': { @@ -740,6 +1245,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'Not multifeed anymore', }, { # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536) @@ -769,16 +1275,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'lsguqyKfVQg', 'ext': 'mp4', 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21', - 'alt_title': 'Dark Walk - Position Music', + 'alt_title': 'Dark Walk', 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a', 'duration': 133, 'upload_date': '20151119', 'uploader_id': 'IronSoulElf', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf', 'uploader': 'IronSoulElf', - 'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan', - 'track': 'Dark Walk - Position Music', - 'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan', + 'creator': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan', + 'track': 'Dark Walk', + 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan', 'album': 'Position Music - Production Music Vol. 143 - Dark Walk', }, 'params': { @@ -1089,6 +1595,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'only_matching': True, }, { + # controversial video, requires bpctr/contentCheckOk + 'url': 'https://www.youtube.com/watch?v=SZJvDhaSDnc', + 'info_dict': { + 'id': 'SZJvDhaSDnc', + 'ext': 'mp4', + 'title': 'San Diego teen commits suicide after bullying over embarrassing video', + 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ', + 'uploader': 'CBS This Morning', + 'uploader_id': 'CBSThisMorning', + 'upload_date': '20140716', + 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7' + } + }, + { # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685 'url': 'cBvYw8_A0vQ', 'info_dict': { @@ -1104,119 +1624,55 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + }, { + # Has multiple audio streams + 'url': 'WaOKSUlf4TM', + 'only_matching': True + }, { + # Requires Premium: has format 141 when requested using YTM url + 'url': 'https://music.youtube.com/watch?v=XclachpHxis', + 'only_matching': True + }, { + # multiple subtitles with same lang_code + 'url': 'https://www.youtube.com/watch?v=wsQiKKfKxug', + 'only_matching': True, + }, { + # Force use android client fallback + 'url': 'https://www.youtube.com/watch?v=YOelRv7fMxY', + 'info_dict': { + 'id': 'YOelRv7fMxY', + 'title': 'DIGGING A SECRET TUNNEL Part 1', + 'ext': '3gp', + 'upload_date': '20210624', + 'channel_id': 'UCp68_FLety0O-n9QU6phsgw', + 'uploader': 'colinfurze', + 'uploader_id': 'colinfurze', + 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw', + 'description': 'md5:b5096f56af7ccd7a555c84db81738b22' + }, + 'params': { + 'format': '17', # 3gp format available on android + 'extractor_args': {'youtube': {'player_client': ['android']}}, + }, + }, + { + # Skip download of additional client configs (remix client config in this case) + 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs', + 'only_matching': True, + 'params': { + 'extractor_args': {'youtube': {'player_skip': ['configs']}}, + }, + }, { + # shorts + 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY', + 'only_matching': True, }, ] - _formats = { - '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, - '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, - '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'}, - '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'}, - '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'}, - '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well - '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'}, - '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, - '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, - '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, - '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, - '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - - - # 3D videos - '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, - '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, - '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, - '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, - '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20}, - '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, - '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, - - # Apple HTTP Live Streaming - '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, - '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, - '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, - '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, - '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, - - # DASH mp4 video - '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559) - '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, - '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, - '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'}, - - # Dash mp4 audio - '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'}, - '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'}, - '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'}, - '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, - '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, - '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'}, - '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'}, - - # Dash webm - '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'}, - '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) - '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - - # Dash webm audio - '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128}, - '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256}, - - # Dash webm audio with opus inside - '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50}, - '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70}, - '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160}, - - # RTMP (unnamed) - '_rtmp': {'protocol': 'rtmp'}, - - # av01 video only formats sometimes served with "unknown" codecs - '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, - '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, - '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, - '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'}, - } @classmethod def suitable(cls, url): - # Hack for lazy extractors until more generic solution is implemented - # (see #28780) - from .youtube import parse_qs + from ..utils import parse_qs + qs = parse_qs(url) if qs.get('list', [None])[0]: return False @@ -1227,6 +1683,29 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self._code_cache = {} self._player_cache = {} + def _extract_player_url(self, *ytcfgs, webpage=None): + player_url = traverse_obj( + ytcfgs, (..., 'PLAYER_JS_URL'), (..., 'WEB_PLAYER_CONTEXT_CONFIGS', ..., 'jsUrl'), + get_all=False, expected_type=compat_str) + if not player_url: + return + if player_url.startswith('//'): + player_url = 'https:' + player_url + elif not re.match(r'https?://', player_url): + player_url = compat_urlparse.urljoin( + 'https://www.youtube.com', player_url) + return player_url + + def _download_player_url(self, video_id, fatal=False): + res = self._download_webpage( + 'https://www.youtube.com/iframe_api', + note='Downloading iframe API JS', video_id=video_id, fatal=fatal) + if res: + player_version = self._search_regex( + r'player\\?/([0-9a-fA-F]{8})\\?/', res, 'player version', fatal=fatal) + if player_version: + return f'https://www.youtube.com/s/player/{player_version}/player_ias.vflset/en_US/base.js' + def _signature_cache_id(self, example_sig): """ Return a string representation of a signature """ return '.'.join(compat_str(len(part)) for part in example_sig.split('.')) @@ -1241,6 +1720,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): raise ExtractorError('Cannot identify player %r' % player_url) return id_m.group('id') + def _load_player(self, video_id, player_url, fatal=True) -> bool: + player_id = self._extract_player_info(player_url) + if player_id not in self._code_cache: + code = self._download_webpage( + player_url, video_id, fatal=fatal, + note='Downloading player ' + player_id, + errnote='Download of %s failed' % player_url) + if code: + self._code_cache[player_id] = code + return player_id in self._code_cache + def _extract_signature_function(self, video_id, player_url, example_sig): player_id = self._extract_player_info(player_url) @@ -1253,20 +1743,16 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if cache_spec is not None: return lambda s: ''.join(s[i] for i in cache_spec) - if player_id not in self._code_cache: - self._code_cache[player_id] = self._download_webpage( - player_url, video_id, - note='Downloading player ' + player_id, - errnote='Download of %s failed' % player_url) - code = self._code_cache[player_id] - res = self._parse_sig_js(code) + if self._load_player(video_id, player_url): + code = self._code_cache[player_id] + res = self._parse_sig_js(code) - test_string = ''.join(map(compat_chr, range(len(example_sig)))) - cache_res = res(test_string) - cache_spec = [ord(c) for c in cache_res] + test_string = ''.join(map(compat_chr, range(len(example_sig)))) + cache_res = res(test_string) + cache_spec = [ord(c) for c in cache_res] - self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) - return res + self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec) + return res def _print_sig_code(self, func, example_sig): def gen_sig_code(idxs): @@ -1311,10 +1797,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): funcname = self._search_regex( (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'(?:\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\(\s*""\s*\)', + 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'(?:\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\(\s*""\s*\)', 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$]+)\(', @@ -1337,11 +1823,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if player_url is None: raise ExtractorError('Cannot decrypt signature without player_url') - if player_url.startswith('//'): - player_url = 'https:' + player_url - elif not re.match(r'https?://', player_url): - player_url = compat_urlparse.urljoin( - 'https://www.youtube.com', player_url) try: player_id = (player_url, self._signature_cache_id(s)) if player_id not in self._player_cache: @@ -1350,7 +1831,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): ) self._player_cache[player_id] = func func = self._player_cache[player_id] - if self._downloader.params.get('youtube_print_sig_code'): + if self.get_param('youtube_print_sig_code'): self._print_sig_code(func, s) return func(s) except Exception as e: @@ -1358,11 +1839,37 @@ class YoutubeIE(YoutubeBaseInfoExtractor): raise ExtractorError( 'Signature extraction failed: ' + tb, cause=e) - def _mark_watched(self, video_id, player_response): - playback_url = url_or_none(try_get( - player_response, - lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl'])) + def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False): + """ + Extract signatureTimestamp (sts) + Required to tell API what sig/player version is in use. + """ + sts = None + if isinstance(ytcfg, dict): + sts = int_or_none(ytcfg.get('STS')) + + if not sts: + # Attempt to extract from player + if player_url is None: + error_msg = 'Cannot extract signature timestamp without player_url.' + if fatal: + raise ExtractorError(error_msg) + self.report_warning(error_msg) + return + if self._load_player(video_id, player_url, fatal=fatal): + player_id = self._extract_player_info(player_url) + code = self._code_cache[player_id] + sts = int_or_none(self._search_regex( + r'(?:signatureTimestamp|sts)\s*:\s*(?P<sts>[0-9]{5})', code, + 'JS player signature timestamp', group='sts', fatal=fatal)) + return sts + + def _mark_watched(self, video_id, player_responses): + playback_url = traverse_obj( + player_responses, (..., 'playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'), + expected_type=url_or_none, get_all=False) if not playback_url: + self.report_warning('Unable to mark watched') return parsed_playback_url = compat_urlparse.urlparse(playback_url) qs = compat_urlparse.parse_qs(parsed_playback_url.query) @@ -1425,47 +1932,58 @@ class YoutubeIE(YoutubeBaseInfoExtractor): mobj = re.match(cls._VALID_URL, url, re.VERBOSE) if mobj is None: raise ExtractorError('Invalid URL: %s' % url) - video_id = mobj.group(2) - return video_id - - def _extract_chapters_from_json(self, data, video_id, duration): - chapters_list = try_get( + return mobj.group('id') + + def _extract_chapters_from_json(self, data, duration): + chapter_list = traverse_obj( + data, ( + 'playerOverlays', 'playerOverlayRenderer', 'decoratedPlayerBarRenderer', + 'decoratedPlayerBarRenderer', 'playerBar', 'chapteredPlayerBarRenderer', 'chapters' + ), expected_type=list) + + return self._extract_chapters( + chapter_list, + chapter_time=lambda chapter: float_or_none( + traverse_obj(chapter, ('chapterRenderer', 'timeRangeStartMillis')), scale=1000), + chapter_title=lambda chapter: traverse_obj( + chapter, ('chapterRenderer', 'title', 'simpleText'), expected_type=str), + duration=duration) + + def _extract_chapters_from_engagement_panel(self, data, duration): + content_list = traverse_obj( data, - lambda x: x['playerOverlays'] - ['playerOverlayRenderer'] - ['decoratedPlayerBarRenderer'] - ['decoratedPlayerBarRenderer'] - ['playerBar'] - ['chapteredPlayerBarRenderer'] - ['chapters'], - list) - if not chapters_list: - return - - def chapter_time(chapter): - return float_or_none( - try_get( - chapter, - lambda x: x['chapterRenderer']['timeRangeStartMillis'], - int), - scale=1000) + ('engagementPanels', ..., 'engagementPanelSectionListRenderer', 'content', 'macroMarkersListRenderer', 'contents'), + expected_type=list, default=[]) + chapter_time = lambda chapter: parse_duration(self._get_text(chapter, 'timeDescription')) + chapter_title = lambda chapter: self._get_text(chapter, 'title') + + return next(( + filter(None, ( + self._extract_chapters( + traverse_obj(contents, (..., 'macroMarkersListItemRenderer')), + chapter_time, chapter_title, duration) + for contents in content_list + ))), []) + + def _extract_chapters(self, chapter_list, chapter_time, chapter_title, duration): chapters = [] - for next_num, chapter in enumerate(chapters_list, start=1): + last_chapter = {'start_time': 0} + for idx, chapter in enumerate(chapter_list or []): + title = chapter_title(chapter) start_time = chapter_time(chapter) if start_time is None: continue - end_time = (chapter_time(chapters_list[next_num]) - if next_num < len(chapters_list) else duration) - if end_time is None: - continue - title = try_get( - chapter, lambda x: x['chapterRenderer']['title']['simpleText'], - compat_str) - chapters.append({ - 'start_time': start_time, - 'end_time': end_time, - 'title': title, - }) + last_chapter['end_time'] = start_time + if start_time < last_chapter['start_time']: + if idx == 1: + chapters.pop() + self.report_warning('Invalid start time for chapter "%s"' % last_chapter['title']) + else: + self.report_warning(f'Invalid start time for chapter "{title}"') + continue + last_chapter = {'start_time': start_time, 'title': title} + chapters.append(last_chapter) + last_chapter['end_time'] = duration return chapters def _extract_yt_initial_variable(self, webpage, regex, video_id, name): @@ -1473,132 +1991,436 @@ class YoutubeIE(YoutubeBaseInfoExtractor): (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE), regex), webpage, name, default='{}'), video_id, fatal=False) - def _real_extract(self, url): - url, smuggled_data = unsmuggle_url(url, {}) - video_id = self._match_id(url) - base_url = self.http_scheme() + '//www.youtube.com/' - webpage_url = base_url + 'watch?v=' + video_id - webpage = self._download_webpage( - webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False) + @staticmethod + def parse_time_text(time_text): + """ + Parse the comment time text + time_text is in the format 'X units ago (edited)' + """ + time_text_split = time_text.split(' ') + if len(time_text_split) >= 3: + try: + return datetime_from_str('now-%s%s' % (time_text_split[0], time_text_split[1]), precision='auto') + except ValueError: + return None + + def _extract_comment(self, comment_renderer, parent=None): + comment_id = comment_renderer.get('commentId') + if not comment_id: + return - player_response = None - if webpage: - player_response = self._extract_yt_initial_variable( - webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, - video_id, 'initial player response') - if not player_response: - player_response = self._call_api( - 'player', {'videoId': video_id}, video_id) - - playability_status = player_response.get('playabilityStatus') or {} - if playability_status.get('reason') == 'Sign in to confirm your age': - pr = self._parse_json(try_get(compat_parse_qs( - self._download_webpage( - base_url + 'get_video_info', video_id, - 'Refetching age-gated info webpage', - 'unable to download video info webpage', query={ - 'video_id': video_id, - 'eurl': 'https://youtube.googleapis.com/v/' + video_id, - 'html5': 1, - }, fatal=False)), - lambda x: x['player_response'][0], - compat_str) or '{}', video_id) - if pr: - player_response = pr + text = self._get_text(comment_renderer, 'contentText') + + # note: timestamp is an estimate calculated from the current time and time_text + time_text = self._get_text(comment_renderer, 'publishedTimeText') or '' + time_text_dt = self.parse_time_text(time_text) + if isinstance(time_text_dt, datetime.datetime): + timestamp = calendar.timegm(time_text_dt.timetuple()) + author = self._get_text(comment_renderer, 'authorText') + author_id = try_get(comment_renderer, + lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str) + + votes = parse_count(try_get(comment_renderer, (lambda x: x['voteCount']['simpleText'], + lambda x: x['likeCount']), compat_str)) or 0 + author_thumbnail = try_get(comment_renderer, + lambda x: x['authorThumbnail']['thumbnails'][-1]['url'], compat_str) + + author_is_uploader = try_get(comment_renderer, lambda x: x['authorIsChannelOwner'], bool) + is_favorited = 'creatorHeart' in (try_get( + comment_renderer, lambda x: x['actionButtons']['commentActionButtonsRenderer'], dict) or {}) + return { + 'id': comment_id, + 'text': text, + 'timestamp': timestamp, + 'time_text': time_text, + 'like_count': votes, + 'is_favorited': is_favorited, + 'author': author, + 'author_id': author_id, + 'author_thumbnail': author_thumbnail, + 'author_is_uploader': author_is_uploader, + 'parent': parent or 'root' + } - trailer_video_id = try_get( - playability_status, - lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'], - compat_str) - if trailer_video_id: - return self.url_result( - trailer_video_id, self.ie_key(), trailer_video_id) + def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, comment_counts=None): - def get_text(x): - if not x: - return - text = x.get('simpleText') - if text and isinstance(text, compat_str): - return text - runs = x.get('runs') - if not isinstance(runs, list): - return - return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)]) - - search_meta = ( - lambda x: self._html_search_meta(x, webpage, default=None)) \ - if webpage else lambda x: None - - video_details = player_response.get('videoDetails') or {} - microformat = try_get( - player_response, - lambda x: x['microformat']['playerMicroformatRenderer'], - dict) or {} - video_title = video_details.get('title') \ - or get_text(microformat.get('title')) \ - or search_meta(['og:title', 'twitter:title', 'title']) - video_description = video_details.get('shortDescription') + def extract_header(contents): + _continuation = None + for content in contents: + comments_header_renderer = try_get(content, lambda x: x['commentsHeaderRenderer']) + expected_comment_count = parse_count(self._get_text( + comments_header_renderer, 'countText', 'commentsCount', max_runs=1)) + + if expected_comment_count: + comment_counts[1] = expected_comment_count + self.to_screen('Downloading ~%d comments' % expected_comment_count) + sort_mode_str = self._configuration_arg('comment_sort', [''])[0] + comment_sort_index = int(sort_mode_str != 'top') # 1 = new, 0 = top + + sort_menu_item = try_get( + comments_header_renderer, + lambda x: x['sortMenu']['sortFilterSubMenuRenderer']['subMenuItems'][comment_sort_index], dict) or {} + sort_continuation_ep = sort_menu_item.get('serviceEndpoint') or {} + + _continuation = self._extract_continuation_ep_data(sort_continuation_ep) or self._extract_continuation(sort_menu_item) + if not _continuation: + continue - if not smuggled_data.get('force_singlefeed', False): - if not self._downloader.params.get('noplaylist'): - multifeed_metadata_list = try_get( - player_response, - lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'], - compat_str) - if multifeed_metadata_list: - entries = [] - feed_ids = [] - for feed in multifeed_metadata_list.split(','): - # Unquote should take place before split on comma (,) since textual - # fields may contain comma as well (see - # https://github.com/ytdl-org/youtube-dl/issues/8536) - feed_data = compat_parse_qs( - compat_urllib_parse_unquote_plus(feed)) + sort_text = sort_menu_item.get('title') + if isinstance(sort_text, compat_str): + sort_text = sort_text.lower() + else: + sort_text = 'top comments' if comment_sort_index == 0 else 'newest first' + self.to_screen('Sorting comments by %s' % sort_text) + break + return _continuation - def feed_entry(name): - return try_get( - feed_data, lambda x: x[name][0], compat_str) + def extract_thread(contents): + if not parent: + comment_counts[2] = 0 + for content in contents: + comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer']) + comment_renderer = try_get( + comment_thread_renderer, (lambda x: x['comment']['commentRenderer'], dict)) or try_get( + content, (lambda x: x['commentRenderer'], dict)) - feed_id = feed_entry('id') - if not feed_id: - continue - feed_title = feed_entry('title') - title = video_title - if feed_title: - title += ' (%s)' % feed_title - entries.append({ - '_type': 'url_transparent', - 'ie_key': 'Youtube', - 'url': smuggle_url( - base_url + 'watch?v=' + feed_data['id'][0], - {'force_singlefeed': True}), - 'title': title, - }) - feed_ids.append(feed_id) - self.to_screen( - 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' - % (', '.join(feed_ids), video_id)) - return self.playlist_result( - entries, video_id, video_title, video_description) + if not comment_renderer: + continue + comment = self._extract_comment(comment_renderer, parent) + if not comment: + continue + comment_counts[0] += 1 + yield comment + # Attempt to get the replies + comment_replies_renderer = try_get( + comment_thread_renderer, lambda x: x['replies']['commentRepliesRenderer'], dict) + + if comment_replies_renderer: + comment_counts[2] += 1 + comment_entries_iter = self._comment_entries( + comment_replies_renderer, ytcfg, video_id, + parent=comment.get('id'), comment_counts=comment_counts) + + for reply_comment in comment_entries_iter: + yield reply_comment + + # YouTube comments have a max depth of 2 + max_depth = int_or_none(self._configuration_arg('max_comment_depth', [''])[0]) or float('inf') + if max_depth == 1 and parent: + return + if not comment_counts: + # comment so far, est. total comments, current comment thread # + comment_counts = [0, 0, 0] + + continuation = self._extract_continuation(root_continuation_data) + if continuation and len(continuation['continuation']) < 27: + self.write_debug('Detected old API continuation token. Generating new API compatible token.') + continuation_token = self._generate_comment_continuation(video_id) + continuation = self._build_api_continuation_query(continuation_token, None) + + message = self._get_text(root_continuation_data, ('contents', ..., 'messageRenderer', 'text'), max_runs=1) + if message and not parent: + self.report_warning(message, video_id=video_id) + + visitor_data = None + is_first_continuation = parent is None + + for page_num in itertools.count(0): + if not continuation: + break + headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=visitor_data) + comment_prog_str = '(%d/%d)' % (comment_counts[0], comment_counts[1]) + if page_num == 0: + if is_first_continuation: + note_prefix = 'Downloading comment section API JSON' + else: + note_prefix = ' Downloading comment API JSON reply thread %d %s' % ( + comment_counts[2], comment_prog_str) else: - self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + note_prefix = '%sDownloading comment%s API JSON page %d %s' % ( + ' ' if parent else '', ' replies' if parent else '', + page_num, comment_prog_str) + + response = self._extract_response( + item_id=None, query=continuation, + ep='next', ytcfg=ytcfg, headers=headers, note=note_prefix, + check_get_keys=('onResponseReceivedEndpoints', 'continuationContents')) + if not response: + break + visitor_data = try_get( + response, + lambda x: x['responseContext']['webResponseContextExtensionData']['ytConfigData']['visitorData'], + compat_str) or visitor_data + + continuation_contents = dict_get(response, ('onResponseReceivedEndpoints', 'continuationContents')) + + continuation = None + if isinstance(continuation_contents, list): + for continuation_section in continuation_contents: + if not isinstance(continuation_section, dict): + continue + continuation_items = try_get( + continuation_section, + (lambda x: x['reloadContinuationItemsCommand']['continuationItems'], + lambda x: x['appendContinuationItemsAction']['continuationItems']), + list) or [] + if is_first_continuation: + continuation = extract_header(continuation_items) + is_first_continuation = False + if continuation: + break + continue + count = 0 + for count, entry in enumerate(extract_thread(continuation_items)): + yield entry + continuation = self._extract_continuation({'contents': continuation_items}) + if continuation: + # Sometimes YouTube provides a continuation without any comments + # In most cases we end up just downloading these with very little comments to come. + if count == 0: + if not parent: + self.report_warning('No comments received - assuming end of comments') + continuation = None + break + + # Deprecated response structure + elif isinstance(continuation_contents, dict): + known_continuation_renderers = ('itemSectionContinuation', 'commentRepliesContinuation') + for key, continuation_renderer in continuation_contents.items(): + if key not in known_continuation_renderers: + continue + if not isinstance(continuation_renderer, dict): + continue + if is_first_continuation: + header_continuation_items = [continuation_renderer.get('header') or {}] + continuation = extract_header(header_continuation_items) + is_first_continuation = False + if continuation: + break + + # Sometimes YouTube provides a continuation without any comments + # In most cases we end up just downloading these with very little comments to come. + count = 0 + for count, entry in enumerate(extract_thread(continuation_renderer.get('contents') or {})): + yield entry + continuation = self._extract_continuation(continuation_renderer) + if count == 0: + if not parent: + self.report_warning('No comments received - assuming end of comments') + continuation = None + break + + @staticmethod + def _generate_comment_continuation(video_id): + """ + Generates initial comment section continuation token from given video id + """ + b64_vid_id = base64.b64encode(bytes(video_id.encode('utf-8'))) + parts = ('Eg0SCw==', b64_vid_id, 'GAYyJyIRIgs=', b64_vid_id, 'MAB4AjAAQhBjb21tZW50cy1zZWN0aW9u') + new_continuation_intlist = list(itertools.chain.from_iterable( + [bytes_to_intlist(base64.b64decode(part)) for part in parts])) + return base64.b64encode(intlist_to_bytes(new_continuation_intlist)).decode('utf-8') + + def _get_comments(self, ytcfg, video_id, contents, webpage): + """Entry for comment extraction""" + def _real_comment_extract(contents): + renderer = next(( + item for item in traverse_obj(contents, (..., 'itemSectionRenderer'), default={}) + if item.get('sectionIdentifier') == 'comment-item-section'), None) + yield from self._comment_entries(renderer, ytcfg, video_id) + + max_comments = int_or_none(self._configuration_arg('max_comments', [''])[0]) + # Force English regardless of account setting to prevent parsing issues + # See: https://github.com/hypervideo/hypervideo/issues/532 + ytcfg = copy.deepcopy(ytcfg) + traverse_obj( + ytcfg, ('INNERTUBE_CONTEXT', 'client'), expected_type=dict, default={})['hl'] = 'en' + return itertools.islice(_real_comment_extract(contents), 0, max_comments) + + @staticmethod + def _get_checkok_params(): + return {'contentCheckOk': True, 'racyCheckOk': True} + + @classmethod + def _generate_player_context(cls, sts=None): + context = { + 'html5Preference': 'HTML5_PREF_WANTS', + } + if sts is not None: + context['signatureTimestamp'] = sts + return { + 'playbackContext': { + 'contentPlaybackContext': context + }, + **cls._get_checkok_params() + } - formats = [] - itags = [] - itag_qualities = {} + @staticmethod + def _is_agegated(player_response): + if traverse_obj(player_response, ('playabilityStatus', 'desktopLegacyAgeGateReason')): + return True + + reasons = traverse_obj(player_response, ('playabilityStatus', ('status', 'reason')), default=[]) + AGE_GATE_REASONS = ( + 'confirm your age', 'age-restricted', 'inappropriate', # reason + 'age_verification_required', 'age_check_required', # status + ) + return any(expected in reason for expected in AGE_GATE_REASONS for reason in reasons) + + @staticmethod + def _is_unplayable(player_response): + return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE' + + def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr): + + session_index = self._extract_session_index(player_ytcfg, master_ytcfg) + syncid = self._extract_account_syncid(player_ytcfg, master_ytcfg, initial_pr) + sts = self._extract_signature_timestamp(video_id, player_url, master_ytcfg, fatal=False) if player_url else None + headers = self.generate_api_headers( + ytcfg=player_ytcfg, account_syncid=syncid, session_index=session_index, default_client=client) + + yt_query = {'videoId': video_id} + yt_query.update(self._generate_player_context(sts)) + return self._extract_response( + item_id=video_id, ep='player', query=yt_query, + ytcfg=player_ytcfg, headers=headers, fatal=True, + default_client=client, + note='Downloading %s player API JSON' % client.replace('_', ' ').strip() + ) or None + + def _get_requested_clients(self, url, smuggled_data): + requested_clients = [] + allowed_clients = sorted( + [client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'], + key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True) + for client in self._configuration_arg('player_client'): + if client in allowed_clients: + requested_clients.append(client) + elif client == 'all': + requested_clients.extend(allowed_clients) + else: + self.report_warning(f'Skipping unsupported client {client}') + if not requested_clients: + requested_clients = ['android', 'web'] + + if smuggled_data.get('is_music_url') or self.is_music_url(url): + requested_clients.extend( + f'{client}_music' for client in requested_clients if f'{client}_music' in INNERTUBE_CLIENTS) + + return orderedSet(requested_clients) + + def _extract_player_ytcfg(self, client, video_id): + url = { + 'web_music': 'https://music.youtube.com', + 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1' + }.get(client) + if not url: + return {} + webpage = self._download_webpage(url, video_id, fatal=False, note=f'Downloading {client} config') + return self.extract_ytcfg(video_id, webpage) or {} + + def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg): + initial_pr = None + if webpage: + initial_pr = self._extract_yt_initial_variable( + webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, + video_id, 'initial player response') + + original_clients = clients + clients = clients[::-1] + prs = [] + + def append_client(client_name): + if client_name in INNERTUBE_CLIENTS and client_name not in original_clients: + clients.append(client_name) + + # Android player_response does not have microFormats which are needed for + # extraction of some data. So we return the initial_pr with formats + # stripped out even if not requested by the user + # See: https://github.com/hypervideo/hypervideo/issues/501 + if initial_pr: + pr = dict(initial_pr) + pr['streamingData'] = None + prs.append(pr) + + last_error = None + tried_iframe_fallback = False player_url = None - q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres']) - streaming_data = player_response.get('streamingData') or {} - streaming_formats = streaming_data.get('formats') or [] - streaming_formats.extend(streaming_data.get('adaptiveFormats') or []) + while clients: + client = clients.pop() + player_ytcfg = master_ytcfg if client == 'web' else {} + if 'configs' not in self._configuration_arg('player_skip'): + player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg + + player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage) + require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER') + if 'js' in self._configuration_arg('player_skip'): + require_js_player = False + player_url = None + + if not player_url and not tried_iframe_fallback and require_js_player: + player_url = self._download_player_url(video_id) + tried_iframe_fallback = True + + try: + pr = initial_pr if client == 'web' and initial_pr else self._extract_player_response( + client, video_id, player_ytcfg or master_ytcfg, player_ytcfg, player_url if require_js_player else None, initial_pr) + except ExtractorError as e: + if last_error: + self.report_warning(last_error) + last_error = e + continue + + if pr: + prs.append(pr) + + # creator clients can bypass AGE_VERIFICATION_REQUIRED if logged in + if client.endswith('_agegate') and self._is_unplayable(pr) and self.is_authenticated: + append_client(client.replace('_agegate', '_creator')) + elif self._is_agegated(pr): + append_client(f'{client}_agegate') + + if last_error: + if not len(prs): + raise last_error + self.report_warning(last_error) + return prs, player_url + + def _extract_formats(self, streaming_data, video_id, player_url, is_live): + itags, stream_ids = [], [] + itag_qualities, res_qualities = {}, {} + q = qualities([ + # Normally tiny is the smallest video-only formats. But + # audio-only formats with unknown quality may get tagged as tiny + 'tiny', + 'audio_quality_ultralow', 'audio_quality_low', 'audio_quality_medium', 'audio_quality_high', # Audio only formats + 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres' + ]) + streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[]) + for fmt in streaming_formats: if fmt.get('targetDurationSec') or fmt.get('drmFamilies'): continue itag = str_or_none(fmt.get('itag')) + audio_track = fmt.get('audioTrack') or {} + stream_id = '%s.%s' % (itag or '', audio_track.get('id', '')) + if stream_id in stream_ids: + continue + quality = fmt.get('quality') - if itag and quality: - itag_qualities[itag] = quality + height = int_or_none(fmt.get('height')) + if quality == 'tiny' or not quality: + quality = fmt.get('audioQuality', '').lower() or quality + # The 3gp format (17) in android client has a quality of "small", + # but is actually worse than other formats + if itag == '17': + quality = 'tiny' + if quality: + if itag: + itag_qualities[itag] = quality + if height: + res_qualities[height] = quality # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment # (adding `&sq=0` to the URL) and parsing emsg box to determine the # number of fragment that would subsequently requested with (`&sq=N`) @@ -1613,12 +2435,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if not (sc and fmt_url and encrypted_sig): continue if not player_url: - if not webpage: - continue - player_url = self._search_regex( - r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"', - webpage, 'player URL', fatal=False) - if not player_url: continue signature = self._decrypt_signature(sc['s'][0], video_id, player_url) sp = try_get(sc, lambda x: x['sp'][0]) or 'signature' @@ -1626,27 +2442,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if itag: itags.append(itag) + stream_ids.append(stream_id) + tbr = float_or_none( fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) dct = { 'asr': int_or_none(fmt.get('audioSampleRate')), 'filesize': int_or_none(fmt.get('contentLength')), 'format_id': itag, - 'format_note': fmt.get('qualityLabel') or quality, + 'format_note': ', '.join(filter(None, ( + '%s%s' % (audio_track.get('displayName') or '', + ' (default)' if audio_track.get('audioIsDefault') else ''), + fmt.get('qualityLabel') or quality.replace('audio_quality_', '')))), 'fps': int_or_none(fmt.get('fps')), - 'height': int_or_none(fmt.get('height')), + 'height': height, 'quality': q(quality), 'tbr': tbr, 'url': fmt_url, - 'width': fmt.get('width'), + 'width': int_or_none(fmt.get('width')), + 'language': audio_track.get('id', '').split('.')[0], + 'language_preference': 1 if audio_track.get('audioIsDefault') else -1, } - mimetype = fmt.get('mimeType') - if mimetype: - mobj = re.match( - r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype) - if mobj: - dct['ext'] = mimetype2ext(mobj.group(1)) - dct.update(parse_codecs(mobj.group(2))) + mime_mobj = re.match( + r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '') + if mime_mobj: + dct['ext'] = mimetype2ext(mime_mobj.group(1)) + dct.update(parse_codecs(mime_mobj.group(2))) no_audio = dct.get('acodec') == 'none' no_video = dct.get('vcodec') == 'none' if no_audio: @@ -1660,61 +2481,181 @@ class YoutubeIE(YoutubeBaseInfoExtractor): } if dct.get('ext'): dct['container'] = dct['ext'] + '_dash' - formats.append(dct) - - hls_manifest_url = streaming_data.get('hlsManifestUrl') - if hls_manifest_url: - for f in self._extract_m3u8_formats( - hls_manifest_url, video_id, 'mp4', fatal=False): - itag = self._search_regex( - r'/itag/(\d+)', f['url'], 'itag', default=None) - if itag: - f['format_id'] = itag - formats.append(f) + yield dct + + skip_manifests = self._configuration_arg('skip') + get_dash = ( + (not is_live or self._configuration_arg('include_live_dash')) + and 'dash' not in skip_manifests and self.get_param('youtube_include_dash_manifest', True)) + get_hls = 'hls' not in skip_manifests and self.get_param('youtube_include_hls_manifest', True) + + def guess_quality(f): + for val, qdict in ((f.get('format_id'), itag_qualities), (f.get('height'), res_qualities)): + if val in qdict: + return q(qdict[val]) + return -1 + + for sd in streaming_data: + hls_manifest_url = get_hls and sd.get('hlsManifestUrl') + if hls_manifest_url: + for f in self._extract_m3u8_formats(hls_manifest_url, video_id, 'mp4', fatal=False): + itag = self._search_regex( + r'/itag/(\d+)', f['url'], 'itag', default=None) + if itag in itags: + itag += '-hls' + if itag in itags: + continue + if itag: + f['format_id'] = itag + itags.append(itag) + f['quality'] = guess_quality(f) + yield f - if self._downloader.params.get('youtube_include_dash_manifest', True): - dash_manifest_url = streaming_data.get('dashManifestUrl') + dash_manifest_url = get_dash and sd.get('dashManifestUrl') if dash_manifest_url: - for f in self._extract_mpd_formats( - dash_manifest_url, video_id, fatal=False): + for f in self._extract_mpd_formats(dash_manifest_url, video_id, fatal=False): itag = f['format_id'] if itag in itags: - continue - if itag in itag_qualities: - f['quality'] = q(itag_qualities[itag]) + itag += '-dash' + if itag in itags: + continue + if itag: + f['format_id'] = itag + itags.append(itag) + f['quality'] = guess_quality(f) filesize = int_or_none(self._search_regex( r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None)) if filesize: f['filesize'] = filesize - formats.append(f) + yield f + + def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + video_id = self._match_id(url) + + base_url = self.http_scheme() + '//www.youtube.com/' + webpage_url = base_url + 'watch?v=' + video_id + webpage = None + if 'webpage' not in self._configuration_arg('player_skip'): + webpage = self._download_webpage( + webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False) + + master_ytcfg = self.extract_ytcfg(video_id, webpage) or self._get_default_ytcfg() + + player_responses, player_url = self._extract_player_responses( + self._get_requested_clients(url, smuggled_data), + video_id, webpage, master_ytcfg) + + get_first = lambda obj, keys, **kwargs: traverse_obj(obj, (..., *variadic(keys)), **kwargs, get_all=False) + + playability_statuses = traverse_obj( + player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[]) + + trailer_video_id = get_first( + playability_statuses, + ('errorScreen', 'playerLegacyDesktopYpcTrailerRenderer', 'trailerVideoId'), + expected_type=str) + if trailer_video_id: + return self.url_result( + trailer_video_id, self.ie_key(), trailer_video_id) + + search_meta = ((lambda x: self._html_search_meta(x, webpage, default=None)) + if webpage else (lambda x: None)) + + video_details = traverse_obj( + player_responses, (..., 'videoDetails'), expected_type=dict, default=[]) + microformats = traverse_obj( + player_responses, (..., 'microformat', 'playerMicroformatRenderer'), + expected_type=dict, default=[]) + video_title = ( + get_first(video_details, 'title') + or self._get_text(microformats, (..., 'title')) + or search_meta(['og:title', 'twitter:title', 'title'])) + video_description = get_first(video_details, 'shortDescription') + + if not smuggled_data.get('force_singlefeed', False): + if not self.get_param('noplaylist'): + multifeed_metadata_list = get_first( + player_responses, + ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'), + expected_type=str) + if multifeed_metadata_list: + entries = [] + feed_ids = [] + for feed in multifeed_metadata_list.split(','): + # Unquote should take place before split on comma (,) since textual + # fields may contain comma as well (see + # https://github.com/ytdl-org/youtube-dl/issues/8536) + feed_data = compat_parse_qs( + compat_urllib_parse_unquote_plus(feed)) + + def feed_entry(name): + return try_get( + feed_data, lambda x: x[name][0], compat_str) + + feed_id = feed_entry('id') + if not feed_id: + continue + feed_title = feed_entry('title') + title = video_title + if feed_title: + title += ' (%s)' % feed_title + entries.append({ + '_type': 'url_transparent', + 'ie_key': 'Youtube', + 'url': smuggle_url( + '%swatch?v=%s' % (base_url, feed_data['id'][0]), + {'force_singlefeed': True}), + 'title': title, + }) + feed_ids.append(feed_id) + self.to_screen( + 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' + % (', '.join(feed_ids), video_id)) + return self.playlist_result( + entries, video_id, video_title, video_description) + else: + self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + + live_broadcast_details = traverse_obj(microformats, (..., 'liveBroadcastDetails')) + is_live = get_first(video_details, 'isLive') + if is_live is None: + is_live = get_first(live_broadcast_details, 'isLiveNow') + + streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[]) + formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live)) if not formats: - if streaming_data.get('licenseInfos'): - raise ExtractorError( - 'This video is DRM protected.', expected=True) - pemr = try_get( - playability_status, - lambda x: x['errorScreen']['playerErrorMessageRenderer'], - dict) or {} - reason = get_text(pemr.get('reason')) or playability_status.get('reason') - subreason = pemr.get('subreason') + if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')): + self.report_drm(video_id) + pemr = get_first( + playability_statuses, + ('errorScreen', 'playerErrorMessageRenderer'), expected_type=dict) or {} + reason = self._get_text(pemr, 'reason') or get_first(playability_statuses, 'reason') + subreason = clean_html(self._get_text(pemr, 'subreason') or '') if subreason: - subreason = clean_html(get_text(subreason)) if subreason == 'The uploader has not made this video available in your country.': - countries = microformat.get('availableCountries') + countries = get_first(microformats, 'availableCountries') if not countries: regions_allowed = search_meta('regionsAllowed') countries = regions_allowed.split(',') if regions_allowed else None - self.raise_geo_restricted( - subreason, countries) - reason += '\n' + subreason + self.raise_geo_restricted(subreason, countries, metadata_available=True) + reason += f'. {subreason}' if reason: - raise ExtractorError(reason, expected=True) + self.raise_no_formats(reason, expected=True) - self._sort_formats(formats) + for f in formats: + if '&c=WEB&' in f['url'] and '&ratebypass=yes&' not in f['url']: # throttled + f['source_preference'] = -10 + # TODO: this method is not reliable + f['format_note'] = format_field(f, 'format_note', '%s ') + '(maybe throttled)' - keywords = video_details.get('keywords') or [] + # Source is given priority since formats that throttle are given lower source_preference + # When throttling issue is fully fixed, remove this + self._sort_formats(formats, ('quality', 'res', 'fps', 'hdr:12', 'source', 'codec:vp9.2', 'lang')) + + keywords = get_first(video_details, 'keywords', expected_type=list) or [] if not keywords and webpage: keywords = [ unescapeHTML(m.group('content')) @@ -1733,35 +2674,71 @@ class YoutubeIE(YoutubeBaseInfoExtractor): break thumbnails = [] - for container in (video_details, microformat): - for thumbnail in (try_get( - container, - lambda x: x['thumbnail']['thumbnails'], list) or []): - thumbnail_url = thumbnail.get('url') - if not thumbnail_url: - continue - thumbnails.append({ - 'height': int_or_none(thumbnail.get('height')), - 'url': thumbnail_url, - 'width': int_or_none(thumbnail.get('width')), - }) - if thumbnails: - break - else: - thumbnail = search_meta(['og:image', 'twitter:image']) - if thumbnail: - thumbnails = [{'url': thumbnail}] - - category = microformat.get('category') or search_meta('genre') - channel_id = video_details.get('channelId') \ - or microformat.get('externalChannelId') \ - or search_meta('channelId') + thumbnail_dicts = traverse_obj( + (video_details, microformats), (..., ..., 'thumbnail', 'thumbnails', ...), + expected_type=dict, default=[]) + for thumbnail in thumbnail_dicts: + thumbnail_url = thumbnail.get('url') + if not thumbnail_url: + continue + # Sometimes youtube gives a wrong thumbnail URL. See: + # https://github.com/hypervideo/hypervideo/issues/233 + # https://github.com/ytdl-org/youtube-dl/issues/28023 + if 'maxresdefault' in thumbnail_url: + thumbnail_url = thumbnail_url.split('?')[0] + thumbnails.append({ + 'url': thumbnail_url, + 'height': int_or_none(thumbnail.get('height')), + 'width': int_or_none(thumbnail.get('width')), + }) + thumbnail_url = search_meta(['og:image', 'twitter:image']) + if thumbnail_url: + thumbnails.append({ + 'url': thumbnail_url, + }) + # The best resolution thumbnails sometimes does not appear in the webpage + # See: https://github.com/ytdl-org/youtube-dl/issues/29049, https://github.com/hypervideo/hypervideo/issues/340 + # List of possible thumbnails - Ref: <https://stackoverflow.com/a/20542029> + thumbnail_names = [ + 'maxresdefault', 'hq720', 'sddefault', 'sd1', 'sd2', 'sd3', + 'hqdefault', 'hq1', 'hq2', 'hq3', '0', + 'mqdefault', 'mq1', 'mq2', 'mq3', + 'default', '1', '2', '3' + ] + n_thumbnail_names = len(thumbnail_names) + + thumbnails.extend({ + 'url': 'https://i.ytimg.com/vi{webp}/{video_id}/{name}{live}.{ext}'.format( + video_id=video_id, name=name, ext=ext, + webp='_webp' if ext == 'webp' else '', live='_live' if is_live else ''), + } for name in thumbnail_names for ext in ('webp', 'jpg')) + for thumb in thumbnails: + i = next((i for i, t in enumerate(thumbnail_names) if f'/{video_id}/{t}' in thumb['url']), n_thumbnail_names) + thumb['preference'] = (0 if '.webp' in thumb['url'] else -1) - (2 * i) + self._remove_duplicate_formats(thumbnails) + + category = get_first(microformats, 'category') or search_meta('genre') + channel_id = str_or_none( + get_first(video_details, 'channelId') + or get_first(microformats, 'externalChannelId') + or search_meta('channelId')) duration = int_or_none( - video_details.get('lengthSeconds') - or microformat.get('lengthSeconds')) \ - or parse_duration(search_meta('duration')) - is_live = video_details.get('isLive') - owner_profile_url = microformat.get('ownerProfileUrl') + get_first(video_details, 'lengthSeconds') + or get_first(microformats, 'lengthSeconds') + or parse_duration(search_meta('duration'))) or None + owner_profile_url = get_first(microformats, 'ownerProfileUrl') + + live_content = get_first(video_details, 'isLiveContent') + is_upcoming = get_first(video_details, 'isUpcoming') + if is_live is None: + if is_upcoming or live_content is False: + is_live = False + if is_upcoming is None and (live_content or is_live): + is_upcoming = False + live_starttime = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp')) + live_endtime = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp')) + if not duration and live_endtime and live_starttime: + duration = live_endtime - live_starttime info = { 'id': video_id, @@ -1770,35 +2747,50 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'thumbnails': thumbnails, 'description': video_description, 'upload_date': unified_strdate( - microformat.get('uploadDate') + get_first(microformats, 'uploadDate') or search_meta('uploadDate')), - 'uploader': video_details['author'], + 'uploader': get_first(video_details, 'author'), 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None, 'uploader_url': owner_profile_url, 'channel_id': channel_id, - 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None, + 'channel_url': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None, 'duration': duration, 'view_count': int_or_none( - video_details.get('viewCount') - or microformat.get('viewCount') + get_first((video_details, microformats), (..., 'viewCount')) or search_meta('interactionCount')), - 'average_rating': float_or_none(video_details.get('averageRating')), + 'average_rating': float_or_none(get_first(video_details, 'averageRating')), 'age_limit': 18 if ( - microformat.get('isFamilySafe') is False + get_first(microformats, 'isFamilySafe') is False or search_meta('isFamilyFriendly') == 'false' or search_meta('og:restrictions:age') == '18+') else 0, 'webpage_url': webpage_url, 'categories': [category] if category else None, 'tags': keywords, + 'playable_in_embed': get_first(playability_statuses, 'playableInEmbed'), 'is_live': is_live, + 'was_live': (False if is_live or is_upcoming or live_content is False + else None if is_live is None or is_upcoming is None + else live_content), + 'live_status': 'is_upcoming' if is_upcoming else None, # rest will be set by YoutubeDL + 'release_timestamp': live_starttime, } - pctr = try_get( - player_response, - lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict) + pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict) if pctr: - def process_language(container, base_url, lang_code, query): - lang_subs = [] + def get_lang_code(track): + return (remove_start(track.get('vssId') or '', '.').replace('.', '-') + or track.get('languageCode')) + + # Converted into dicts to remove duplicates + captions = { + get_lang_code(sub): sub + for sub in traverse_obj(pctr, (..., 'captionTracks', ...), default=[])} + translation_languages = { + lang.get('languageCode'): self._get_text(lang.get('languageName'), max_runs=1) + for lang in traverse_obj(pctr, (..., 'translationLanguages', ...), default=[])} + + def process_language(container, base_url, lang_code, sub_name, query): + lang_subs = container.setdefault(lang_code, []) for fmt in self._SUBTITLE_FORMATS: query.update({ 'fmt': fmt, @@ -1806,30 +2798,31 @@ class YoutubeIE(YoutubeBaseInfoExtractor): lang_subs.append({ 'ext': fmt, 'url': update_url_query(base_url, query), + 'name': sub_name, }) - container[lang_code] = lang_subs - subtitles = {} - for caption_track in (pctr.get('captionTracks') or []): + subtitles, automatic_captions = {}, {} + for lang_code, caption_track in captions.items(): base_url = caption_track.get('baseUrl') if not base_url: continue + lang_name = self._get_text(caption_track, 'name', max_runs=1) if caption_track.get('kind') != 'asr': - lang_code = caption_track.get('languageCode') if not lang_code: continue process_language( - subtitles, base_url, lang_code, {}) - continue - automatic_captions = {} - for translation_language in (pctr.get('translationLanguages') or []): - translation_language_code = translation_language.get('languageCode') - if not translation_language_code: + subtitles, base_url, lang_code, lang_name, {}) + if not caption_track.get('isTranslatable'): + continue + for trans_code, trans_name in translation_languages.items(): + if not trans_code: continue + if caption_track.get('kind') != 'asr': + trans_code += f'-{lang_code}' + trans_name += format_field(lang_name, template=' from %s') process_language( - automatic_captions, base_url, translation_language_code, - {'tlang': translation_language_code}) - info['automatic_captions'] = automatic_captions + automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code}) + info['automatic_captions'] = automatic_captions info['subtitles'] = subtitles parsed_url = compat_urllib_parse_urlparse(url) @@ -1841,6 +2834,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if d_k not in info and k in s_ks: info[d_k] = parse_duration(query[k][0]) + # Youtube Music Auto-generated description if video_description: mobj = re.search(r'(?s)(?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?.+\nAuto-generated by YouTube\.\s*$', video_description) if mobj: @@ -1864,42 +2858,31 @@ class YoutubeIE(YoutubeBaseInfoExtractor): webpage, self._YT_INITIAL_DATA_RE, video_id, 'yt initial data') if not initial_data: - initial_data = self._call_api( - 'next', {'videoId': video_id}, video_id, fatal=False) + query = {'videoId': video_id} + query.update(self._get_checkok_params()) + initial_data = self._extract_response( + item_id=video_id, ep='next', fatal=False, + ytcfg=master_ytcfg, query=query, + headers=self.generate_api_headers(ytcfg=master_ytcfg), + note='Downloading initial data API JSON') - if initial_data: - chapters = self._extract_chapters_from_json( - initial_data, video_id, duration) - if not chapters: - for engagment_pannel in (initial_data.get('engagementPanels') or []): - contents = try_get( - engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'], - list) - if not contents: - continue + try: + # This will error if there is no livechat + initial_data['contents']['twoColumnWatchNextResults']['conversationBar']['liveChatRenderer']['continuations'][0]['reloadContinuationData']['continuation'] + info.setdefault('subtitles', {})['live_chat'] = [{ + 'url': 'https://www.youtube.com/watch?v=%s' % video_id, # url is needed to set cookies + 'video_id': video_id, + 'ext': 'json', + 'protocol': 'youtube_live_chat' if is_live or is_upcoming else 'youtube_live_chat_replay', + }] + except (KeyError, IndexError, TypeError): + pass - def chapter_time(mmlir): - return parse_duration( - get_text(mmlir.get('timeDescription'))) - - chapters = [] - for next_num, content in enumerate(contents, start=1): - mmlir = content.get('macroMarkersListItemRenderer') or {} - start_time = chapter_time(mmlir) - end_time = chapter_time(try_get( - contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \ - if next_num < len(contents) else duration - if start_time is None or end_time is None: - continue - chapters.append({ - 'start_time': start_time, - 'end_time': end_time, - 'title': get_text(mmlir.get('title')), - }) - if chapters: - break - if chapters: - info['chapters'] = chapters + if initial_data: + info['chapters'] = ( + self._extract_chapters_from_json(initial_data, duration) + or self._extract_chapters_from_engagement_panel(initial_data, duration) + or None) contents = try_get( initial_data, @@ -1910,7 +2893,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if vpir: stl = vpir.get('superTitleLink') if stl: - stl = get_text(stl) + stl = self._get_text(stl) if try_get( vpir, lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN': @@ -1950,10 +2933,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }) vsir = content.get('videoSecondaryInfoRenderer') if vsir: - info['channel'] = get_text(try_get( - vsir, - lambda x: x['owner']['videoOwnerRenderer']['title'], - dict)) + info['channel'] = self._get_text(vsir, ('owner', 'videoOwnerRenderer', 'title')) rows = try_get( vsir, lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'], @@ -1968,8 +2948,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): mrr_title = mrr.get('title') if not mrr_title: continue - mrr_title = get_text(mrr['title']) - mrr_contents_text = get_text(mrr['contents'][0]) + mrr_title = self._get_text(mrr, 'title') + mrr_contents_text = self._get_text(mrr, ('contents', 0)) if mrr_title == 'License': info['license'] = mrr_contents_text elif not multiple_songs: @@ -1980,12 +2960,51 @@ class YoutubeIE(YoutubeBaseInfoExtractor): elif mrr_title == 'Song': info['track'] = mrr_contents_text + fallbacks = { + 'channel': 'uploader', + 'channel_id': 'uploader_id', + 'channel_url': 'uploader_url', + } + for to, frm in fallbacks.items(): + if not info.get(to): + info[to] = info.get(frm) + for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]: v = info.get(s_k) if v: info[d_k] = v - self.mark_watched(video_id, player_response) + is_private = get_first(video_details, 'isPrivate', expected_type=bool) + is_unlisted = get_first(microformats, 'isUnlisted', expected_type=bool) + is_membersonly = None + is_premium = None + if initial_data and is_private is not None: + is_membersonly = False + is_premium = False + contents = try_get(initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or [] + badge_labels = set() + for content in contents: + if not isinstance(content, dict): + continue + badge_labels.update(self._extract_badges(content.get('videoPrimaryInfoRenderer'))) + for badge_label in badge_labels: + if badge_label.lower() == 'members only': + is_membersonly = True + elif badge_label.lower() == 'premium': + is_premium = True + elif badge_label.lower() == 'unlisted': + is_unlisted = True + + info['availability'] = self._availability( + is_private=is_private, + needs_premium=is_premium, + needs_subscription=is_membersonly, + needs_auth=info['age_limit'] >= 18, + is_unlisted=None if is_private is None else is_unlisted) + + info['__post_extractor'] = self.extract_comments(master_ytcfg, video_id, contents, webpage) + + self.mark_watched(video_id, player_responses) return info @@ -2000,127 +3019,161 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): invidio\.us )/ (?: - (?:channel|c|user|feed|hashtag)/| - (?:playlist|watch)\?.*?\blist=| - (?!(?:watch|embed|v|e)\b) + (?P<channel_type>channel|c|user|browse)/| + (?P<not_channel> + feed/|hashtag/| + (?:playlist|watch)\?.*?\blist= + )| + (?!(?:%s)\b) # Direct URLs ) (?P<id>[^/?\#&]+) - ''' + ''' % YoutubeBaseInfoExtractor._RESERVED_NAMES IE_NAME = 'youtube:tab' _TESTS = [{ - # playlists, multipage + 'note': 'playlists, multipage', 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid', 'playlist_mincount': 94, 'info_dict': { 'id': 'UCqj7Cz7revf5maW9g5pgNcg', 'title': 'Игорь Клейнер - Playlists', 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2', + 'uploader': 'Игорь Клейнер', + 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg', }, }, { - # playlists, multipage, different order + 'note': 'playlists, multipage, different order', 'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd', 'playlist_mincount': 94, 'info_dict': { 'id': 'UCqj7Cz7revf5maW9g5pgNcg', 'title': 'Игорь Клейнер - Playlists', 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2', + 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'uploader': 'Игорь Клейнер', }, }, { - # playlists, series + 'note': 'playlists, series', 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3', 'playlist_mincount': 5, 'info_dict': { 'id': 'UCYO_jab_esuFRV4b17AJtAw', 'title': '3Blue1Brown - Playlists', 'description': 'md5:e1384e8a133307dd10edee76e875d62f', + 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw', + 'uploader': '3Blue1Brown', }, }, { - # playlists, singlepage + 'note': 'playlists, singlepage', 'url': 'https://www.youtube.com/user/ThirstForScience/playlists', 'playlist_mincount': 4, 'info_dict': { 'id': 'UCAEtajcuhQ6an9WEzY9LEMQ', 'title': 'ThirstForScience - Playlists', 'description': 'md5:609399d937ea957b0f53cbffb747a14c', + 'uploader': 'ThirstForScience', + 'uploader_id': 'UCAEtajcuhQ6an9WEzY9LEMQ', } }, { 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists', 'only_matching': True, }, { - # basic, single video playlist + 'note': 'basic, single video playlist', 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', 'info_dict': { 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA', 'uploader': 'Sergey M.', 'id': 'PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', - 'title': 'hypervideo public playlist', + 'title': 'youtube-dl public playlist', }, 'playlist_count': 1, }, { - # empty playlist + 'note': 'empty playlist', 'url': 'https://www.youtube.com/playlist?list=PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf', 'info_dict': { 'uploader_id': 'UCmlqkdCBesrv2Lak1mF_MxA', 'uploader': 'Sergey M.', 'id': 'PL4lCao7KL_QFodcLWhDpGCYnngnHtQ-Xf', - 'title': 'hypervideo empty playlist', + 'title': 'youtube-dl empty playlist', }, 'playlist_count': 0, }, { - # Home tab + 'note': 'Home tab', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/featured', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Home', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', }, 'playlist_mincount': 2, }, { - # Videos tab + 'note': 'Videos tab', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Videos', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', }, 'playlist_mincount': 975, }, { - # Videos tab, sorted by popular + 'note': 'Videos tab, sorted by popular', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/videos?view=0&sort=p&flow=grid', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Videos', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', }, 'playlist_mincount': 199, }, { - # Playlists tab + 'note': 'Playlists tab', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/playlists', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Playlists', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', }, 'playlist_mincount': 17, }, { - # Community tab + 'note': 'Community tab', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/community', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Community', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', }, 'playlist_mincount': 18, }, { - # Channels tab + 'note': 'Channels tab', 'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w/channels', 'info_dict': { 'id': 'UCKfVa3S1e4PHvxWcwyMMg8w', 'title': 'lex will - Channels', 'description': 'md5:2163c5d0ff54ed5f598d6a7e6211e488', + 'uploader': 'lex will', + 'uploader_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + }, + 'playlist_mincount': 12, + }, { + 'note': 'Search tab', + 'url': 'https://www.youtube.com/c/3blue1brown/search?query=linear%20algebra', + 'playlist_mincount': 40, + 'info_dict': { + 'id': 'UCYO_jab_esuFRV4b17AJtAw', + 'title': '3Blue1Brown - Search - linear algebra', + 'description': 'md5:e1384e8a133307dd10edee76e875d62f', + 'uploader': '3Blue1Brown', + 'uploader_id': 'UCYO_jab_esuFRV4b17AJtAw', }, - 'playlist_mincount': 138, }, { 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA', 'only_matching': True, @@ -2138,6 +3191,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', 'uploader': 'Christiaan008', 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg', + 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268', }, 'playlist_count': 96, }, { @@ -2151,7 +3205,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): }, 'playlist_mincount': 1123, }, { - # even larger playlist, 8832 videos + 'note': 'even larger playlist, 8832 videos', 'url': 'http://www.youtube.com/user/NASAgovVideo/videos', 'only_matching': True, }, { @@ -2165,20 +3219,41 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): }, 'playlist_mincount': 21, }, { - # https://github.com/ytdl-org/youtube-dl/issues/21844 + 'note': 'Playlist with "show unavailable videos" button', + 'url': 'https://www.youtube.com/playlist?list=UUTYLiWFZy8xtPwxFwX9rV7Q', + 'info_dict': { + 'title': 'Uploads from Phim Siêu Nhân Nhật Bản', + 'id': 'UUTYLiWFZy8xtPwxFwX9rV7Q', + 'uploader': 'Phim Siêu Nhân Nhật Bản', + 'uploader_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q', + }, + 'playlist_mincount': 200, + }, { + 'note': 'Playlist with unavailable videos in page 7', + 'url': 'https://www.youtube.com/playlist?list=UU8l9frL61Yl5KFOl87nIm2w', + 'info_dict': { + 'title': 'Uploads from BlankTV', + 'id': 'UU8l9frL61Yl5KFOl87nIm2w', + 'uploader': 'BlankTV', + 'uploader_id': 'UC8l9frL61Yl5KFOl87nIm2w', + }, + 'playlist_mincount': 1000, + }, { + 'note': 'https://github.com/ytdl-org/youtube-dl/issues/21844', 'url': 'https://www.youtube.com/playlist?list=PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba', 'info_dict': { 'title': 'Data Analysis with Dr Mike Pound', 'id': 'PLzH6n4zXuckpfMu_4Ff8E7Z1behQks5ba', 'uploader_id': 'UC9-y-6csu5WGm29I7JiwpnA', 'uploader': 'Computerphile', + 'description': 'md5:7f567c574d13d3f8c0954d9ffee4e487', }, 'playlist_mincount': 11, }, { 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', 'only_matching': True, }, { - # Playlist URL that does not actually serve a playlist + 'note': 'Playlist URL that does not actually serve a playlist', 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4', 'info_dict': { 'id': 'FqZTN594JQw', @@ -2210,14 +3285,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): }, { 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live', 'info_dict': { - 'id': '9Auq9mYxFEE', + 'id': '3yImotZU3tw', # This will keep changing 'ext': 'mp4', - 'title': 'Watch Sky News live', + 'title': compat_str, 'uploader': 'Sky News', 'uploader_id': 'skynews', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews', - 'upload_date': '20191102', - 'description': 'md5:78de4e1c2359d0ea3ed829678e38b662', + 'upload_date': r're:\d{8}', + 'description': compat_str, 'categories': ['News & Politics'], 'tags': list, 'like_count': int, @@ -2226,6 +3301,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'expected_warnings': ['Downloading just video ', 'Ignoring subtitle tracks found in '], }, { 'url': 'https://www.youtube.com/user/TheYoungTurks/live', 'info_dict': { @@ -2254,30 +3330,30 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'url': 'https://www.youtube.com/c/CommanderVideoHq/live', 'only_matching': True, }, { + 'note': 'A channel that is not live. Should raise error', + 'url': 'https://www.youtube.com/user/numberphile/live', + 'only_matching': True, + }, { 'url': 'https://www.youtube.com/feed/trending', 'only_matching': True, }, { - # needs auth 'url': 'https://www.youtube.com/feed/library', 'only_matching': True, }, { - # needs auth 'url': 'https://www.youtube.com/feed/history', 'only_matching': True, }, { - # needs auth 'url': 'https://www.youtube.com/feed/subscriptions', 'only_matching': True, }, { - # needs auth 'url': 'https://www.youtube.com/feed/watch_later', 'only_matching': True, }, { - # no longer available? + 'note': 'Recommended - redirects to home page.', 'url': 'https://www.youtube.com/feed/recommended', 'only_matching': True, }, { - # inline playlist with not always working continuations + 'note': 'inline playlist with not always working continuations', 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C', 'only_matching': True, }, { @@ -2305,6 +3381,116 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): }, { 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU', 'only_matching': True, + }, { + 'note': 'Requires Premium: should request additional YTM-info webpage (and have format 141) for videos in playlist', + 'url': 'https://music.youtube.com/playlist?list=PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq', + 'only_matching': True + }, { + 'note': '/browse/ should redirect to /channel/', + 'url': 'https://music.youtube.com/browse/UC1a8OFewdjuLq6KlF8M_8Ng', + 'only_matching': True + }, { + 'note': 'VLPL, should redirect to playlist?list=PL...', + 'url': 'https://music.youtube.com/browse/VLPLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq', + 'info_dict': { + 'id': 'PLRBp0Fe2GpgmgoscNFLxNyBVSFVdYmFkq', + 'uploader': 'NoCopyrightSounds', + 'description': 'Providing you with copyright free / safe music for gaming, live streaming, studying and more!', + 'uploader_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg', + 'title': 'NCS Releases', + }, + 'playlist_mincount': 166, + }, { + 'note': 'Topic, should redirect to playlist?list=UU...', + 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw', + 'info_dict': { + 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw', + 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw', + 'title': 'Uploads from Royalty Free Music - Topic', + 'uploader': 'Royalty Free Music - Topic', + }, + 'expected_warnings': [ + 'A channel/user page was given', + 'The URL does not have a videos tab', + ], + 'playlist_mincount': 101, + }, { + 'note': 'Topic without a UU playlist', + 'url': 'https://www.youtube.com/channel/UCtFRv9O2AHqOZjjynzrv-xg', + 'info_dict': { + 'id': 'UCtFRv9O2AHqOZjjynzrv-xg', + 'title': 'UCtFRv9O2AHqOZjjynzrv-xg', + }, + 'expected_warnings': [ + 'A channel/user page was given', + 'The URL does not have a videos tab', + 'Falling back to channel URL', + ], + 'playlist_mincount': 9, + }, { + 'note': 'Youtube music Album', + 'url': 'https://music.youtube.com/browse/MPREb_gTAcphH99wE', + 'info_dict': { + 'id': 'OLAK5uy_l1m0thk3g31NmIIz_vMIbWtyv7eZixlH0', + 'title': 'Album - Royalty Free Music Library V2 (50 Songs)', + }, + 'playlist_count': 50, + }, { + 'note': 'unlisted single video playlist', + 'url': 'https://www.youtube.com/playlist?list=PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf', + 'info_dict': { + 'uploader_id': 'UC9zHu_mHU96r19o-wV5Qs1Q', + 'uploader': 'colethedj', + 'id': 'PLwL24UFy54GrB3s2KMMfjZscDi1x5Dajf', + 'title': 'hypervideo unlisted playlist test', + 'availability': 'unlisted' + }, + 'playlist_count': 1, + }, { + 'note': 'API Fallback: Recommended - redirects to home page. Requires visitorData', + 'url': 'https://www.youtube.com/feed/recommended', + 'info_dict': { + 'id': 'recommended', + 'title': 'recommended', + }, + 'playlist_mincount': 50, + 'params': { + 'skip_download': True, + 'extractor_args': {'youtubetab': {'skip': ['webpage']}} + }, + }, { + 'note': 'API Fallback: /videos tab, sorted by oldest first', + 'url': 'https://www.youtube.com/user/theCodyReeder/videos?view=0&sort=da&flow=grid', + 'info_dict': { + 'id': 'UCu6mSoMNzHQiBIOCkHUa2Aw', + 'title': 'Cody\'sLab - Videos', + 'description': 'md5:d083b7c2f0c67ee7a6c74c3e9b4243fa', + 'uploader': 'Cody\'sLab', + 'uploader_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw', + }, + 'playlist_mincount': 650, + 'params': { + 'skip_download': True, + 'extractor_args': {'youtubetab': {'skip': ['webpage']}} + }, + }, { + 'note': 'API Fallback: Topic, should redirect to playlist?list=UU...', + 'url': 'https://music.youtube.com/browse/UC9ALqqC4aIeG5iDs7i90Bfw', + 'info_dict': { + 'id': 'UU9ALqqC4aIeG5iDs7i90Bfw', + 'uploader_id': 'UC9ALqqC4aIeG5iDs7i90Bfw', + 'title': 'Uploads from Royalty Free Music - Topic', + 'uploader': 'Royalty Free Music - Topic', + }, + 'expected_warnings': [ + 'A channel/user page was given', + 'The URL does not have a videos tab', + ], + 'playlist_mincount': 101, + 'params': { + 'skip_download': True, + 'extractor_args': {'youtubetab': {'skip': ['webpage']}} + }, }] @classmethod @@ -2326,25 +3512,28 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): channel_url, 'channel id') @staticmethod - def _extract_grid_item_renderer(item): - assert isinstance(item, dict) + def _extract_basic_item_renderer(item): + # Modified from _extract_grid_item_renderer + known_basic_renderers = ( + 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer' + ) for key, renderer in item.items(): - if not key.startswith('grid') or not key.endswith('Renderer'): - continue if not isinstance(renderer, dict): continue - return renderer + elif key in known_basic_renderers: + return renderer + elif key.startswith('grid') and key.endswith('Renderer'): + return renderer def _grid_entries(self, grid_renderer): for item in grid_renderer['items']: if not isinstance(item, dict): continue - renderer = self._extract_grid_item_renderer(item) + renderer = self._extract_basic_item_renderer(item) if not isinstance(renderer, dict): continue - title = try_get( - renderer, (lambda x: x['title']['runs'][0]['text'], - lambda x: x['title']['simpleText']), compat_str) + title = self._get_text(renderer, 'title') + # playlist playlist_id = renderer.get('playlistId') if playlist_id: @@ -2361,8 +3550,6 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): # channel channel_id = renderer.get('channelId') if channel_id: - title = try_get( - renderer, lambda x: x['title']['simpleText'], compat_str) yield self.url_result( 'https://www.youtube.com/channel/%s' % channel_id, ie=YoutubeTabIE.ie_key(), video_title=title) @@ -2382,7 +3569,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): content = shelf_renderer.get('content') if not isinstance(content, dict): return - renderer = content.get('gridRenderer') + renderer = content.get('gridRenderer') or content.get('expandedShelfContentsRenderer') if renderer: # TODO: add support for nested playlists so each shelf is processed # as separate playlist @@ -2405,8 +3592,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): # will not work if skip_channels and '/channels?' in shelf_url: return - title = try_get( - shelf_renderer, lambda x: x['title']['runs'][0]['text'], compat_str) + title = self._get_text(shelf_renderer, 'title') yield self.url_result(shelf_url, video_title=title) # Shelf may not contain shelf URL, fallback to extraction from content for entry in self._shelf_entries_from_content(shelf_renderer): @@ -2424,6 +3610,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): continue yield self._extract_video(renderer) + def _rich_entries(self, rich_grid_renderer): + renderer = try_get( + rich_grid_renderer, lambda x: x['content']['videoRenderer'], dict) or {} + video_id = renderer.get('videoId') + if not video_id: + return + yield self._extract_video(renderer) + def _video_entry(self, video_renderer): video_id = video_renderer.get('videoId') if video_id: @@ -2436,12 +3630,19 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): return # video attachment video_renderer = try_get( - post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) - video_id = None - if video_renderer: - entry = self._video_entry(video_renderer) + post_renderer, lambda x: x['backstageAttachment']['videoRenderer'], dict) or {} + video_id = video_renderer.get('videoId') + if video_id: + entry = self._extract_video(video_renderer) if entry: yield entry + # playlist attachment + playlist_id = try_get( + post_renderer, lambda x: x['backstageAttachment']['playlistRenderer']['playlistId'], compat_str) + if playlist_id: + yield self.url_result( + 'https://www.youtube.com/playlist?list=%s' % playlist_id, + ie=YoutubeTabIE.ie_key(), video_id=playlist_id) # inline video links runs = try_get(post_renderer, lambda x: x['contentText']['runs'], list) or [] for run in runs: @@ -2456,7 +3657,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): ep_video_id = YoutubeIE._match_id(ep_url) if video_id == ep_video_id: continue - yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=video_id) + yield self.url_result(ep_url, ie=YoutubeIE.ie_key(), video_id=ep_video_id) def _post_thread_continuation_entries(self, post_thread_continuation): contents = post_thread_continuation.get('contents') @@ -2469,6 +3670,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): for entry in self._post_thread_entries(renderer): yield entry + r''' # unused def _rich_grid_entries(self, contents): for content in contents: video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict) @@ -2476,316 +3678,264 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): entry = self._video_entry(video_renderer) if entry: yield entry + ''' + def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data): - @staticmethod - def _build_continuation_query(continuation, ctp=None): - query = { - 'ctoken': continuation, - 'continuation': continuation, - } - if ctp: - query['itct'] = ctp - return query - - @staticmethod - def _extract_next_continuation_data(renderer): - next_continuation = try_get( - renderer, lambda x: x['continuations'][0]['nextContinuationData'], dict) - if not next_continuation: - return - continuation = next_continuation.get('continuation') - if not continuation: - return - ctp = next_continuation.get('clickTrackingParams') - return YoutubeTabIE._build_continuation_query(continuation, ctp) - - @classmethod - def _extract_continuation(cls, renderer): - next_continuation = cls._extract_next_continuation_data(renderer) - if next_continuation: - return next_continuation - contents = [] - for key in ('contents', 'items'): - contents.extend(try_get(renderer, lambda x: x[key], list) or []) - for content in contents: - if not isinstance(content, dict): - continue - continuation_ep = try_get( - content, lambda x: x['continuationItemRenderer']['continuationEndpoint'], - dict) - if not continuation_ep: - continue - continuation = try_get( - continuation_ep, lambda x: x['continuationCommand']['token'], compat_str) - if not continuation: - continue - ctp = continuation_ep.get('clickTrackingParams') - return YoutubeTabIE._build_continuation_query(continuation, ctp) - - def _entries(self, tab, item_id, webpage): - tab_content = try_get(tab, lambda x: x['content'], dict) - if not tab_content: - return - slr_renderer = try_get(tab_content, lambda x: x['sectionListRenderer'], dict) - if slr_renderer: - is_channels_tab = tab.get('title') == 'Channels' - continuation = None - slr_contents = try_get(slr_renderer, lambda x: x['contents'], list) or [] - for slr_content in slr_contents: - if not isinstance(slr_content, dict): + def extract_entries(parent_renderer): # this needs to called again for continuation to work with feeds + contents = try_get(parent_renderer, lambda x: x['contents'], list) or [] + for content in contents: + if not isinstance(content, dict): continue - is_renderer = try_get(slr_content, lambda x: x['itemSectionRenderer'], dict) + is_renderer = try_get(content, lambda x: x['itemSectionRenderer'], dict) if not is_renderer: + renderer = content.get('richItemRenderer') + if renderer: + for entry in self._rich_entries(renderer): + yield entry + continuation_list[0] = self._extract_continuation(parent_renderer) continue isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or [] for isr_content in isr_contents: if not isinstance(isr_content, dict): continue - renderer = isr_content.get('playlistVideoListRenderer') - if renderer: - for entry in self._playlist_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) - continue - renderer = isr_content.get('gridRenderer') - if renderer: - for entry in self._grid_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) - continue - renderer = isr_content.get('shelfRenderer') - if renderer: - for entry in self._shelf_entries(renderer, not is_channels_tab): - yield entry - continue - renderer = isr_content.get('backstagePostThreadRenderer') - if renderer: - for entry in self._post_thread_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) - continue - renderer = isr_content.get('videoRenderer') - if renderer: - entry = self._video_entry(renderer) - if entry: - yield entry - if not continuation: - continuation = self._extract_continuation(is_renderer) - if not continuation: - continuation = self._extract_continuation(slr_renderer) - else: - rich_grid_renderer = tab_content.get('richGridRenderer') - if not rich_grid_renderer: - return - for entry in self._rich_grid_entries(rich_grid_renderer.get('contents') or []): - yield entry - continuation = self._extract_continuation(rich_grid_renderer) + known_renderers = { + 'playlistVideoListRenderer': self._playlist_entries, + 'gridRenderer': self._grid_entries, + 'shelfRenderer': lambda x: self._shelf_entries(x, tab.get('title') != 'Channels'), + 'backstagePostThreadRenderer': self._post_thread_entries, + 'videoRenderer': lambda x: [self._video_entry(x)], + } + for key, renderer in isr_content.items(): + if key not in known_renderers: + continue + for entry in known_renderers[key](renderer): + if entry: + yield entry + continuation_list[0] = self._extract_continuation(renderer) + break - ytcfg = self._extract_ytcfg(item_id, webpage) - client_version = try_get( - ytcfg, lambda x: x['INNERTUBE_CLIENT_VERSION'], compat_str) or '2.20210407.08.00' + if not continuation_list[0]: + continuation_list[0] = self._extract_continuation(is_renderer) - headers = { - 'x-youtube-client-name': '1', - 'x-youtube-client-version': client_version, - 'content-type': 'application/json', - } + if not continuation_list[0]: + continuation_list[0] = self._extract_continuation(parent_renderer) - context = try_get(ytcfg, lambda x: x['INNERTUBE_CONTEXT'], dict) or { - 'client': { - 'clientName': 'WEB', - 'clientVersion': client_version, - } - } - visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str) - - identity_token = self._extract_identity_token(ytcfg, webpage) - if identity_token: - headers['x-youtube-identity-token'] = identity_token - - data = { - 'context': context, - } + continuation_list = [None] # Python 2 does not support nonlocal + tab_content = try_get(tab, lambda x: x['content'], dict) + if not tab_content: + return + parent_renderer = ( + try_get(tab_content, lambda x: x['sectionListRenderer'], dict) + or try_get(tab_content, lambda x: x['richGridRenderer'], dict) or {}) + for entry in extract_entries(parent_renderer): + yield entry + continuation = continuation_list[0] for page_num in itertools.count(1): if not continuation: break - if visitor_data: - headers['x-goog-visitor-id'] = visitor_data - data['continuation'] = continuation['continuation'] - data['clickTracking'] = { - 'clickTrackingParams': continuation['itct'] - } - count = 0 - retries = 3 - while count <= retries: - try: - # Downloading page may result in intermittent 5xx HTTP error - # that is usually worked around with a retry - response = self._download_json( - 'https://www.youtube.com/youtubei/v1/browse?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', - None, 'Downloading page %d%s' % (page_num, ' (retry #%d)' % count if count else ''), - headers=headers, data=json.dumps(data).encode('utf8')) - break - except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503): - count += 1 - if count <= retries: - continue - raise + headers = self.generate_api_headers( + ytcfg=ytcfg, account_syncid=account_syncid, visitor_data=visitor_data) + response = self._extract_response( + item_id='%s page %s' % (item_id, page_num), + query=continuation, headers=headers, ytcfg=ytcfg, + check_get_keys=('continuationContents', 'onResponseReceivedActions', 'onResponseReceivedEndpoints')) + if not response: break - - visitor_data = try_get( - response, lambda x: x['responseContext']['visitorData'], compat_str) or visitor_data - + # Extracting updated visitor data is required to prevent an infinite extraction loop in some cases + # See: https://github.com/ytdl-org/youtube-dl/issues/28702 + visitor_data = self._extract_visitor_data(response) or visitor_data + + known_continuation_renderers = { + 'playlistVideoListContinuation': self._playlist_entries, + 'gridContinuation': self._grid_entries, + 'itemSectionContinuation': self._post_thread_continuation_entries, + 'sectionListContinuation': extract_entries, # for feeds + } continuation_contents = try_get( - response, lambda x: x['continuationContents'], dict) - if continuation_contents: - continuation_renderer = continuation_contents.get('playlistVideoListContinuation') - if continuation_renderer: - for entry in self._playlist_entries(continuation_renderer): - yield entry - continuation = self._extract_continuation(continuation_renderer) - continue - continuation_renderer = continuation_contents.get('gridContinuation') - if continuation_renderer: - for entry in self._grid_entries(continuation_renderer): - yield entry - continuation = self._extract_continuation(continuation_renderer) - continue - continuation_renderer = continuation_contents.get('itemSectionContinuation') - if continuation_renderer: - for entry in self._post_thread_continuation_entries(continuation_renderer): - yield entry - continuation = self._extract_continuation(continuation_renderer) + response, lambda x: x['continuationContents'], dict) or {} + continuation_renderer = None + for key, value in continuation_contents.items(): + if key not in known_continuation_renderers: continue + continuation_renderer = value + continuation_list = [None] + for entry in known_continuation_renderers[key](continuation_renderer): + yield entry + continuation = continuation_list[0] or self._extract_continuation(continuation_renderer) + break + if continuation_renderer: + continue + known_renderers = { + 'gridPlaylistRenderer': (self._grid_entries, 'items'), + 'gridVideoRenderer': (self._grid_entries, 'items'), + 'gridChannelRenderer': (self._grid_entries, 'items'), + 'playlistVideoRenderer': (self._playlist_entries, 'contents'), + 'itemSectionRenderer': (extract_entries, 'contents'), # for feeds + 'richItemRenderer': (extract_entries, 'contents'), # for hashtag + 'backstagePostThreadRenderer': (self._post_thread_continuation_entries, 'contents') + } on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints')) continuation_items = try_get( on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list) - if continuation_items: - continuation_item = continuation_items[0] - if not isinstance(continuation_item, dict): - continue - renderer = self._extract_grid_item_renderer(continuation_item) - if renderer: - grid_renderer = {'items': continuation_items} - for entry in self._grid_entries(grid_renderer): - yield entry - continuation = self._extract_continuation(grid_renderer) - continue - renderer = continuation_item.get('playlistVideoRenderer') or continuation_item.get('itemSectionRenderer') - if renderer: - video_list_renderer = {'contents': continuation_items} - for entry in self._playlist_entries(video_list_renderer): - yield entry - continuation = self._extract_continuation(video_list_renderer) - continue - renderer = continuation_item.get('backstagePostThreadRenderer') - if renderer: - continuation_renderer = {'contents': continuation_items} - for entry in self._post_thread_continuation_entries(continuation_renderer): - yield entry - continuation = self._extract_continuation(continuation_renderer) - continue - renderer = continuation_item.get('richItemRenderer') - if renderer: - for entry in self._rich_grid_entries(continuation_items): - yield entry - continuation = self._extract_continuation({'contents': continuation_items}) + continuation_item = try_get(continuation_items, lambda x: x[0], dict) or {} + video_items_renderer = None + for key, value in continuation_item.items(): + if key not in known_renderers: continue - + video_items_renderer = {known_renderers[key][1]: continuation_items} + continuation_list = [None] + for entry in known_renderers[key][0](video_items_renderer): + yield entry + continuation = continuation_list[0] or self._extract_continuation(video_items_renderer) + break + if video_items_renderer: + continue break @staticmethod def _extract_selected_tab(tabs): for tab in tabs: - if try_get(tab, lambda x: x['tabRenderer']['selected'], bool): - return tab['tabRenderer'] + renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {} + if renderer.get('selected') is True: + return renderer else: raise ExtractorError('Unable to find selected tab') - @staticmethod - def _extract_uploader(data): + @classmethod + def _extract_uploader(cls, data): uploader = {} - sidebar_renderer = try_get( - data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) - if sidebar_renderer: - for item in sidebar_renderer: - if not isinstance(item, dict): - continue - renderer = item.get('playlistSidebarSecondaryInfoRenderer') - if not isinstance(renderer, dict): - continue - owner = try_get( - renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict) - if owner: - uploader['uploader'] = owner.get('text') - uploader['uploader_id'] = try_get( - owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str) - uploader['uploader_url'] = urljoin( - 'https://www.youtube.com/', - try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str)) - return uploader + renderer = cls._extract_sidebar_info_renderer(data, 'playlistSidebarSecondaryInfoRenderer') or {} + owner = try_get( + renderer, lambda x: x['videoOwner']['videoOwnerRenderer']['title']['runs'][0], dict) + if owner: + uploader['uploader'] = owner.get('text') + uploader['uploader_id'] = try_get( + owner, lambda x: x['navigationEndpoint']['browseEndpoint']['browseId'], compat_str) + uploader['uploader_url'] = urljoin( + 'https://www.youtube.com/', + try_get(owner, lambda x: x['navigationEndpoint']['browseEndpoint']['canonicalBaseUrl'], compat_str)) + return {k: v for k, v in uploader.items() if v is not None} + + def _extract_from_tabs(self, item_id, ytcfg, data, tabs): + playlist_id = title = description = channel_url = channel_name = channel_id = None + thumbnails_list = [] + tags = [] - @staticmethod - def _extract_alert(data): - alerts = [] - for alert in try_get(data, lambda x: x['alerts'], list) or []: - if not isinstance(alert, dict): - continue - alert_text = try_get( - alert, lambda x: x['alertRenderer']['text'], dict) - if not alert_text: - continue - text = try_get( - alert_text, - (lambda x: x['simpleText'], lambda x: x['runs'][0]['text']), - compat_str) - if text: - alerts.append(text) - return '\n'.join(alerts) - - def _extract_from_tabs(self, item_id, webpage, data, tabs): selected_tab = self._extract_selected_tab(tabs) renderer = try_get( data, lambda x: x['metadata']['channelMetadataRenderer'], dict) - playlist_id = item_id - title = description = None if renderer: - channel_title = renderer.get('title') or item_id - tab_title = selected_tab.get('title') - title = channel_title or item_id - if tab_title: - title += ' - %s' % tab_title - description = renderer.get('description') - playlist_id = renderer.get('externalId') + channel_name = renderer.get('title') + channel_url = renderer.get('channelUrl') + channel_id = renderer.get('externalId') else: renderer = try_get( data, lambda x: x['metadata']['playlistMetadataRenderer'], dict) - if renderer: - title = renderer.get('title') - else: - renderer = try_get( - data, lambda x: x['header']['hashtagHeaderRenderer'], dict) - if renderer: - title = try_get(renderer, lambda x: x['hashtag']['simpleText']) - playlist = self.playlist_result( - self._entries(selected_tab, item_id, webpage), - playlist_id=playlist_id, playlist_title=title, - playlist_description=description) - playlist.update(self._extract_uploader(data)) - return playlist - - def _extract_from_playlist(self, item_id, url, data, playlist): + + if renderer: + title = renderer.get('title') + description = renderer.get('description', '') + playlist_id = channel_id + tags = renderer.get('keywords', '').split() + thumbnails_list = ( + try_get(renderer, lambda x: x['avatar']['thumbnails'], list) + or try_get( + self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer'), + lambda x: x['thumbnailRenderer']['playlistVideoThumbnailRenderer']['thumbnail']['thumbnails'], + list) + or []) + + thumbnails = [] + for t in thumbnails_list: + if not isinstance(t, dict): + continue + thumbnail_url = url_or_none(t.get('url')) + if not thumbnail_url: + continue + thumbnails.append({ + 'url': thumbnail_url, + 'width': int_or_none(t.get('width')), + 'height': int_or_none(t.get('height')), + }) + if playlist_id is None: + playlist_id = item_id + if title is None: + title = ( + try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText']) + or playlist_id) + title += format_field(selected_tab, 'title', ' - %s') + title += format_field(selected_tab, 'expandedText', ' - %s') + metadata = { + 'playlist_id': playlist_id, + 'playlist_title': title, + 'playlist_description': description, + 'uploader': channel_name, + 'uploader_id': channel_id, + 'uploader_url': channel_url, + 'thumbnails': thumbnails, + 'tags': tags, + } + availability = self._extract_availability(data) + if availability: + metadata['availability'] = availability + if not channel_id: + metadata.update(self._extract_uploader(data)) + metadata.update({ + 'channel': metadata['uploader'], + 'channel_id': metadata['uploader_id'], + 'channel_url': metadata['uploader_url']}) + return self.playlist_result( + self._entries( + selected_tab, playlist_id, ytcfg, + self._extract_account_syncid(ytcfg, data), + self._extract_visitor_data(data, ytcfg)), + **metadata) + + def _extract_mix_playlist(self, playlist, playlist_id, data, ytcfg): + first_id = last_id = response = None + for page_num in itertools.count(1): + videos = list(self._playlist_entries(playlist)) + if not videos: + return + start = next((i for i, v in enumerate(videos) if v['id'] == last_id), -1) + 1 + if start >= len(videos): + return + for video in videos[start:]: + if video['id'] == first_id: + self.to_screen('First video %s found again; Assuming end of Mix' % first_id) + return + yield video + first_id = first_id or videos[0]['id'] + last_id = videos[-1]['id'] + watch_endpoint = try_get( + playlist, lambda x: x['contents'][-1]['playlistPanelVideoRenderer']['navigationEndpoint']['watchEndpoint']) + headers = self.generate_api_headers( + ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data), + visitor_data=self._extract_visitor_data(response, data, ytcfg)) + query = { + 'playlistId': playlist_id, + 'videoId': watch_endpoint.get('videoId') or last_id, + 'index': watch_endpoint.get('index') or len(videos), + 'params': watch_endpoint.get('params') or 'OAE%3D' + } + response = self._extract_response( + item_id='%s page %d' % (playlist_id, page_num), + query=query, ep='next', headers=headers, ytcfg=ytcfg, + check_get_keys='contents' + ) + playlist = try_get( + response, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict) + + def _extract_from_playlist(self, item_id, url, data, playlist, ytcfg): title = playlist.get('title') or try_get( data, lambda x: x['titleText']['simpleText'], compat_str) playlist_id = playlist.get('playlistId') or item_id - # Inline playlist rendition continuation does not always work - # at Youtube side, so delegating regular tab-based playlist URL - # processing whenever possible. + + # Delegating everything except mix playlists to regular tab-based playlist URL playlist_url = urljoin(url, try_get( playlist, lambda x: x['endpoint']['commandMetadata']['webCommandMetadata']['url'], compat_str)) @@ -2793,54 +3943,297 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): return self.url_result( playlist_url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id, video_title=title) + return self.playlist_result( - self._playlist_entries(playlist), playlist_id=playlist_id, - playlist_title=title) + self._extract_mix_playlist(playlist, playlist_id, data, ytcfg), + playlist_id=playlist_id, playlist_title=title) - def _extract_identity_token(self, ytcfg, webpage): - if ytcfg: - token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str) - if token: - return token - return self._search_regex( - r'\bID_TOKEN["\']\s*:\s*["\'](.+?)["\']', webpage, - 'identity token', default=None) + def _extract_availability(self, data): + """ + Gets the availability of a given playlist/tab. + Note: Unless YouTube tells us explicitly, we do not assume it is public + @param data: response + """ + is_private = is_unlisted = None + renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') or {} + badge_labels = self._extract_badges(renderer) + + # Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge + privacy_dropdown_entries = try_get( + renderer, lambda x: x['privacyForm']['dropdownFormFieldRenderer']['dropdown']['dropdownRenderer']['entries'], list) or [] + for renderer_dict in privacy_dropdown_entries: + is_selected = try_get( + renderer_dict, lambda x: x['privacyDropdownItemRenderer']['isSelected'], bool) or False + if not is_selected: + continue + label = self._get_text(renderer_dict, ('privacyDropdownItemRenderer', 'label')) + if label: + badge_labels.add(label.lower()) + break + + for badge_label in badge_labels: + if badge_label == 'unlisted': + is_unlisted = True + elif badge_label == 'private': + is_private = True + elif badge_label == 'public': + is_unlisted = is_private = False + return self._availability(is_private, False, False, False, is_unlisted) + + @staticmethod + def _extract_sidebar_info_renderer(data, info_renderer, expected_type=dict): + sidebar_renderer = try_get( + data, lambda x: x['sidebar']['playlistSidebarRenderer']['items'], list) or [] + for item in sidebar_renderer: + renderer = try_get(item, lambda x: x[info_renderer], expected_type) + if renderer: + return renderer + + def _reload_with_unavailable_videos(self, item_id, data, ytcfg): + """ + Get playlist with unavailable videos if the 'show unavailable videos' button exists. + """ + browse_id = params = None + renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') + if not renderer: + return + menu_renderer = try_get( + renderer, lambda x: x['menu']['menuRenderer']['items'], list) or [] + for menu_item in menu_renderer: + if not isinstance(menu_item, dict): + continue + nav_item_renderer = menu_item.get('menuNavigationItemRenderer') + text = try_get( + nav_item_renderer, lambda x: x['text']['simpleText'], compat_str) + if not text or text.lower() != 'show unavailable videos': + continue + browse_endpoint = try_get( + nav_item_renderer, lambda x: x['navigationEndpoint']['browseEndpoint'], dict) or {} + browse_id = browse_endpoint.get('browseId') + params = browse_endpoint.get('params') + break + + headers = self.generate_api_headers( + ytcfg=ytcfg, account_syncid=self._extract_account_syncid(ytcfg, data), + visitor_data=self._extract_visitor_data(data, ytcfg)) + query = { + 'params': params or 'wgYCCAA=', + 'browseId': browse_id or 'VL%s' % item_id + } + return self._extract_response( + item_id=item_id, headers=headers, query=query, + check_get_keys='contents', fatal=False, ytcfg=ytcfg, + note='Downloading API JSON with unavailable videos') + + def _extract_webpage(self, url, item_id, fatal=True): + retries = self.get_param('extractor_retries', 3) + count = -1 + webpage = data = last_error = None + while count < retries: + count += 1 + # Sometimes youtube returns a webpage with incomplete ytInitialData + # See: https://github.com/hypervideo/hypervideo/issues/116 + if last_error: + self.report_warning('%s. Retrying ...' % last_error) + try: + webpage = self._download_webpage( + url, item_id, + note='Downloading webpage%s' % (' (retry #%d)' % count if count else '',)) + data = self.extract_yt_initial_data(item_id, webpage or '', fatal=fatal) or {} + except ExtractorError as e: + if isinstance(e.cause, network_exceptions): + if not isinstance(e.cause, compat_HTTPError) or e.cause.code not in (403, 429): + last_error = error_to_compat_str(e.cause or e.msg) + if count < retries: + continue + if fatal: + raise + self.report_warning(error_to_compat_str(e)) + break + else: + try: + self._extract_and_report_alerts(data) + except ExtractorError as e: + if fatal: + raise + self.report_warning(error_to_compat_str(e)) + break + + if dict_get(data, ('contents', 'currentVideoEndpoint')): + break + + last_error = 'Incomplete yt initial data received' + if count >= retries: + if fatal: + raise ExtractorError(last_error) + self.report_warning(last_error) + break + + return webpage, data + + def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'): + data = None + if 'webpage' not in self._configuration_arg('skip'): + webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal) + ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage) + if not data: + if not ytcfg and self.is_authenticated: + msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.' + if 'authcheck' not in self._configuration_arg('skip') and fatal: + raise ExtractorError( + msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,' + ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check', + expected=True) + self.report_warning(msg, only_once=True) + data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client) + return data, ytcfg + + def _extract_tab_endpoint(self, url, item_id, ytcfg=None, fatal=True, default_client='web'): + headers = self.generate_api_headers(ytcfg=ytcfg, default_client=default_client) + resolve_response = self._extract_response( + item_id=item_id, query={'url': url}, check_get_keys='endpoint', headers=headers, ytcfg=ytcfg, fatal=fatal, + ep='navigation/resolve_url', note='Downloading API parameters API JSON', default_client=default_client) + endpoints = {'browseEndpoint': 'browse', 'watchEndpoint': 'next'} + for ep_key, ep in endpoints.items(): + params = try_get(resolve_response, lambda x: x['endpoint'][ep_key], dict) + if params: + return self._extract_response( + item_id=item_id, query=params, ep=ep, headers=headers, + ytcfg=ytcfg, fatal=fatal, default_client=default_client, + check_get_keys=('contents', 'currentVideoEndpoint')) + err_note = 'Failed to resolve url (does the playlist exist?)' + if fatal: + raise ExtractorError(err_note, expected=True) + self.report_warning(err_note, item_id) + + @staticmethod + def _smuggle_data(entries, data): + for entry in entries: + if data: + entry['url'] = smuggle_url(entry['url'], data) + yield entry def _real_extract(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + if self.is_music_url(url): + smuggled_data['is_music_url'] = True + info_dict = self.__real_extract(url, smuggled_data) + if info_dict.get('entries'): + info_dict['entries'] = self._smuggle_data(info_dict['entries'], smuggled_data) + return info_dict + + _url_re = re.compile(r'(?P<pre>%s)(?(channel_type)(?P<tab>/\w+))?(?P<post>.*)$' % _VALID_URL) + + def __real_extract(self, url, smuggled_data): item_id = self._match_id(url) url = compat_urlparse.urlunparse( compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com')) + compat_opts = self.get_param('compat_opts', []) + + def get_mobj(url): + mobj = self._url_re.match(url).groupdict() + mobj.update((k, '') for k, v in mobj.items() if v is None) + return mobj + + mobj = get_mobj(url) + # Youtube returns incomplete data if tabname is not lower case + pre, tab, post, is_channel = mobj['pre'], mobj['tab'].lower(), mobj['post'], not mobj['not_channel'] + if is_channel: + if smuggled_data.get('is_music_url'): + if item_id[:2] == 'VL': + # Youtube music VL channels have an equivalent playlist + item_id = item_id[2:] + pre, tab, post, is_channel = 'https://www.youtube.com/playlist?list=%s' % item_id, '', '', False + elif item_id[:2] == 'MP': + # Resolve albums (/[channel/browse]/MP...) to their equivalent playlist + mdata = self._extract_tab_endpoint( + 'https://music.youtube.com/channel/%s' % item_id, item_id, default_client='web_music') + murl = traverse_obj( + mdata, ('microformat', 'microformatDataRenderer', 'urlCanonical'), get_all=False, expected_type=compat_str) + if not murl: + raise ExtractorError('Failed to resolve album to playlist.') + return self.url_result(murl, ie=YoutubeTabIE.ie_key()) + elif mobj['channel_type'] == 'browse': + # Youtube music /browse/ should be changed to /channel/ + pre = 'https://www.youtube.com/channel/%s' % item_id + if is_channel and not tab and 'no-youtube-channel-redirect' not in compat_opts: + # Home URLs should redirect to /videos/ + self.report_warning( + 'A channel/user page was given. All the channel\'s videos will be downloaded. ' + 'To download only the videos in the home page, add a "/featured" to the URL') + tab = '/videos' + + url = ''.join((pre, tab, post)) + mobj = get_mobj(url) + # Handle both video/playlist URLs qs = parse_qs(url) video_id = qs.get('v', [None])[0] playlist_id = qs.get('list', [None])[0] + + if not video_id and mobj['not_channel'].startswith('watch'): + if not playlist_id: + # If there is neither video or playlist ids, youtube redirects to home page, which is undesirable + raise ExtractorError('Unable to recognize tab page') + # Common mistake: https://www.youtube.com/watch?list=playlist_id + self.report_warning('A video URL was given without video ID. Trying to download playlist %s' % playlist_id) + url = 'https://www.youtube.com/playlist?list=%s' % playlist_id + mobj = get_mobj(url) + if video_id and playlist_id: - if self._downloader.params.get('noplaylist'): + if self.get_param('noplaylist'): self.to_screen('Downloading just video %s because of --no-playlist' % video_id) - return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id) - self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) - webpage = self._download_webpage(url, item_id) - data = self._extract_yt_initial_data(item_id, webpage) + return self.url_result(f'https://www.youtube.com/watch?v={video_id}', ie=YoutubeIE.ie_key(), video_id=video_id) + self.to_screen('Downloading playlist %s; add --no-playlist to just download video %s' % (playlist_id, video_id)) + + data, ytcfg = self._extract_data(url, item_id) + + tabs = try_get( + data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list) + if tabs: + selected_tab = self._extract_selected_tab(tabs) + tab_name = selected_tab.get('title', '') + if 'no-youtube-channel-redirect' not in compat_opts: + if mobj['tab'] == '/live': + # Live tab should have redirected to the video + raise ExtractorError('The channel is not currently live', expected=True) + if mobj['tab'] == '/videos' and tab_name.lower() != mobj['tab'][1:]: + if not mobj['not_channel'] and item_id[:2] == 'UC': + # Topic channels don't have /videos. Use the equivalent playlist instead + self.report_warning('The URL does not have a %s tab. Trying to redirect to playlist UU%s instead' % (mobj['tab'][1:], item_id[2:])) + pl_id = 'UU%s' % item_id[2:] + pl_url = 'https://www.youtube.com/playlist?list=%s%s' % (pl_id, mobj['post']) + try: + data, ytcfg, item_id, url = *self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True), pl_id, pl_url + except ExtractorError: + self.report_warning('The playlist gave error. Falling back to channel URL') + else: + self.report_warning('The URL does not have a %s tab. %s is being downloaded instead' % (mobj['tab'][1:], tab_name)) + + self.write_debug('Final URL: %s' % url) + + # YouTube sometimes provides a button to reload playlist with unavailable videos. + if 'no-youtube-unavailable-videos' not in compat_opts: + data = self._reload_with_unavailable_videos(item_id, data, ytcfg) or data + self._extract_and_report_alerts(data, only_once=True) tabs = try_get( data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list) if tabs: - return self._extract_from_tabs(item_id, webpage, data, tabs) + return self._extract_from_tabs(item_id, ytcfg, data, tabs) + playlist = try_get( data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict) if playlist: - return self._extract_from_playlist(item_id, url, data, playlist) - # Fallback to video extraction if no playlist alike page is recognized. - # First check for the current video then try the v attribute of URL query. + return self._extract_from_playlist(item_id, url, data, playlist, ytcfg) + video_id = try_get( data, lambda x: x['currentVideoEndpoint']['watchEndpoint']['videoId'], compat_str) or video_id if video_id: - return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id) - # Capture and output alerts - alert = self._extract_alert(data) - if alert: - raise ExtractorError(alert, expected=True) - # Failed to recognize + if mobj['tab'] != '/live': # live tab is expected to redirect to video + self.report_warning('Unable to recognize playlist. Downloading just video %s' % video_id) + return self.url_result(f'https://www.youtube.com/watch?v={video_id}', ie=YoutubeIE.ie_key(), video_id=video_id) + raise ExtractorError('Unable to recognize tab page') @@ -2867,6 +4260,7 @@ class YoutubePlaylistIE(InfoExtractor): 'id': 'PLBB231211A4F62143', 'uploader': 'Wickydoo', 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q', + 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2', }, 'playlist_mincount': 29, }, { @@ -2889,12 +4283,13 @@ class YoutubePlaylistIE(InfoExtractor): } }, { 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', - 'playlist_mincount': 982, + 'playlist_mincount': 654, 'info_dict': { 'title': '2018 Chinese New Singles (11/6 updated)', 'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', 'uploader': 'LBK', 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA', + 'description': 'md5:da521864744d60a198e3a88af4db0d9d', } }, { 'url': 'TLGGrESM50VT6acwMjAyMjAxNw', @@ -2919,15 +4314,17 @@ class YoutubePlaylistIE(InfoExtractor): def _real_extract(self, url): playlist_id = self._match_id(url) - qs = parse_qs(url) - if not qs: - qs = {'list': playlist_id} - return self.url_result( - update_url_query('https://www.youtube.com/playlist', qs), - ie=YoutubeTabIE.ie_key(), video_id=playlist_id) + is_music_url = YoutubeBaseInfoExtractor.is_music_url(url) + url = update_url_query( + 'https://www.youtube.com/playlist', + parse_qs(url) or {'list': playlist_id}) + if is_music_url: + url = smuggle_url(url, {'is_music_url': True}) + return self.url_result(url, ie=YoutubeTabIE.ie_key(), video_id=playlist_id) class YoutubeYtBeIE(InfoExtractor): + IE_DESC = 'youtu.be' _VALID_URL = r'https?://youtu\.be/(?P<id>[0-9A-Za-z_-]{11})/*?.*?\blist=(?P<playlist_id>%(playlist_id)s)' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE} _TESTS = [{ 'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5', @@ -2955,7 +4352,7 @@ class YoutubeYtBeIE(InfoExtractor): }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) + mobj = self._match_valid_url(url) video_id = mobj.group('id') playlist_id = mobj.group('playlist_id') return self.url_result( @@ -2967,6 +4364,7 @@ class YoutubeYtBeIE(InfoExtractor): class YoutubeYtUserIE(InfoExtractor): + IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword' _VALID_URL = r'ytuser:(?P<id>.+)' _TESTS = [{ 'url': 'ytuser:phihag', @@ -2982,8 +4380,8 @@ class YoutubeYtUserIE(InfoExtractor): class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): IE_NAME = 'youtube:favorites' - IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)' - _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?' + IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)' + _VALID_URL = r':ytfav(?:ou?rite)?s?' _LOGIN_REQUIRED = True _TESTS = [{ 'url': ':ytfav', @@ -2999,8 +4397,8 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): ie=YoutubeTabIE.ie_key()) -class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor): - IE_DESC = 'YouTube.com searches' +class YoutubeSearchIE(SearchInfoExtractor, YoutubeTabIE): + IE_DESC = 'YouTube.com searches, "ytsearch" keyword' # there doesn't appear to be a real limit, for example if you search for # 'python' you get more than 8.000.000 results _MAX_RESULTS = float('inf') @@ -3009,27 +4407,17 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor): _SEARCH_PARAMS = None _TESTS = [] - def _entries(self, query, n): - data = { - 'context': { - 'client': { - 'clientName': 'WEB', - 'clientVersion': '2.20201021.03.00', - } - }, - 'query': query, - } + def _search_results(self, query): + data = {'query': query} if self._SEARCH_PARAMS: data['params'] = self._SEARCH_PARAMS - total = 0 + continuation = {} for page_num in itertools.count(1): - search = self._download_json( - 'https://www.youtube.com/youtubei/v1/search?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', - video_id='query "%s"' % query, - note='Downloading page %s' % page_num, - errnote='Unable to download API page', fatal=False, - data=json.dumps(data).encode('utf8'), - headers={'content-type': 'application/json'}) + data.update(continuation) + search = self._extract_response( + item_id='query "%s" page %s' % (query, page_num), ep='search', query=data, + check_get_keys=('contents', 'onResponseReceivedCommands') + ) if not search: break slr_contents = try_get( @@ -3039,7 +4427,15 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor): list) if not slr_contents: break + + # Youtube sometimes adds promoted content to searches, + # changing the index location of videos and token. + # So we search through all entries till we find them. + continuation = None for slr_content in slr_contents: + if not continuation: + continuation = self._extract_continuation({'contents': [slr_content]}) + isr_contents = try_get( slr_content, lambda x: x['itemSectionRenderer']['contents'], @@ -3055,52 +4451,46 @@ class YoutubeSearchIE(SearchInfoExtractor, YoutubeBaseInfoExtractor): video_id = video.get('videoId') if not video_id: continue + yield self._extract_video(video) - total += 1 - if total == n: - return - token = try_get( - slr_contents, - lambda x: x[-1]['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'], - compat_str) - if not token: - break - data['continuation'] = token - def _get_n_results(self, query, n): - """Get a specified number of results for a query""" - return self.playlist_result(self._entries(query, n), query) + if not continuation: + break class YoutubeSearchDateIE(YoutubeSearchIE): IE_NAME = YoutubeSearchIE.IE_NAME + ':date' _SEARCH_KEY = 'ytsearchdate' - IE_DESC = 'YouTube.com searches, newest videos first' + IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword' _SEARCH_PARAMS = 'CAI%3D' -r""" class YoutubeSearchURLIE(YoutubeSearchIE): IE_DESC = 'YouTube.com search URLs' - IE_NAME = 'youtube:search_url' - _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)' + IE_NAME = YoutubeSearchIE.IE_NAME + '_url' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)' + # _MAX_RESULTS = 100 _TESTS = [{ 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', 'playlist_mincount': 5, 'info_dict': { - 'title': 'hypervideo test video', + 'id': 'youtube-dl test video', + 'title': 'youtube-dl test video', } }, { 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB', 'only_matching': True, }] + @classmethod + def _make_valid_url(cls): + return cls._VALID_URL + def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - query = compat_urllib_parse_unquote_plus(mobj.group('query')) - webpage = self._download_webpage(url, query) - return self.playlist_result(self._process_page(webpage), playlist_title=query) -""" + qs = parse_qs(url) + query = (qs.get('search_query') or qs.get('q'))[0] + self._SEARCH_PARAMS = qs.get('sp', ('',))[0] + return self._get_n_results(query, self._MAX_RESULTS) class YoutubeFeedsInfoExtractor(YoutubeTabIE): @@ -3109,14 +4499,12 @@ class YoutubeFeedsInfoExtractor(YoutubeTabIE): Subclasses must define the _FEED_NAME property. """ _LOGIN_REQUIRED = True + _TESTS = [] @property def IE_NAME(self): return 'youtube:%s' % self._FEED_NAME - def _real_initialize(self): - self._login() - def _real_extract(self, url): return self.url_result( 'https://www.youtube.com/feed/%s' % self._FEED_NAME, @@ -3139,20 +4527,24 @@ class YoutubeWatchLaterIE(InfoExtractor): class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)' - _VALID_URL = r':ytrec(?:ommended)?' + _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?' _FEED_NAME = 'recommended' + _LOGIN_REQUIRED = False _TESTS = [{ 'url': ':ytrec', 'only_matching': True, }, { 'url': ':ytrecommended', 'only_matching': True, + }, { + 'url': 'https://youtube.com', + 'only_matching': True, }] class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): - IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)' - _VALID_URL = r':ytsubs(?:criptions)?' + IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)' + _VALID_URL = r':ytsub(?:scription)?s?' _FEED_NAME = 'subscriptions' _TESTS = [{ 'url': ':ytsubs', @@ -3164,8 +4556,8 @@ class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): - IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)' - _VALID_URL = r':ythistory' + IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)' + _VALID_URL = r':ythis(?:tory)?' _FEED_NAME = 'history' _TESTS = [{ 'url': ':ythistory', @@ -3216,12 +4608,22 @@ class YoutubeTruncatedURLIE(InfoExtractor): raise ExtractorError( 'Did you forget to quote the URL? Remember that & is a meta ' 'character in most shells, so you want to put the URL in quotes, ' - 'like hypervideo ' + 'like youtube-dl ' '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" ' - ' or simply hypervideo BaW_jenozKc .', + ' or simply youtube-dl BaW_jenozKc .', expected=True) +class YoutubeClipIE(InfoExtractor): + IE_NAME = 'youtube:clip' + IE_DESC = False # Do not list + _VALID_URL = r'https?://(?:www\.)?youtube\.com/clip/' + + def _real_extract(self, url): + self.report_warning('YouTube clips are not currently supported. The entire video will be downloaded instead') + return self.url_result(url, 'Generic') + + class YoutubeTruncatedIDIE(InfoExtractor): IE_NAME = 'youtube:truncated_id' IE_DESC = False # Do not list |