diff options
Diffstat (limited to 'hypervideo_dl/extractor/youtube.py')
| -rw-r--r-- | hypervideo_dl/extractor/youtube.py | 3631 |
1 files changed, 2373 insertions, 1258 deletions
diff --git a/hypervideo_dl/extractor/youtube.py b/hypervideo_dl/extractor/youtube.py index dc5ee63..dec3b14 100644 --- a/hypervideo_dl/extractor/youtube.py +++ b/hypervideo_dl/extractor/youtube.py @@ -2,18 +2,21 @@ from __future__ import unicode_literals -import base64 import calendar import copy import datetime +import functools import hashlib import itertools import json +import math import os.path import random import re +import sys import time import traceback +import threading from .common import InfoExtractor, SearchInfoExtractor from ..compat import ( @@ -28,7 +31,7 @@ from ..compat import ( ) from ..jsinterp import JSInterpreter from ..utils import ( - bytes_to_intlist, + bug_reports_message, clean_html, datetime_from_str, dict_get, @@ -36,11 +39,14 @@ from ..utils import ( ExtractorError, float_or_none, format_field, + get_first, int_or_none, - intlist_to_bytes, is_html, + join_nonempty, + js_to_json, mimetype2ext, network_exceptions, + NO_DEFAULT, orderedSet, parse_codecs, parse_count, @@ -53,10 +59,12 @@ from ..utils import ( smuggle_url, str_or_none, str_to_int, + strftime_or_none, traverse_obj, try_get, unescapeHTML, unified_strdate, + unified_timestamp, unsmuggle_url, update_url_query, url_or_none, @@ -72,7 +80,7 @@ INNERTUBE_CLIENTS = { 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'WEB', - 'clientVersion': '2.20210622.10.00', + 'clientVersion': '2.20211221.00.00', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 1 @@ -82,7 +90,7 @@ INNERTUBE_CLIENTS = { 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'WEB_EMBEDDED_PLAYER', - 'clientVersion': '1.20210620.0.1', + 'clientVersion': '1.20211215.00.01', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 56 @@ -93,96 +101,96 @@ INNERTUBE_CLIENTS = { 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'WEB_REMIX', - 'clientVersion': '1.20210621.00.00', + 'clientVersion': '1.20211213.00.00', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 67, }, 'web_creator': { - 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_API_KEY': 'AIzaSyBUPetSUmoZL-OhlxA7wSac5XinrygCqMo', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'WEB_CREATOR', - 'clientVersion': '1.20210621.00.00', + 'clientVersion': '1.20211220.02.00', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 62, }, 'android': { - 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_API_KEY': 'AIzaSyA8eiZmM1FaDVjRy-df2KTyQ_vz_yYM39w', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'ANDROID', - 'clientVersion': '16.20', + 'clientVersion': '16.49', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 3, 'REQUIRE_JS_PLAYER': False }, 'android_embedded': { - 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_API_KEY': 'AIzaSyCjc_pVEDi4qsv5MtC2dMXzpIaDoRFLsxw', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'ANDROID_EMBEDDED_PLAYER', - 'clientVersion': '16.20', + 'clientVersion': '16.49', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 55, 'REQUIRE_JS_PLAYER': False }, 'android_music': { - 'INNERTUBE_API_KEY': 'AIzaSyC9XL3ZjWddXya6X74dJoCTL-WEYFDNX30', - 'INNERTUBE_HOST': 'music.youtube.com', + 'INNERTUBE_API_KEY': 'AIzaSyAOghZGza2MQSZkY_zfZ370N-PUdXEo8AI', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'ANDROID_MUSIC', - 'clientVersion': '4.32', + 'clientVersion': '4.57', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 21, 'REQUIRE_JS_PLAYER': False }, 'android_creator': { + 'INNERTUBE_API_KEY': 'AIzaSyD_qjV8zaaUMehtLkrKFgVeSX_Iqbtyws8', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'ANDROID_CREATOR', - 'clientVersion': '21.24.100', + 'clientVersion': '21.47', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 14, 'REQUIRE_JS_PLAYER': False }, - # ios has HLS live streams - # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680 + # iOS clients have HLS live streams. Setting device model to get 60fps formats. + # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558 'ios': { - 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_API_KEY': 'AIzaSyB-63vPrdThhKuerbB2N_l7Kwwcxj6yUAc', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'IOS', - 'clientVersion': '16.20', + 'clientVersion': '16.46', + 'deviceModel': 'iPhone14,3', } }, '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', + 'clientVersion': '16.46', + 'deviceModel': 'iPhone14,3', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 66, 'REQUIRE_JS_PLAYER': False }, 'ios_music': { - 'INNERTUBE_API_KEY': 'AIzaSyDK3iBpDP9nHVTk2qL73FLJICfOC3c51Og', - 'INNERTUBE_HOST': 'music.youtube.com', + 'INNERTUBE_API_KEY': 'AIzaSyBAETezhkwP0ZWA02RsqT1zu78Fpt0bC_s', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'IOS_MUSIC', - 'clientVersion': '4.32', + 'clientVersion': '4.57', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 26, @@ -192,7 +200,7 @@ INNERTUBE_CLIENTS = { 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'IOS_CREATOR', - 'clientVersion': '21.24.100', + 'clientVersion': '21.47', }, }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 15, @@ -201,39 +209,61 @@ INNERTUBE_CLIENTS = { # mweb has 'ultralow' formats # See: https://github.com/hypervideo/hypervideo/pull/557 'mweb': { - 'INNERTUBE_API_KEY': 'AIzaSyDCU8hByM-4DrUqRUYnGn-3llEO78bcxq8', + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', 'INNERTUBE_CONTEXT': { 'client': { 'clientName': 'MWEB', - 'clientVersion': '2.20210721.07.00', + 'clientVersion': '2.20211221.01.00', } }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 2 }, + # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option) + # See: https://github.com/zerodytrash/YouTube-Internal-Clients + 'tv_embedded': { + 'INNERTUBE_API_KEY': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'TVHTML5_SIMPLY_EMBEDDED_PLAYER', + 'clientVersion': '2.0', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 85 + }, } +def _split_innertube_client(client_name): + variant, *base = client_name.rsplit('.', 1) + if base: + return variant, base[0], variant + base, *variant = client_name.split('_', 1) + return client_name, base, variant[0] if variant else None + + def build_innertube_clients(): - third_party = { - 'embedUrl': 'https://google.com', # Can be any valid URL + THIRD_PARTY = { + 'embedUrl': 'https://www.youtube.com/', # Can be any valid URL } - base_clients = ('android', 'web', 'ios', 'mweb') - priority = qualities(base_clients[::-1]) + BASE_CLIENTS = ('android', 'web', 'tv', '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 + + _, base_client, variant = _split_innertube_client(client) + ytcfg['priority'] = 10 * priority(base_client) + + if not variant: + INNERTUBE_CLIENTS[f'{client}_embedscreen'] = embedscreen = copy.deepcopy(ytcfg) + embedscreen['INNERTUBE_CONTEXT']['client']['clientScreen'] = 'EMBED' + embedscreen['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY + embedscreen['priority'] -= 3 + elif variant == 'embedded': + ytcfg['INNERTUBE_CONTEXT']['thirdParty'] = THIRD_PARTY ytcfg['priority'] -= 2 else: ytcfg['priority'] -= 3 @@ -247,31 +277,82 @@ class YoutubeBaseInfoExtractor(InfoExtractor): _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'shorts|movies|results|search|shared|hashtag|trending|explore|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' + # _NETRC_MACHINE = 'youtube' # If True it will raise an error if no login info is provided _LOGIN_REQUIRED = False - def _login(self): - """ - Attempt to log in to YouTube. - If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised. - """ - - 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"]}') + _INVIDIOUS_SITES = ( + # invidious-redirect websites + 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', + r'(?:www\.)?osbivz6guyeahrwp2lnwyjk2xos342h4ocsxyqrlaopqjuhwn2djiiyd\.onion', + r'(?:www\.)?u2cvlit75owumwpy4dj2hsmvkq7nvrclkpht7xgyye2pyoxhpmclkrad\.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\.mastodon\.host', + r'(?:www\.)?invidious\.zapashcanon\.fr', + r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks', + r'(?:www\.)?invidious\.tinfoil-hat\.net', + r'(?:www\.)?invidious\.himiko\.cloud', + r'(?:www\.)?invidious\.reallyancient\.tech', + r'(?:www\.)?invidious\.tube', + r'(?:www\.)?invidiou\.site', + r'(?:www\.)?invidious\.site', + r'(?:www\.)?invidious\.xyz', + r'(?:www\.)?invidious\.nixnet\.xyz', + r'(?:www\.)?invidious\.048596\.xyz', + r'(?:www\.)?invidious\.drycat\.fr', + r'(?:www\.)?inv\.skyn3t\.in', + r'(?:www\.)?tube\.poal\.co', + r'(?:www\.)?tube\.connect\.cafe', + r'(?:www\.)?vid\.wxzm\.sx', + r'(?:www\.)?vid\.mint\.lgbt', + r'(?:www\.)?vid\.puffyan\.us', + r'(?:www\.)?yewtu\.be', + r'(?:www\.)?yt\.elukerio\.org', + r'(?:www\.)?yt\.lelux\.fi', + r'(?:www\.)?invidious\.ggc-project\.de', + r'(?:www\.)?yt\.maisputain\.ovh', + r'(?:www\.)?ytprivate\.com', + r'(?:www\.)?invidious\.13ad\.de', + 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', + r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion', + r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion', + 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', + ) def _initialize_consent(self): cookies = self._get_cookies('https://www.youtube.com/') @@ -288,9 +369,25 @@ class YoutubeBaseInfoExtractor(InfoExtractor): consent_id = random.randint(100, 999) self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id) + def _initialize_pref(self): + cookies = self._get_cookies('https://www.youtube.com/') + pref_cookie = cookies.get('PREF') + pref = {} + if pref_cookie: + try: + pref = dict(compat_urlparse.parse_qsl(pref_cookie.value)) + except ValueError: + self.report_warning('Failed to parse user PREF cookie' + bug_reports_message()) + pref.update({'hl': 'en', 'tz': 'UTC'}) + self._set_cookie('.youtube.com', name='PREF', value=compat_urllib_parse_urlencode(pref)) + def _real_initialize(self): + self._initialize_pref() self._initialize_consent() - self._login() + 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') _YT_INITIAL_DATA_RE = r'(?:window\s*\[\s*["\']ytInitialData["\']\s*\]|ytInitialData)\s*=\s*({.+?})\s*;' _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;' @@ -321,23 +418,11 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 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 + context = get_first( + (ytcfg, self._get_default_ytcfg(default_client)), 'INNERTUBE_CONTEXT', expected_type=dict) + # Enforce language and tz for extraction + client_context = traverse_obj(context, 'client', expected_type=dict, default={}) + client_context.update({'hl': 'en', 'timeZone': 'UTC', 'utcOffsetMinutes': 0}) return context _SAPISID = None @@ -381,7 +466,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor): '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()}) + query={'key': api_key or self._extract_api_key(), 'prettyPrint': 'false'}) def extract_yt_initial_data(self, item_id, webpage, fatal=True): data = self._search_regex( @@ -437,9 +522,9 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 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) + return get_first( + args, [('VISITOR_DATA', ('INNERTUBE_CONTEXT', 'client', 'visitorData'), ('responseContext', 'visitorData'))], + expected_type=str) @property def is_authenticated(self): @@ -594,6 +679,72 @@ class YoutubeBaseInfoExtractor(InfoExtractor): if text: return text + def _get_count(self, data, *path_list): + count_text = self._get_text(data, *path_list) or '' + count = parse_count(count_text) + if count is None: + count = str_to_int( + self._search_regex(r'^([\d,]+)', re.sub(r'\s', '', count_text), 'count', default=None)) + return count + + @staticmethod + def _extract_thumbnails(data, *path_list): + """ + Extract thumbnails from thumbnails dict + @param path_list: path list to level that contains 'thumbnails' key + """ + thumbnails = [] + for path in path_list or [()]: + for thumbnail in traverse_obj(data, (*variadic(path), 'thumbnails', ...), default=[]): + thumbnail_url = url_or_none(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')), + }) + return thumbnails + + @staticmethod + def extract_relative_time(relative_time_text): + """ + Extracts a relative time from string and converts to dt object + e.g. 'streamed 6 days ago', '5 seconds ago (edited)', 'updated today' + """ + mobj = re.search(r'(?P<start>today|yesterday|now)|(?P<time>\d+)\s*(?P<unit>microsecond|second|minute|hour|day|week|month|year)s?\s*ago', relative_time_text) + if mobj: + start = mobj.group('start') + if start: + return datetime_from_str(start) + try: + return datetime_from_str('now-%s%s' % (mobj.group('time'), mobj.group('unit'))) + except ValueError: + return None + + def _extract_time_text(self, renderer, *path_list): + text = self._get_text(renderer, *path_list) or '' + dt = self.extract_relative_time(text) + timestamp = None + if isinstance(dt, datetime.datetime): + timestamp = calendar.timegm(dt.timetuple()) + + if timestamp is None: + timestamp = ( + unified_timestamp(text) or unified_timestamp( + self._search_regex( + (r'([a-z]+\s*\d{1,2},?\s*20\d{2})', r'(?:.+|^)(?:live|premieres|ed|ing)(?:\s*(?:on|for))?\s*(.+\d)'), + text.lower(), 'time text', default=None))) + + if text and timestamp is None: + self.report_warning(f"Cannot parse localized time text '{text}'" + bug_reports_message(), only_once=True) + return timestamp, 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'): @@ -617,13 +768,15 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 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) + if isinstance(e.cause, compat_HTTPError): + first_bytes = e.cause.read(512) + if not is_html(first_bytes): + yt_error = try_get( + self._parse_json( + self._webpage_read_content(e.cause, None, item_id, prefix=first_bytes) or '{}', 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 @@ -674,91 +827,58 @@ class YoutubeBaseInfoExtractor(InfoExtractor): 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)) + if duration is None: + duration = parse_duration(self._search_regex( + r'(?i)(ago)(?!.*\1)\s+(?P<duration>[a-z0-9 ,]+?)(?:\s+[\d,]+\s+views)?(?:\s+-\s+play\s+short)?$', + traverse_obj(renderer, ('title', 'accessibility', 'accessibilityData', 'label'), default='', expected_type=str), + video_id, default=None, group='duration')) + + view_count = self._get_count(renderer, 'viewCountText') uploader = self._get_text(renderer, 'ownerText', 'shortBylineText') + channel_id = traverse_obj( + renderer, ('shortBylineText', 'runs', ..., 'navigationEndpoint', 'browseEndpoint', 'browseId'), + expected_type=str, get_all=False) + timestamp, time_text = self._extract_time_text(renderer, 'publishedTimeText') + scheduled_timestamp = str_to_int(traverse_obj(renderer, ('upcomingEventData', 'startTime'), get_all=False)) + overlay_style = traverse_obj( + renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'), + get_all=False, expected_type=str) + badges = self._extract_badges(renderer) + thumbnails = self._extract_thumbnails(renderer, 'thumbnail') + navigation_url = urljoin('https://www.youtube.com/', traverse_obj( + renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'), + expected_type=str)) or '' + url = f'https://www.youtube.com/watch?v={video_id}' + if overlay_style == 'SHORTS' or '/shorts/' in navigation_url: + url = f'https://www.youtube.com/shorts/{video_id}' return { '_type': 'url', 'ie_key': YoutubeIE.ie_key(), 'id': video_id, - 'url': f'https://www.youtube.com/watch?v={video_id}', + 'url': url, 'title': title, 'description': description, 'duration': duration, 'view_count': view_count, 'uploader': uploader, + 'channel_id': channel_id, + 'thumbnails': thumbnails, + 'upload_date': (strftime_or_none(timestamp, '%Y%m%d') + if self._configuration_arg('approximate_date', ie_key='youtubetab') + else None), + 'live_status': ('is_upcoming' if scheduled_timestamp is not None + else 'was_live' if 'streamed' in time_text.lower() + else 'is_live' if overlay_style is not None and overlay_style == 'LIVE' or 'live now' in badges + else None), + 'release_timestamp': scheduled_timestamp, + 'availability': self._availability(needs_premium='premium' in badges, needs_subscription='members only' in badges) } class YoutubeIE(YoutubeBaseInfoExtractor): - IE_DESC = 'YouTube.com' - _INVIDIOUS_SITES = ( - # invidious-redirect websites - 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\.mastodon\.host', - r'(?:www\.)?invidious\.zapashcanon\.fr', - r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks', - r'(?:www\.)?invidious\.tinfoil-hat\.net', - r'(?:www\.)?invidious\.himiko\.cloud', - r'(?:www\.)?invidious\.reallyancient\.tech', - r'(?:www\.)?invidious\.tube', - r'(?:www\.)?invidiou\.site', - r'(?:www\.)?invidious\.site', - r'(?:www\.)?invidious\.xyz', - r'(?:www\.)?invidious\.nixnet\.xyz', - r'(?:www\.)?invidious\.048596\.xyz', - r'(?:www\.)?invidious\.drycat\.fr', - r'(?:www\.)?inv\.skyn3t\.in', - r'(?:www\.)?tube\.poal\.co', - r'(?:www\.)?tube\.connect\.cafe', - r'(?:www\.)?vid\.wxzm\.sx', - r'(?:www\.)?vid\.mint\.lgbt', - r'(?:www\.)?vid\.puffyan\.us', - r'(?:www\.)?yewtu\.be', - r'(?:www\.)?yt\.elukerio\.org', - r'(?:www\.)?yt\.lelux\.fi', - r'(?:www\.)?invidious\.ggc-project\.de', - r'(?:www\.)?yt\.maisputain\.ovh', - r'(?:www\.)?ytprivate\.com', - r'(?:www\.)?invidious\.13ad\.de', - 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', - r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion', - r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion', - 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', - ) + IE_DESC = 'YouTube' _VALID_URL = r"""(?x)^ ( (?:https?://|//) # http(s):// or protocol-independent URL @@ -772,7 +892,7 @@ 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|shorts)/(?!videoseries)) # v/ or embed/ or e/ or shorts/ + (?:(?:v|embed|e|shorts)/(?!videoseries|live_stream)) # 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 #! @@ -792,7 +912,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): (?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), + 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES), } _PLAYER_INFO_RE = ( r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player', @@ -923,18 +1043,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Philipp Hagemeister', 'uploader_id': 'phihag', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag', + 'channel': 'Philipp Hagemeister', '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 youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22', 'categories': ['Science & Technology'], 'tags': ['youtube-dl'], 'duration': 10, 'view_count': int, 'like_count': int, - 'dislike_count': int, + 'availability': 'public', + 'playable_in_embed': True, + 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg', + 'live_status': 'not_live', + 'age_limit': 0, 'start_time': 1, 'end_time': 9, + 'channel_follower_count': int } }, { @@ -963,14 +1089,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Philipp Hagemeister', 'uploader_id': 'phihag', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag', + 'channel': 'Philipp Hagemeister', + '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 youtube-dl.\n\nFor more information, contact phihag@phihag.de .', + 'description': 'md5:8fb536f4877b8a7455c2ec23794dbc22', 'categories': ['Science & Technology'], 'tags': ['youtube-dl'], 'duration': 10, 'view_count': int, 'like_count': int, - 'dislike_count': int, + 'availability': 'public', + 'playable_in_embed': True, + 'thumbnail': 'https://i.ytimg.com/vi/BaW_jenozKc/maxresdefault.jpg', + 'live_status': 'not_live', + 'age_limit': 0, + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1008,6 +1142,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'AfrojackVEVO', 'upload_date': '20131011', 'abr': 129.495, + 'like_count': int, + 'channel_id': 'UChuZAo1RKL85gev3Eal9_zg', + 'playable_in_embed': True, + 'channel_url': 'https://www.youtube.com/channel/UChuZAo1RKL85gev3Eal9_zg', + 'view_count': int, + 'track': 'The Spark', + 'live_status': 'not_live', + 'thumbnail': 'https://i.ytimg.com/vi_webp/IB3lcPjvWLA/maxresdefault.webp', + 'channel': 'Afrojack', + 'uploader_url': 'http://www.youtube.com/user/AfrojackVEVO', + 'tags': 'count:19', + 'availability': 'public', + 'categories': ['Music'], + 'age_limit': 0, + 'alt_title': 'The Spark', + 'channel_follower_count': int }, 'params': { 'youtube_include_dash_manifest': True, @@ -1029,6 +1179,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame', 'upload_date': '20140605', 'age_limit': 18, + 'categories': ['Gaming'], + 'thumbnail': 'https://i.ytimg.com/vi_webp/HtVdAasjOgU/maxresdefault.webp', + 'availability': 'needs_auth', + 'channel_url': 'https://www.youtube.com/channel/UCzybXLxv08IApdjdN0mJhEg', + 'like_count': int, + 'channel': 'The Witcher', + 'live_status': 'not_live', + 'tags': 'count:17', + 'channel_id': 'UCzybXLxv08IApdjdN0mJhEg', + 'playable_in_embed': True, + 'view_count': int, + 'channel_follower_count': int }, }, { @@ -1043,6 +1205,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'FlyingKitty900', 'uploader': 'FlyingKitty', 'age_limit': 18, + 'availability': 'needs_auth', + 'channel_id': 'UCYQT13AtrJC0gsM1far_zJg', + 'uploader_url': 'http://www.youtube.com/user/FlyingKitty900', + 'channel': 'FlyingKitty', + 'channel_url': 'https://www.youtube.com/channel/UCYQT13AtrJC0gsM1far_zJg', + 'view_count': int, + 'categories': ['Entertainment'], + 'live_status': 'not_live', + 'tags': ['Flyingkitty', 'godzilla 2'], + 'thumbnail': 'https://i.ytimg.com/vi/HsUATh_Nc2U/maxresdefault.jpg', + 'like_count': int, + 'duration': 177, + 'playable_in_embed': True, + 'channel_follower_count': int }, }, { @@ -1052,11 +1228,25 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'Tq92D6wQ1mg', 'title': '[MMD] Adios - EVERGLOW [+Motion DL]', 'ext': 'mp4', - 'upload_date': '20191227', + 'upload_date': '20191228', 'uploader_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ', 'uploader': 'Projekt Melody', 'description': 'md5:17eccca93a786d51bc67646756894066', 'age_limit': 18, + 'like_count': int, + 'availability': 'needs_auth', + 'uploader_url': 'http://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ', + 'channel_id': 'UC1yoRdFoFJaCY-AGfD9W0wQ', + 'view_count': int, + 'thumbnail': 'https://i.ytimg.com/vi_webp/Tq92D6wQ1mg/sddefault.webp', + 'channel': 'Projekt Melody', + 'live_status': 'not_live', + 'tags': ['mmd', 'dance', 'mikumikudance', 'kpop', 'vtuber'], + 'playable_in_embed': True, + 'categories': ['Entertainment'], + 'duration': 106, + 'channel_url': 'https://www.youtube.com/channel/UC1yoRdFoFJaCY-AGfD9W0wQ', + 'channel_follower_count': int }, }, { @@ -1070,6 +1260,25 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'st3in234', 'description': 'Fan Video. Music & Lyrics by OOMPH!.', 'upload_date': '20130730', + 'track': 'Such mich find mich', + 'age_limit': 0, + 'tags': ['oomph', 'such mich find mich', 'lyrics', 'german industrial', 'musica industrial'], + 'like_count': int, + 'playable_in_embed': False, + 'creator': 'OOMPH!', + 'thumbnail': 'https://i.ytimg.com/vi/MeJVWBSsPAY/sddefault.jpg', + 'view_count': int, + 'alt_title': 'Such mich find mich', + 'duration': 210, + 'channel': 'Herr Lurik', + 'channel_id': 'UCdR3RSDPqub28LjZx0v9-aA', + 'categories': ['Music'], + 'availability': 'public', + 'uploader_url': 'http://www.youtube.com/user/st3in234', + 'channel_url': 'https://www.youtube.com/channel/UCdR3RSDPqub28LjZx0v9-aA', + 'live_status': 'not_live', + 'artist': 'OOMPH!', + 'channel_follower_count': int }, }, { @@ -1093,6 +1302,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'deadmau5', 'title': 'Deadmau5 - Some Chords (HD)', 'alt_title': 'Some Chords', + 'availability': 'public', + 'tags': 'count:14', + 'channel_id': 'UCYEK6xds6eo-3tr4xRdflmQ', + 'view_count': int, + 'live_status': 'not_live', + 'channel': 'deadmau5', + 'thumbnail': 'https://i.ytimg.com/vi_webp/__2ABJjxzNo/maxresdefault.webp', + 'like_count': int, + 'track': 'Some Chords', + 'artist': 'deadmau5', + 'playable_in_embed': True, + 'age_limit': 0, + 'channel_url': 'https://www.youtube.com/channel/UCYEK6xds6eo-3tr4xRdflmQ', + 'categories': ['Music'], + 'album': 'Some Chords', + 'channel_follower_count': int }, 'expected_warnings': [ 'DASH manifest missing', @@ -1111,6 +1336,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'HO09 - Women - GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games', 'uploader': 'Olympics', 'title': 'Hockey - Women - GER-AUS - London 2012 Olympic Games', + 'like_count': int, + 'release_timestamp': 1343767800, + 'playable_in_embed': True, + 'categories': ['Sports'], + 'release_date': '20120731', + 'channel': 'Olympics', + 'tags': ['Hockey', '2012-07-31', '31 July 2012', 'Riverbank Arena', 'Session', 'Olympics', 'Olympic Games', 'London 2012', '2012 Summer Olympics', 'Summer Games'], + 'channel_id': 'UCTl3QQTvqHFjurroKxexy2Q', + 'thumbnail': 'https://i.ytimg.com/vi/lqQg6PlCWgI/maxresdefault.jpg', + 'age_limit': 0, + 'availability': 'public', + 'live_status': 'was_live', + 'view_count': int, + 'channel_url': 'https://www.youtube.com/channel/UCTl3QQTvqHFjurroKxexy2Q', + 'channel_follower_count': int }, 'params': { 'skip_download': 'requires avconv', @@ -1130,6 +1370,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'description': 'made by Wacom from Korea | 字幕&加油添醋 by TY\'s Allen | 感謝heylisa00cavey1001同學熱情提供梗及翻譯', 'uploader': '孫ᄋᄅ', 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人', + 'playable_in_embed': True, + 'channel': '孫ᄋᄅ', + 'age_limit': 0, + 'tags': 'count:11', + 'channel_url': 'https://www.youtube.com/channel/UCS-xxCmRaA6BFdmgDPA_BIw', + 'channel_id': 'UCS-xxCmRaA6BFdmgDPA_BIw', + 'thumbnail': 'https://i.ytimg.com/vi/_b-2C3KPAM0/maxresdefault.jpg', + 'view_count': int, + 'categories': ['People & Blogs'], + 'like_count': int, + 'live_status': 'not_live', + 'availability': 'unlisted', + 'channel_follower_count': int }, }, # url_encoded_fmt_stream_map is empty string @@ -1286,6 +1539,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'track': 'Dark Walk', 'artist': 'Todd Haberman;\nDaniel Law Heath and Aaron Kaplan', 'album': 'Position Music - Production Music Vol. 143 - Dark Walk', + 'thumbnail': 'https://i.ytimg.com/vi_webp/lsguqyKfVQg/maxresdefault.webp', + 'categories': ['Film & Animation'], + 'view_count': int, + 'live_status': 'not_live', + 'channel_url': 'https://www.youtube.com/channel/UCTSRgz5jylBvFt_S7wnsqLQ', + 'channel_id': 'UCTSRgz5jylBvFt_S7wnsqLQ', + 'tags': 'count:13', + 'availability': 'public', + 'channel': 'IronSoulElf', + 'playable_in_embed': True, + 'like_count': int, + 'age_limit': 0, + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1327,11 +1593,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'md5:e41008789470fc2533a3252216f1c1d1', 'description': 'md5:a677553cf0840649b731a3024aeff4cc', 'duration': 721, - 'upload_date': '20150127', + 'upload_date': '20150128', 'uploader_id': 'BerkmanCenter', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter', 'uploader': 'The Berkman Klein Center for Internet & Society', 'license': 'Creative Commons Attribution license (reuse allowed)', + 'channel_id': 'UCuLGmD72gJDBwmLw06X58SA', + 'channel_url': 'https://www.youtube.com/channel/UCuLGmD72gJDBwmLw06X58SA', + 'like_count': int, + 'age_limit': 0, + 'tags': ['Copyright (Legal Subject)', 'Law (Industry)', 'William W. Fisher (Author)'], + 'channel': 'The Berkman Klein Center for Internet & Society', + 'availability': 'public', + 'view_count': int, + 'categories': ['Education'], + 'thumbnail': 'https://i.ytimg.com/vi_webp/M4gD1WSo5mA/maxresdefault.webp', + 'live_status': 'not_live', + 'playable_in_embed': True, + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1346,11 +1625,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders', 'description': 'md5:13a2503d7b5904ef4b223aa101628f39', 'duration': 4060, - 'upload_date': '20151119', + 'upload_date': '20151120', 'uploader': 'Bernie Sanders', 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg', 'license': 'Creative Commons Attribution license (reuse allowed)', + 'playable_in_embed': True, + 'tags': 'count:12', + 'like_count': int, + 'channel_id': 'UCH1dpzjCEiGAt8CXkryhkZg', + 'age_limit': 0, + 'availability': 'public', + 'categories': ['News & Politics'], + 'channel': 'Bernie Sanders', + 'thumbnail': 'https://i.ytimg.com/vi_webp/eQcmzGIKrzg/maxresdefault.webp', + 'view_count': int, + 'live_status': 'not_live', + 'channel_url': 'https://www.youtube.com/channel/UCH1dpzjCEiGAt8CXkryhkZg', + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1400,6 +1692,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'series': 'Mind Field', 'season_number': 1, 'episode_number': 1, + 'thumbnail': 'https://i.ytimg.com/vi_webp/iqKdEhx-dD4/maxresdefault.webp', + 'tags': 'count:12', + 'view_count': int, + 'availability': 'public', + 'age_limit': 0, + 'channel': 'Vsauce', + 'episode': 'Episode 1', + 'categories': ['Entertainment'], + 'season': 'Season 1', + 'channel_id': 'UC6nSFpj9HTCZ5t-N3Rm3-HA', + 'channel_url': 'https://www.youtube.com/channel/UC6nSFpj9HTCZ5t-N3Rm3-HA', + 'like_count': int, + 'playable_in_embed': True, + 'live_status': 'not_live', + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1493,6 +1800,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'album': 'it\'s too much love to know my dear', 'release_date': '20190313', 'release_year': 2019, + 'alt_title': 'Voyeur Girl', + 'view_count': int, + 'uploader_url': 'http://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA', + 'playable_in_embed': True, + 'like_count': int, + 'categories': ['Music'], + 'channel_url': 'https://www.youtube.com/channel/UC-pWHpBjdGG69N9mM2auIAA', + 'channel': 'Stephen', + 'availability': 'public', + 'creator': 'Stephen', + 'duration': 169, + 'thumbnail': 'https://i.ytimg.com/vi_webp/MgNrAu2pzNs/maxresdefault.webp', + 'age_limit': 0, + 'channel_id': 'UC-pWHpBjdGG69N9mM2auIAA', + 'tags': 'count:11', + 'live_status': 'not_live', + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1534,6 +1858,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20170613', 'uploader_id': 'ElevageOrVert', 'uploader': 'ElevageOrVert', + 'view_count': int, + 'thumbnail': 'https://i.ytimg.com/vi_webp/x41yOUIvK2k/maxresdefault.webp', + 'uploader_url': 'http://www.youtube.com/user/ElevageOrVert', + 'like_count': int, + 'channel_id': 'UCo03ZQPBW5U4UC3regpt1nw', + 'tags': [], + 'channel_url': 'https://www.youtube.com/channel/UCo03ZQPBW5U4UC3regpt1nw', + 'availability': 'public', + 'age_limit': 0, + 'categories': ['Pets & Animals'], + 'duration': 7, + 'playable_in_embed': True, + 'live_status': 'not_live', + 'channel': 'ElevageOrVert', + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1553,6 +1892,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20130831', 'uploader_id': 'kudvenkat', 'uploader': 'kudvenkat', + 'channel_id': 'UCCTVrRB5KpIiK6V2GGVsR1Q', + 'like_count': int, + 'uploader_url': 'http://www.youtube.com/user/kudvenkat', + 'channel_url': 'https://www.youtube.com/channel/UCCTVrRB5KpIiK6V2GGVsR1Q', + 'live_status': 'not_live', + 'categories': ['Education'], + 'availability': 'public', + 'thumbnail': 'https://i.ytimg.com/vi/CHqg6qOn4no/sddefault.jpg', + 'tags': 'count:12', + 'playable_in_embed': True, + 'age_limit': 0, + 'view_count': int, + 'duration': 522, + 'channel': 'kudvenkat', + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1582,8 +1936,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'artist': 'The Cinematic Orchestra', 'track': 'Burn Out', 'album': 'Every Day', - 'release_data': None, - 'release_year': None, + 'like_count': int, + 'live_status': 'not_live', + 'alt_title': 'Burn Out', + 'duration': 614, + 'age_limit': 0, + 'view_count': int, + 'channel_url': 'https://www.youtube.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw', + 'creator': 'The Cinematic Orchestra', + 'channel': 'The Cinematic Orchestra', + 'tags': ['The Cinematic Orchestra', 'Every Day', 'Burn Out'], + 'channel_id': 'UCIzsJBIyo8hhpFm1NK0uLgw', + 'availability': 'public', + 'thumbnail': 'https://i.ytimg.com/vi/OtqTfy26tG0/maxresdefault.jpg', + 'categories': ['Music'], + 'playable_in_embed': True, + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1602,10 +1970,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'ext': 'mp4', 'title': 'San Diego teen commits suicide after bullying over embarrassing video', 'channel_id': 'UC-SJ6nODDmufqBzPBwCvYvQ', - 'uploader': 'CBS This Morning', + 'uploader': 'CBS Mornings', 'uploader_id': 'CBSThisMorning', 'upload_date': '20140716', - 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7' + 'description': 'md5:acde3a73d3f133fc97e837a9f76b53b7', + 'duration': 170, + 'categories': ['News & Politics'], + 'uploader_url': 'http://www.youtube.com/user/CBSThisMorning', + 'view_count': int, + 'channel': 'CBS Mornings', + 'tags': ['suicide', 'bullying', 'video', 'cbs', 'news'], + 'thumbnail': 'https://i.ytimg.com/vi/SZJvDhaSDnc/hqdefault.jpg', + 'age_limit': 18, + 'availability': 'needs_auth', + 'channel_url': 'https://www.youtube.com/channel/UC-SJ6nODDmufqBzPBwCvYvQ', + 'like_count': int, + 'live_status': 'not_live', + 'playable_in_embed': True, + 'channel_follower_count': int } }, { @@ -1620,6 +2002,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Walk around Japan', 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw', + 'duration': 1456, + 'categories': ['Travel & Events'], + 'channel_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw', + 'view_count': int, + 'channel': 'Walk around Japan', + 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'], + 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp', + 'age_limit': 0, + 'availability': 'public', + 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw', + 'live_status': 'not_live', + 'playable_in_embed': True, + 'channel_follower_count': int }, 'params': { 'skip_download': True, @@ -1648,7 +2043,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'colinfurze', 'uploader_id': 'colinfurze', 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCp68_FLety0O-n9QU6phsgw', - 'description': 'md5:b5096f56af7ccd7a555c84db81738b22' + 'description': 'md5:5d5991195d599b56cd0c4148907eec50', + 'duration': 596, + 'categories': ['Entertainment'], + 'uploader_url': 'http://www.youtube.com/user/colinfurze', + 'view_count': int, + 'channel': 'colinfurze', + 'tags': ['Colin', 'furze', 'Terry', 'tunnel', 'underground', 'bunker'], + 'thumbnail': 'https://i.ytimg.com/vi/YOelRv7fMxY/maxresdefault.jpg', + 'age_limit': 0, + 'availability': 'public', + 'like_count': int, + 'live_status': 'not_live', + 'playable_in_embed': True, + 'channel_follower_count': int }, 'params': { 'format': '17', # 3gp format available on android @@ -1666,6 +2074,120 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # shorts 'url': 'https://www.youtube.com/shorts/BGQWPY4IigY', 'only_matching': True, + }, { + 'note': 'Storyboards', + 'url': 'https://www.youtube.com/watch?v=5KLPxDtMqe8', + 'info_dict': { + 'id': '5KLPxDtMqe8', + 'ext': 'mhtml', + 'format_id': 'sb0', + 'title': 'Your Brain is Plastic', + 'uploader_id': 'scishow', + 'description': 'md5:89cd86034bdb5466cd87c6ba206cd2bc', + 'upload_date': '20140324', + 'uploader': 'SciShow', + 'like_count': int, + 'channel_id': 'UCZYTClx2T1of7BRZ86-8fow', + 'channel_url': 'https://www.youtube.com/channel/UCZYTClx2T1of7BRZ86-8fow', + 'view_count': int, + 'thumbnail': 'https://i.ytimg.com/vi/5KLPxDtMqe8/maxresdefault.jpg', + 'playable_in_embed': True, + 'tags': 'count:12', + 'uploader_url': 'http://www.youtube.com/user/scishow', + 'availability': 'public', + 'channel': 'SciShow', + 'live_status': 'not_live', + 'duration': 248, + 'categories': ['Education'], + 'age_limit': 0, + 'channel_follower_count': int + }, 'params': {'format': 'mhtml', 'skip_download': True} + }, { + # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939) + 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4', + 'info_dict': { + 'id': '2NUZ8W2llS4', + 'ext': 'mp4', + 'title': 'The NP that test your phone performance 🙂', + 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d', + 'uploader': 'Leon Nguyen', + 'uploader_id': 'VNSXIII', + 'uploader_url': 'http://www.youtube.com/user/VNSXIII', + 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA', + 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA', + 'duration': 21, + 'view_count': int, + 'age_limit': 0, + 'categories': ['Gaming'], + 'tags': 'count:23', + 'playable_in_embed': True, + 'live_status': 'not_live', + 'upload_date': '20220103', + 'like_count': int, + 'availability': 'public', + 'channel': 'Leon Nguyen', + 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp', + 'channel_follower_count': int + } + }, { + # date text is premiered video, ensure upload date in UTC (published 1641172509) + 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM', + 'info_dict': { + 'id': 'mzZzzBU6lrM', + 'ext': 'mp4', + 'title': 'I Met GeorgeNotFound In Real Life...', + 'description': 'md5:cca98a355c7184e750f711f3a1b22c84', + 'uploader': 'Quackity', + 'uploader_id': 'QuackityHQ', + 'uploader_url': 'http://www.youtube.com/user/QuackityHQ', + 'channel_id': 'UC_8NknAFiyhOUaZqHR3lq3Q', + 'channel_url': 'https://www.youtube.com/channel/UC_8NknAFiyhOUaZqHR3lq3Q', + 'duration': 955, + 'view_count': int, + 'age_limit': 0, + 'categories': ['Entertainment'], + 'tags': 'count:26', + 'playable_in_embed': True, + 'live_status': 'not_live', + 'release_timestamp': 1641172509, + 'release_date': '20220103', + 'upload_date': '20220103', + 'like_count': int, + 'availability': 'public', + 'channel': 'Quackity', + 'thumbnail': 'https://i.ytimg.com/vi/mzZzzBU6lrM/maxresdefault.jpg', + 'channel_follower_count': int + } + }, + { # continuous livestream. Microformat upload date should be preferred. + # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27 + 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU', + 'info_dict': { + 'id': 'kgx4WGK0oNU', + 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}', + 'ext': 'mp4', + 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA', + 'availability': 'public', + 'age_limit': 0, + 'release_timestamp': 1637975704, + 'upload_date': '20210619', + 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA', + 'live_status': 'is_live', + 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg', + 'uploader': '阿鲍Abao', + 'uploader_url': 'http://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA', + 'channel': 'Abao in Tokyo', + 'channel_follower_count': int, + 'release_date': '20211127', + 'tags': 'count:39', + 'categories': ['People & Blogs'], + 'like_count': int, + 'uploader_id': 'UC84whx2xxsiA1gXHXXqKGOA', + 'view_count': int, + 'playable_in_embed': True, + 'description': 'md5:2ef1d002cad520f65825346e2084e49d', + }, + 'params': {'skip_download': True} }, ] @@ -1683,18 +2205,158 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self._code_cache = {} self._player_cache = {} + def _prepare_live_from_start_formats(self, formats, video_id, live_start_time, url, webpage_url, smuggled_data): + lock = threading.Lock() + + is_live = True + start_time = time.time() + formats = [f for f in formats if f.get('is_from_start')] + + def refetch_manifest(format_id, delay): + nonlocal formats, start_time, is_live + if time.time() <= start_time + delay: + return + + _, _, prs, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url) + video_details = traverse_obj( + prs, (..., 'videoDetails'), expected_type=dict, default=[]) + microformats = traverse_obj( + prs, (..., 'microformat', 'playerMicroformatRenderer'), + expected_type=dict, default=[]) + _, is_live, _, formats = self._list_formats(video_id, microformats, video_details, prs, player_url) + start_time = time.time() + + def mpd_feed(format_id, delay): + """ + @returns (manifest_url, manifest_stream_number, is_live) or None + """ + with lock: + refetch_manifest(format_id, delay) + + f = next((f for f in formats if f['format_id'] == format_id), None) + if not f: + if not is_live: + self.to_screen(f'{video_id}: Video is no longer live') + else: + self.report_warning( + f'Cannot find refreshed manifest for format {format_id}{bug_reports_message()}') + return None + return f['manifest_url'], f['manifest_stream_number'], is_live + + for f in formats: + f['is_live'] = True + f['protocol'] = 'http_dash_segments_generator' + f['fragments'] = functools.partial( + self._live_dash_fragments, f['format_id'], live_start_time, mpd_feed) + + def _live_dash_fragments(self, format_id, live_start_time, mpd_feed, ctx): + FETCH_SPAN, MAX_DURATION = 5, 432000 + + mpd_url, stream_number, is_live = None, None, True + + begin_index = 0 + download_start_time = ctx.get('start') or time.time() + + lack_early_segments = download_start_time - (live_start_time or download_start_time) > MAX_DURATION + if lack_early_segments: + self.report_warning(bug_reports_message( + 'Starting download from the last 120 hours of the live stream since ' + 'YouTube does not have data before that. If you think this is wrong,'), only_once=True) + lack_early_segments = True + + known_idx, no_fragment_score, last_segment_url = begin_index, 0, None + fragments, fragment_base_url = None, None + + def _extract_sequence_from_mpd(refresh_sequence, immediate): + nonlocal mpd_url, stream_number, is_live, no_fragment_score, fragments, fragment_base_url + # Obtain from MPD's maximum seq value + old_mpd_url = mpd_url + last_error = ctx.pop('last_error', None) + expire_fast = immediate or last_error and isinstance(last_error, compat_HTTPError) and last_error.code == 403 + mpd_url, stream_number, is_live = (mpd_feed(format_id, 5 if expire_fast else 18000) + or (mpd_url, stream_number, False)) + if not refresh_sequence: + if expire_fast and not is_live: + return False, last_seq + elif old_mpd_url == mpd_url: + return True, last_seq + try: + fmts, _ = self._extract_mpd_formats_and_subtitles( + mpd_url, None, note=False, errnote=False, fatal=False) + except ExtractorError: + fmts = None + if not fmts: + no_fragment_score += 2 + return False, last_seq + fmt_info = next(x for x in fmts if x['manifest_stream_number'] == stream_number) + fragments = fmt_info['fragments'] + fragment_base_url = fmt_info['fragment_base_url'] + assert fragment_base_url + + _last_seq = int(re.search(r'(?:/|^)sq/(\d+)', fragments[-1]['path']).group(1)) + return True, _last_seq + + while is_live: + fetch_time = time.time() + if no_fragment_score > 30: + return + if last_segment_url: + # Obtain from "X-Head-Seqnum" header value from each segment + try: + urlh = self._request_webpage( + last_segment_url, None, note=False, errnote=False, fatal=False) + except ExtractorError: + urlh = None + last_seq = try_get(urlh, lambda x: int_or_none(x.headers['X-Head-Seqnum'])) + if last_seq is None: + no_fragment_score += 2 + last_segment_url = None + continue + else: + should_continue, last_seq = _extract_sequence_from_mpd(True, no_fragment_score > 15) + no_fragment_score += 2 + if not should_continue: + continue + + if known_idx > last_seq: + last_segment_url = None + continue + + last_seq += 1 + + if begin_index < 0 and known_idx < 0: + # skip from the start when it's negative value + known_idx = last_seq + begin_index + if lack_early_segments: + known_idx = max(known_idx, last_seq - int(MAX_DURATION // fragments[-1]['duration'])) + try: + for idx in range(known_idx, last_seq): + # do not update sequence here or you'll get skipped some part of it + should_continue, _ = _extract_sequence_from_mpd(False, False) + if not should_continue: + known_idx = idx - 1 + raise ExtractorError('breaking out of outer loop') + last_segment_url = urljoin(fragment_base_url, 'sq/%d' % idx) + yield { + 'url': last_segment_url, + } + if known_idx == last_seq: + no_fragment_score += 5 + else: + no_fragment_score = 0 + known_idx = last_seq + except ExtractorError: + continue + + time.sleep(max(0, FETCH_SPAN + fetch_time - time.time())) + 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 + return urljoin('https://www.youtube.com', player_url) def _download_player_url(self, video_id, fatal=False): res = self._download_webpage( @@ -1720,7 +2382,7 @@ 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: + def _load_player(self, video_id, player_url, fatal=True): player_id = self._extract_player_info(player_url) if player_id not in self._code_cache: code = self._download_webpage( @@ -1729,7 +2391,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): errnote='Download of %s failed' % player_url) if code: self._code_cache[player_id] = code - return player_id in self._code_cache + return self._code_cache.get(player_id) def _extract_signature_function(self, video_id, player_url, example_sig): player_id = self._extract_player_info(player_url) @@ -1743,8 +2405,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if cache_spec is not None: return lambda s: ''.join(s[i] for i in cache_spec) - if self._load_player(video_id, player_url): - code = self._code_cache[player_id] + code = self._load_player(video_id, player_url) + if code: res = self._parse_sig_js(code) test_string = ''.join(map(compat_chr, range(len(example_sig)))) @@ -1755,6 +2417,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return res def _print_sig_code(self, func, example_sig): + if not self.get_param('youtube_print_sig_code'): + return + def gen_sig_code(idxs): def _genslice(start, end, step): starts = '' if start == 0 else str(start) @@ -1831,13 +2496,59 @@ class YoutubeIE(YoutubeBaseInfoExtractor): ) self._player_cache[player_id] = func func = self._player_cache[player_id] - if self.get_param('youtube_print_sig_code'): - self._print_sig_code(func, s) + self._print_sig_code(func, s) return func(s) except Exception as e: - tb = traceback.format_exc() - raise ExtractorError( - 'Signature extraction failed: ' + tb, cause=e) + raise ExtractorError('Signature extraction failed: ' + traceback.format_exc(), cause=e) + + def _decrypt_nsig(self, s, video_id, player_url): + """Turn the encrypted n field into a working signature""" + if player_url is None: + raise ExtractorError('Cannot decrypt nsig without player_url') + player_url = urljoin('https://www.youtube.com', player_url) + + sig_id = ('nsig_value', s) + if sig_id in self._player_cache: + return self._player_cache[sig_id] + + try: + player_id = ('nsig', player_url) + if player_id not in self._player_cache: + self._player_cache[player_id] = self._extract_n_function(video_id, player_url) + func = self._player_cache[player_id] + self._player_cache[sig_id] = func(s) + self.write_debug(f'Decrypted nsig {s} => {self._player_cache[sig_id]}') + return self._player_cache[sig_id] + except Exception as e: + raise ExtractorError(traceback.format_exc(), cause=e, video_id=video_id) + + def _extract_n_function_name(self, jscode): + nfunc, idx = self._search_regex( + r'\.get\("n"\)\)&&\(b=(?P<nfunc>[a-zA-Z0-9$]+)(?:\[(?P<idx>\d+)\])?\([a-zA-Z0-9]\)', + jscode, 'Initial JS player n function name', group=('nfunc', 'idx')) + if not idx: + return nfunc + return json.loads(js_to_json(self._search_regex( + rf'var {re.escape(nfunc)}\s*=\s*(\[.+?\]);', jscode, + f'Initial JS player n function list ({nfunc}.{idx})')))[int(idx)] + + def _extract_n_function(self, video_id, player_url): + player_id = self._extract_player_info(player_url) + func_code = self._downloader.cache.load('youtube-nsig', player_id) + + if func_code: + jsi = JSInterpreter(func_code) + else: + jscode = self._load_player(video_id, player_url) + funcname = self._extract_n_function_name(jscode) + jsi = JSInterpreter(jscode) + func_code = jsi.extract_function_code(funcname) + self._downloader.cache.store('youtube-nsig', player_id, func_code) + + if self.get_param('youtube_print_sig_code'): + self.to_screen(f'Extracted nsig function from {player_id}:\n{func_code[1]}\n') + + return lambda s: jsi.extract_function_from_code(*func_code)([s]) def _extract_signature_timestamp(self, video_id, player_url, ytcfg=None, fatal=False): """ @@ -1856,18 +2567,17 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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] + code = self._load_player(video_id, player_url, fatal=fatal) + if code: 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) + playback_url = get_first( + player_responses, ('playbackTracking', 'videostatsPlaybackUrl', 'baseUrl'), + expected_type=url_or_none) if not playback_url: self.report_warning('Unable to mark watched') return @@ -1991,19 +2701,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE), regex), webpage, name, default='{}'), 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: @@ -2012,10 +2709,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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()) + timestamp, time_text = self._extract_time_text(comment_renderer, 'publishedTimeText') author = self._get_text(comment_renderer, 'authorText') author_id = try_get(comment_renderer, lambda x: x['authorEndpoint']['browseEndpoint']['browseId'], compat_str) @@ -2042,20 +2736,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'parent': parent or 'root' } - def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, comment_counts=None): + def _comment_entries(self, root_continuation_data, ytcfg, video_id, parent=None, tracker=None): + + get_single_config_arg = lambda c: self._configuration_arg(c, [''])[0] 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)) + comments_header_renderer = traverse_obj(content, 'commentsHeaderRenderer') + expected_comment_count = self._get_count( + comments_header_renderer, 'countText', 'commentsCount') 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 + tracker['est_total'] = expected_comment_count + self.to_screen(f'Downloading ~{expected_comment_count} comments') + comment_sort_index = int(get_single_config_arg('comment_sort') != 'top') # 1 = new, 0 = top sort_menu_item = try_get( comments_header_renderer, @@ -2066,76 +2761,84 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if not _continuation: continue - sort_text = sort_menu_item.get('title') - if isinstance(sort_text, compat_str): - sort_text = sort_text.lower() - else: + sort_text = str_or_none(sort_menu_item.get('title')) + if not sort_text: sort_text = 'top comments' if comment_sort_index == 0 else 'newest first' - self.to_screen('Sorting comments by %s' % sort_text) + self.to_screen('Sorting comments by %s' % sort_text.lower()) break return _continuation def extract_thread(contents): if not parent: - comment_counts[2] = 0 + tracker['current_page_thread'] = 0 for content in contents: + if not parent and tracker['total_parent_comments'] >= max_parents: + yield 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)) + comment_renderer = get_first( + (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]], + expected_type=dict, default={}) - if not comment_renderer: - continue comment = self._extract_comment(comment_renderer, parent) if not comment: continue - comment_counts[0] += 1 + + tracker['running_total'] += 1 + tracker['total_reply_comments' if parent else 'total_parent_comments'] += 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 + tracker['current_page_thread'] += 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: + parent=comment.get('id'), tracker=tracker) + for reply_comment in itertools.islice(comment_entries_iter, min(max_replies_per_thread, max(0, max_replies - tracker['total_reply_comments']))): yield reply_comment + # Keeps track of counts across recursive calls + if not tracker: + tracker = dict( + running_total=0, + est_total=0, + current_page_thread=0, + total_parent_comments=0, + total_reply_comments=0) + + # TODO: Deprecated # YouTube comments have a max depth of 2 - max_depth = int_or_none(self._configuration_arg('max_comment_depth', [''])[0]) or float('inf') + max_depth = int_or_none(get_single_config_arg('max_comment_depth')) + if max_depth: + self._downloader.deprecation_warning( + '[youtube] max_comment_depth extractor argument is deprecated. Set max replies in the max-comments extractor argument instead.') 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) + max_comments, max_parents, max_replies, max_replies_per_thread, *_ = map( + lambda p: int_or_none(p, default=sys.maxsize), self._configuration_arg('max_comments', ) + [''] * 4) + continuation = self._extract_continuation(root_continuation_data) 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 + response = 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]) + headers = self.generate_api_headers(ytcfg=ytcfg, visitor_data=self._extract_visitor_data(response)) + comment_prog_str = f"({tracker['running_total']}/{tracker['est_total']})" 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) + tracker['current_page_thread'], comment_prog_str) else: note_prefix = '%sDownloading comment%s API JSON page %d %s' % ( ' ' if parent else '', ' replies' if parent else '', @@ -2144,83 +2847,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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 + check_get_keys='onResponseReceivedEndpoints') - continuation_contents = dict_get(response, ('onResponseReceivedEndpoints', 'continuationContents')) + continuation_contents = traverse_obj( + response, 'onResponseReceivedEndpoints', expected_type=list, default=[]) 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}) + for continuation_section in continuation_contents: + continuation_items = traverse_obj( + continuation_section, + (('reloadContinuationItemsCommand', 'appendContinuationItemsAction'), 'continuationItems'), + get_all=False, expected_type=list) or [] + if is_first_continuation: + continuation = extract_header(continuation_items) + is_first_continuation = False 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 + continue - # 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 + for entry in extract_thread(continuation_items): + if not entry: + return + yield entry + continuation = self._extract_continuation({'contents': continuation_items}) + if continuation: 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): @@ -2230,11 +2882,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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 @@ -2290,18 +2937,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _get_requested_clients(self, url, smuggled_data): requested_clients = [] + default = ['android', 'web'] 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 == 'default': + requested_clients.extend(default) 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'] + requested_clients = default if smuggled_data.get('is_music_url') or self.is_music_url(url): requested_clients.extend( @@ -2316,7 +2966,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }.get(client) if not url: return {} - webpage = self._download_webpage(url, video_id, fatal=False, note=f'Downloading {client} config') + webpage = self._download_webpage(url, video_id, fatal=False, note='Downloading %s config' % client.replace('_', ' ').strip()) return self.extract_ytcfg(video_id, webpage) or {} def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg): @@ -2326,13 +2976,19 @@ class YoutubeIE(YoutubeBaseInfoExtractor): webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, video_id, 'initial player response') - original_clients = clients + all_clients = set(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) + def append_client(*client_names): + """ Append the first client name that exists but not already used """ + for client_name in client_names: + actual_client = _split_innertube_client(client_name)[0] + if actual_client in INNERTUBE_CLIENTS: + if actual_client not in all_clients: + clients.append(client_name) + all_clients.add(actual_client) + return # Android player_response does not have microFormats which are needed for # extraction of some data. So we return the initial_pr with formats @@ -2347,7 +3003,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): tried_iframe_fallback = False player_url = None while clients: - client = clients.pop() + client, base_client, variant = _split_innertube_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 @@ -2375,10 +3031,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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')) + if variant == 'embedded' and self._is_unplayable(pr) and self.is_authenticated: + append_client(f'{base_client}_creator') elif self._is_agegated(pr): - append_client(f'{client}_agegate') + if variant == 'tv_embedded': + append_client(f'{base_client}_embedded') + elif not variant: + append_client(f'tv_embedded.{base_client}', f'{base_client}_embedded') if last_error: if not len(prs): @@ -2386,8 +3045,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self.report_warning(last_error) return prs, player_url - def _extract_formats(self, streaming_data, video_id, player_url, is_live): - itags, stream_ids = [], [] + def _extract_formats(self, streaming_data, video_id, player_url, is_live, duration): + itags, stream_ids = {}, [] itag_qualities, res_qualities = {}, {} q = qualities([ # Normally tiny is the smallest video-only formats. But @@ -2399,7 +3058,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...), default=[]) for fmt in streaming_formats: - if fmt.get('targetDurationSec') or fmt.get('drmFamilies'): + if fmt.get('targetDurationSec'): continue itag = str_or_none(fmt.get('itag')) @@ -2440,28 +3099,56 @@ class YoutubeIE(YoutubeBaseInfoExtractor): sp = try_get(sc, lambda x: x['sp'][0]) or 'signature' fmt_url += '&' + sp + '=' + signature + query = parse_qs(fmt_url) + throttled = False + if query.get('n'): + try: + fmt_url = update_url_query(fmt_url, { + 'n': self._decrypt_nsig(query['n'][0], video_id, player_url)}) + except ExtractorError as e: + self.report_warning( + f'nsig extraction failed: You may experience throttling for some formats\n' + f'n = {query["n"][0]} ; player = {player_url}\n{e}', only_once=True) + throttled = True + if itag: - itags.append(itag) + itags[itag] = 'https' stream_ids.append(stream_id) - tbr = float_or_none( - fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) + tbr = float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) + language_preference = ( + 10 if audio_track.get('audioIsDefault') and 10 + else -10 if 'descriptive' in (audio_track.get('displayName') or '').lower() and -10 + else -1) + # Some formats may have much smaller duration than others (possibly damaged during encoding) + # Eg: 2-nOtRESiUc Ref: https://github.com/hypervideo/hypervideo/issues/2823 + # Make sure to avoid false positives with small duration differences. + # Eg: __2ABJjxzNo, ySuUZEjARPY + is_damaged = try_get(fmt, lambda x: float(x['approxDurationMs']) / duration < 500) + if is_damaged: + self.report_warning(f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True) dct = { 'asr': int_or_none(fmt.get('audioSampleRate')), 'filesize': int_or_none(fmt.get('contentLength')), 'format_id': itag, - 'format_note': ', '.join(filter(None, ( + 'format_note': join_nonempty( '%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')), + ' (default)' if language_preference > 0 else ''), + fmt.get('qualityLabel') or quality.replace('audio_quality_', ''), + throttled and 'THROTTLED', is_damaged and 'DAMAGED', delim=', '), + 'source_preference': -10 if throttled else -1, + 'fps': int_or_none(fmt.get('fps')) or None, 'height': height, 'quality': q(quality), + 'has_drm': bool(fmt.get('drmFamilies')), 'tbr': tbr, 'url': fmt_url, 'width': int_or_none(fmt.get('width')), - 'language': audio_track.get('id', '').split('.')[0], - 'language_preference': 1 if audio_track.get('audioIsDefault') else -1, + 'language': join_nonempty(audio_track.get('id', '').split('.')[0], + 'desc' if language_preference < -1 else ''), + 'language_preference': language_preference, + # Strictly de-prioritize damaged and 3gp formats + 'preference': -10 if is_damaged else -2 if itag == '17' else None, } mime_mobj = re.match( r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '') @@ -2483,59 +3170,84 @@ class YoutubeIE(YoutubeBaseInfoExtractor): dct['container'] = dct['ext'] + '_dash' yield dct + live_from_start = is_live and self.get_param('live_from_start') 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) + if not self.get_param('youtube_include_hls_manifest', True): + skip_manifests.append('hls') + get_dash = 'dash' not in skip_manifests and ( + not is_live or live_from_start or self._configuration_arg('include_live_dash')) + get_hls = not live_from_start and 'hls' not in skip_manifests + + def process_manifest_format(f, proto, itag): + if itag in itags: + if itags[itag] == proto or f'{itag}-{proto}' in itags: + return False + itag = f'{itag}-{proto}' + if itag: + f['format_id'] = itag + itags[itag] = proto - 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 + f['quality'] = next(( + q(qdict[val]) + for val, qdict in ((f.get('format_id', '').split('-')[0], itag_qualities), (f.get('height'), res_qualities)) + if val in qdict), -1) + return True 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 process_manifest_format(f, 'hls', self._search_regex( + r'/itag/(\d+)', f['url'], 'itag', default=None)): + yield f 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): - itag = f['format_id'] - if itag in itags: - 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 - yield f - - def _real_extract(self, url): - url, smuggled_data = unsmuggle_url(url, {}) - video_id = self._match_id(url) + if process_manifest_format(f, 'dash', f['format_id']): + f['filesize'] = int_or_none(self._search_regex( + r'/clen/(\d+)', f.get('fragment_base_url') or f['url'], 'file size', default=None)) + if live_from_start: + f['is_from_start'] = True + + yield f + + def _extract_storyboard(self, player_responses, duration): + spec = get_first( + player_responses, ('storyboards', 'playerStoryboardSpecRenderer', 'spec'), default='').split('|')[::-1] + base_url = url_or_none(urljoin('https://i.ytimg.com/', spec.pop() or None)) + if not base_url: + return + L = len(spec) - 1 + for i, args in enumerate(spec): + args = args.split('#') + counts = list(map(int_or_none, args[:5])) + if len(args) != 8 or not all(counts): + self.report_warning(f'Malformed storyboard {i}: {"#".join(args)}{bug_reports_message()}') + continue + width, height, frame_count, cols, rows = counts + N, sigh = args[6:] + + url = base_url.replace('$L', str(L - i)).replace('$N', N) + f'&sigh={sigh}' + fragment_count = frame_count / (cols * rows) + fragment_duration = duration / fragment_count + yield { + 'format_id': f'sb{i}', + 'format_note': 'storyboard', + 'ext': 'mhtml', + 'protocol': 'mhtml', + 'acodec': 'none', + 'vcodec': 'none', + 'url': url, + 'width': width, + 'height': height, + 'fragments': [{ + 'url': url.replace('$M', str(j)), + 'duration': min(fragment_duration, duration - (j * fragment_duration)), + } for j in range(math.ceil(fragment_count))], + } - base_url = self.http_scheme() + '//www.youtube.com/' - webpage_url = base_url + 'watch?v=' + video_id + def _download_player_responses(self, url, smuggled_data, video_id, webpage_url): webpage = None if 'webpage' not in self._configuration_arg('player_skip'): webpage = self._download_webpage( @@ -2547,7 +3259,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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) + return webpage, master_ytcfg, player_responses, player_url + + def _list_formats(self, video_id, microformats, video_details, player_responses, player_url, duration=None): + 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, duration)) + + return live_broadcast_details, is_live, streaming_data, formats + + 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, master_ytcfg, player_responses, player_url = self._download_player_responses(url, smuggled_data, video_id, webpage_url) playability_statuses = traverse_obj( player_responses, (..., 'playabilityStatus'), expected_type=dict, default=[]) @@ -2574,57 +3306,56 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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: + multifeed_metadata_list = get_first( + player_responses, + ('multicamera', 'playerLegacyMulticameraRenderer', 'metadataList'), + expected_type=str) + if multifeed_metadata_list and not smuggled_data.get('force_singlefeed'): + if self.get_param('noplaylist'): self.to_screen('Downloading just video %s because of --no-playlist' % video_id) + else: + 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) - 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') + duration = int_or_none( + get_first(video_details, 'lengthSeconds') + or get_first(microformats, 'lengthSeconds') + or parse_duration(search_meta('duration'))) or None - streaming_data = traverse_obj(player_responses, (..., 'streamingData'), default=[]) - formats = list(self._extract_formats(streaming_data, video_id, player_url, is_live)) + live_broadcast_details, is_live, streaming_data, formats = self._list_formats( + video_id, microformats, video_details, player_responses, player_url, duration) if not formats: if not self.get_param('allow_unplayable_formats') and traverse_obj(streaming_data, (..., 'licenseInfos')): @@ -2645,16 +3376,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if reason: self.raise_no_formats(reason, expected=True) - 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)' - - # 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 = [ @@ -2672,30 +3393,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): if f.get('vcodec') != 'none': f['stretched_ratio'] = ratio break - - thumbnails = [] - 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')), - }) + thumbnails = self._extract_thumbnails((video_details, microformats), (..., ..., 'thumbnail')) thumbnail_url = search_meta(['og:image', 'twitter:image']) if thumbnail_url: thumbnails.append({ 'url': thumbnail_url, }) + original_thumbnails = thumbnails.copy() + # 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> @@ -2706,7 +3411,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '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, @@ -2716,16 +3420,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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) + self._downloader._sort_thumbnails(original_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( - 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') @@ -2735,25 +3436,34 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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 + live_start_time = parse_iso8601(get_first(live_broadcast_details, 'startTimestamp')) + live_end_time = parse_iso8601(get_first(live_broadcast_details, 'endTimestamp')) + if not duration and live_end_time and live_start_time: + duration = live_end_time - live_start_time + + if is_live and self.get_param('live_from_start'): + self._prepare_live_from_start_formats(formats, video_id, live_start_time, url, webpage_url, smuggled_data) + + formats.extend(self._extract_storyboard(player_responses, duration)) + + # 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', 'proto')) info = { 'id': video_id, - 'title': self._live_title(video_title) if is_live else video_title, + 'title': video_title, 'formats': formats, 'thumbnails': thumbnails, + # The best thumbnail that we are sure exists. Prevents unnecessary + # URL checking if user don't care about getting the best possible thumbnail + 'thumbnail': traverse_obj(original_thumbnails, (-1, 'url')), 'description': video_description, - 'upload_date': unified_strdate( - get_first(microformats, 'uploadDate') - or search_meta('uploadDate')), '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': f'https://www.youtube.com/channel/{channel_id}' if channel_id else None, + 'channel_url': format_field(channel_id, template='https://www.youtube.com/channel/%s'), 'duration': duration, 'view_count': int_or_none( get_first((video_details, microformats), (..., 'viewCount')) @@ -2772,7 +3482,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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, + 'release_timestamp': live_start_time, } pctr = traverse_obj(player_responses, (..., 'captions', 'playerCaptionsTracklistRenderer'), expected_type=dict) @@ -2797,13 +3507,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }) lang_subs.append({ 'ext': fmt, - 'url': update_url_query(base_url, query), + 'url': urljoin('https://www.youtube.com', update_url_query(base_url, query)), 'name': sub_name, }) subtitles, automatic_captions = {}, {} for lang_code, caption_track in captions.items(): base_url = caption_track.get('baseUrl') + orig_lang = parse_qs(base_url).get('lang', [None])[-1] if not base_url: continue lang_name = self._get_text(caption_track, 'name', max_runs=1) @@ -2817,11 +3528,20 @@ class YoutubeIE(YoutubeBaseInfoExtractor): for trans_code, trans_name in translation_languages.items(): if not trans_code: continue + orig_trans_code = trans_code if caption_track.get('kind') != 'asr': + if 'translated_subs' in self._configuration_arg('skip'): + continue trans_code += f'-{lang_code}' trans_name += format_field(lang_name, template=' from %s') - process_language( - automatic_captions, base_url, trans_code, trans_name, {'tlang': trans_code}) + # Add an "-orig" label to the original language so that it can be distinguished. + # The subs are returned without "-orig" as well for compatibility + if lang_code == f'a-{orig_trans_code}': + process_language( + automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {}) + # Setting tlang=lang returns damaged subtitles. + process_language(automatic_captions, base_url, trans_code, trans_name, + {} if orig_lang == orig_trans_code else {'tlang': trans_code}) info['automatic_captions'] = automatic_captions info['subtitles'] = subtitles @@ -2884,87 +3604,101 @@ class YoutubeIE(YoutubeBaseInfoExtractor): or self._extract_chapters_from_engagement_panel(initial_data, duration) or None) - contents = try_get( - initial_data, - lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], - list) or [] - for content in contents: - vpir = content.get('videoPrimaryInfoRenderer') - if vpir: - stl = vpir.get('superTitleLink') - if stl: - stl = self._get_text(stl) - if try_get( - vpir, - lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN': - info['location'] = stl - else: - mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl) - if mobj: - info.update({ - 'series': mobj.group(1), - 'season_number': int(mobj.group(2)), - 'episode_number': int(mobj.group(3)), - }) - for tlb in (try_get( - vpir, - lambda x: x['videoActions']['menuRenderer']['topLevelButtons'], - list) or []): - tbr = tlb.get('toggleButtonRenderer') or {} - for getter, regex in [( - lambda x: x['defaultText']['accessibility']['accessibilityData'], - r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([ - lambda x: x['accessibility'], - lambda x: x['accessibilityData']['accessibilityData'], - ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]: - label = (try_get(tbr, getter, dict) or {}).get('label') - if label: - mobj = re.match(regex, label) - if mobj: - info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count')) - break - sbr_tooltip = try_get( - vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip']) - if sbr_tooltip: - like_count, dislike_count = sbr_tooltip.split(' / ') + contents = traverse_obj( + initial_data, ('contents', 'twoColumnWatchNextResults', 'results', 'results', 'contents'), + expected_type=list, default=[]) + + vpir = get_first(contents, 'videoPrimaryInfoRenderer') + if vpir: + stl = vpir.get('superTitleLink') + if stl: + stl = self._get_text(stl) + if try_get( + vpir, + lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN': + info['location'] = stl + else: + mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl) + if mobj: info.update({ - 'like_count': str_to_int(like_count), - 'dislike_count': str_to_int(dislike_count), + 'series': mobj.group(1), + 'season_number': int(mobj.group(2)), + 'episode_number': int(mobj.group(3)), }) - vsir = content.get('videoSecondaryInfoRenderer') - if vsir: - info['channel'] = self._get_text(vsir, ('owner', 'videoOwnerRenderer', 'title')) - rows = try_get( - vsir, - lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'], - list) or [] - multiple_songs = False - for row in rows: - if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True: - multiple_songs = True + for tlb in (try_get( + vpir, + lambda x: x['videoActions']['menuRenderer']['topLevelButtons'], + list) or []): + tbr = tlb.get('toggleButtonRenderer') or {} + for getter, regex in [( + lambda x: x['defaultText']['accessibility']['accessibilityData'], + r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([ + lambda x: x['accessibility'], + lambda x: x['accessibilityData']['accessibilityData'], + ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]: + label = (try_get(tbr, getter, dict) or {}).get('label') + if label: + mobj = re.match(regex, label) + if mobj: + info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count')) break - for row in rows: - mrr = row.get('metadataRowRenderer') or {} - mrr_title = mrr.get('title') - if not mrr_title: - continue - 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: - if mrr_title == 'Album': - info['album'] = mrr_contents_text - elif mrr_title == 'Artist': - info['artist'] = mrr_contents_text - elif mrr_title == 'Song': - info['track'] = mrr_contents_text + sbr_tooltip = try_get( + vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip']) + if sbr_tooltip: + like_count, dislike_count = sbr_tooltip.split(' / ') + info.update({ + 'like_count': str_to_int(like_count), + 'dislike_count': str_to_int(dislike_count), + }) + vsir = get_first(contents, 'videoSecondaryInfoRenderer') + if vsir: + vor = traverse_obj(vsir, ('owner', 'videoOwnerRenderer')) + info.update({ + 'channel': self._get_text(vor, 'title'), + 'channel_follower_count': self._get_count(vor, 'subscriberCountText')}) + + rows = try_get( + vsir, + lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'], + list) or [] + multiple_songs = False + for row in rows: + if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True: + multiple_songs = True + break + for row in rows: + mrr = row.get('metadataRowRenderer') or {} + mrr_title = mrr.get('title') + if not mrr_title: + continue + 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: + if mrr_title == 'Album': + info['album'] = mrr_contents_text + elif mrr_title == 'Artist': + info['artist'] = mrr_contents_text + elif mrr_title == 'Song': + info['track'] = mrr_contents_text fallbacks = { 'channel': 'uploader', 'channel_id': 'uploader_id', 'channel_url': 'uploader_url', } + + # The upload date for scheduled, live and past live streams / premieres in microformats + # may be different from the stream date. Although not in UTC, we will prefer it in this case. + # See: https://github.com/hypervideo/hypervideo/pull/2223#issuecomment-1008485139 + upload_date = ( + unified_strdate(get_first(microformats, 'uploadDate')) + or unified_strdate(search_meta('uploadDate'))) + if not upload_date or (not info.get('is_live') and not info.get('was_live') and info.get('live_status') != 'is_upcoming'): + upload_date = strftime_or_none(self._extract_time_text(vpir, 'dateText')[0], '%Y%m%d') + info['upload_date'] = upload_date + for to, frm in fallbacks.items(): if not info.get(to): info[to] = info.get(frm) @@ -3009,494 +3743,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor): return info -class YoutubeTabIE(YoutubeBaseInfoExtractor): - IE_DESC = 'YouTube.com tab' - _VALID_URL = r'''(?x) - https?:// - (?:\w+\.)? - (?: - youtube(?:kids)?\.com| - invidio\.us - )/ - (?: - (?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' +class YoutubeTabBaseInfoExtractor(YoutubeBaseInfoExtractor): - _TESTS = [{ - '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', - }, - }, { - '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', - }, - }, { - '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', - }, - }, { - '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, - }, { - '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': 'youtube-dl public playlist', - }, - 'playlist_count': 1, - }, { - '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': 'youtube-dl empty playlist', - }, - 'playlist_count': 0, - }, { - '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, - }, { - '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, - }, { - '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, - }, { - '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, - }, { - '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, - }, { - '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', - }, - }, { - 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA', - 'only_matching': True, - }, { - 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', - 'only_matching': True, - }, { - 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', - 'only_matching': True, - }, { - 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', - 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', - 'info_dict': { - 'title': '29C3: Not my department', - 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', - 'uploader': 'Christiaan008', - 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg', - 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268', - }, - 'playlist_count': 96, - }, { - 'note': 'Large playlist', - 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', - 'info_dict': { - 'title': 'Uploads from Cauchemar', - 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q', - 'uploader': 'Cauchemar', - 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q', - }, - 'playlist_mincount': 1123, - }, { - 'note': 'even larger playlist, 8832 videos', - 'url': 'http://www.youtube.com/user/NASAgovVideo/videos', - 'only_matching': True, - }, { - 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos', - 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA', - 'info_dict': { - 'title': 'Uploads from Interstellar Movie', - 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', - 'uploader': 'Interstellar Movie', - 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA', - }, - 'playlist_mincount': 21, - }, { - '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, - }, { - 'note': 'Playlist URL that does not actually serve a playlist', - 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4', - 'info_dict': { - 'id': 'FqZTN594JQw', - 'ext': 'webm', - 'title': "Smiley's People 01 detective, Adventure Series, Action", - 'uploader': 'STREEM', - 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng', - 'upload_date': '20150526', - 'license': 'Standard YouTube License', - 'description': 'md5:507cdcb5a49ac0da37a920ece610be80', - 'categories': ['People & Blogs'], - 'tags': list, - 'view_count': int, - 'like_count': int, - 'dislike_count': int, - }, - 'params': { - 'skip_download': True, - }, - 'skip': 'This video is not available.', - 'add_ie': [YoutubeIE.ie_key()], - }, { - 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live', - 'info_dict': { - 'id': '3yImotZU3tw', # This will keep changing - 'ext': 'mp4', - 'title': compat_str, - 'uploader': 'Sky News', - 'uploader_id': 'skynews', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews', - 'upload_date': r're:\d{8}', - 'description': compat_str, - 'categories': ['News & Politics'], - 'tags': list, - 'like_count': int, - 'dislike_count': int, - }, - 'params': { - 'skip_download': True, - }, - 'expected_warnings': ['Downloading just video ', 'Ignoring subtitle tracks found in '], - }, { - 'url': 'https://www.youtube.com/user/TheYoungTurks/live', - 'info_dict': { - 'id': 'a48o2S1cPoo', - 'ext': 'mp4', - 'title': 'The Young Turks - Live Main Show', - 'uploader': 'The Young Turks', - 'uploader_id': 'TheYoungTurks', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks', - 'upload_date': '20150715', - 'license': 'Standard YouTube License', - 'description': 'md5:438179573adcdff3c97ebb1ee632b891', - 'categories': ['News & Politics'], - 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'], - 'like_count': int, - 'dislike_count': int, - }, - 'params': { - 'skip_download': True, - }, - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live', - 'only_matching': True, - }, { - '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, - }, { - 'url': 'https://www.youtube.com/feed/library', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/feed/history', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/feed/subscriptions', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/feed/watch_later', - 'only_matching': True, - }, { - 'note': 'Recommended - redirects to home page.', - 'url': 'https://www.youtube.com/feed/recommended', - 'only_matching': True, - }, { - 'note': 'inline playlist with not always working continuations', - 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/course', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/zsecurity', - 'only_matching': True, - }, { - 'url': 'http://www.youtube.com/NASAgovVideo/videos', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/TheYoungTurks/live', - 'only_matching': True, - }, { - 'url': 'https://www.youtube.com/hashtag/cctv9', - 'info_dict': { - 'id': 'cctv9', - 'title': '#cctv9', - }, - 'playlist_mincount': 350, - }, { - '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']}} - }, - }] + @staticmethod + def passthrough_smuggled_data(func): + def _smuggle(entries, smuggled_data): + for entry in entries: + # TODO: Convert URL to music.youtube instead. + # Do we need to passthrough any other smuggled_data? + entry['url'] = smuggle_url(entry['url'], smuggled_data) + yield entry - @classmethod - def suitable(cls, url): - return False if YoutubeIE.suitable(url) else super( - YoutubeTabIE, cls).suitable(url) + @functools.wraps(func) + def wrapper(self, url): + url, smuggled_data = unsmuggle_url(url, {}) + if self.is_music_url(url): + smuggled_data['is_music_url'] = True + info_dict = func(self, url, smuggled_data) + if smuggled_data and info_dict.get('entries'): + info_dict['entries'] = _smuggle(info_dict['entries'], smuggled_data) + return info_dict + return wrapper def _extract_channel_id(self, webpage): channel_id = self._html_search_meta( @@ -3515,7 +3782,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): def _extract_basic_item_renderer(item): # Modified from _extract_grid_item_renderer known_basic_renderers = ( - 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer' + 'playlistRenderer', 'videoRenderer', 'channelRenderer', 'showRenderer', 'reelItemRenderer' ) for key, renderer in item.items(): if not isinstance(renderer, dict): @@ -3565,6 +3832,24 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title) break + def _music_reponsive_list_entry(self, renderer): + video_id = traverse_obj(renderer, ('playlistItemData', 'videoId')) + if video_id: + return self.url_result(f'https://music.youtube.com/watch?v={video_id}', + ie=YoutubeIE.ie_key(), video_id=video_id) + playlist_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'playlistId')) + if playlist_id: + video_id = traverse_obj(renderer, ('navigationEndpoint', 'watchEndpoint', 'videoId')) + if video_id: + return self.url_result(f'https://music.youtube.com/watch?v={video_id}&list={playlist_id}', + ie=YoutubeTabIE.ie_key(), video_id=playlist_id) + return self.url_result(f'https://music.youtube.com/playlist?list={playlist_id}', + ie=YoutubeTabIE.ie_key(), video_id=playlist_id) + browse_id = traverse_obj(renderer, ('navigationEndpoint', 'browseEndpoint', 'browseId')) + if browse_id: + return self.url_result(f'https://music.youtube.com/browse/{browse_id}', + ie=YoutubeTabIE.ie_key(), video_id=browse_id) + def _shelf_entries_from_content(self, shelf_renderer): content = shelf_renderer.get('content') if not isinstance(content, dict): @@ -3623,6 +3908,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): if video_id: return self._extract_video(video_renderer) + def _hashtag_tile_entry(self, hashtag_tile_renderer): + url = urljoin('https://youtube.com', traverse_obj( + hashtag_tile_renderer, ('onTapCommand', 'commandMetadata', 'webCommandMetadata', 'url'))) + if url: + return self.url_result( + url, ie=YoutubeTabIE.ie_key(), title=self._get_text(hashtag_tile_renderer, 'hashtag')) + def _post_thread_entries(self, post_thread_renderer): post_renderer = try_get( post_thread_renderer, lambda x: x['post']['backstagePostRenderer'], dict) @@ -3679,49 +3971,59 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): if entry: yield entry ''' - def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data): - 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(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) + def _extract_entries(self, parent_renderer, continuation_list): + # continuation_list is modified in-place with continuation_list = [continuation_token] + continuation_list[:] = [None] + contents = try_get(parent_renderer, lambda x: x['contents'], list) or [] + for content in contents: + if not isinstance(content, dict): + continue + is_renderer = traverse_obj( + content, 'itemSectionRenderer', 'musicShelfRenderer', 'musicShelfContinuation', + expected_type=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 - 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 - - 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 - if not continuation_list[0]: - continuation_list[0] = self._extract_continuation(is_renderer) + known_renderers = { + 'playlistVideoListRenderer': self._playlist_entries, + 'gridRenderer': self._grid_entries, + 'reelShelfRenderer': self._grid_entries, + 'shelfRenderer': self._shelf_entries, + 'musicResponsiveListItemRenderer': lambda x: [self._music_reponsive_list_entry(x)], + 'backstagePostThreadRenderer': self._post_thread_entries, + 'videoRenderer': lambda x: [self._video_entry(x)], + 'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}), + 'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}), + 'hashtagTileRenderer': lambda x: [self._hashtag_tile_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 if not continuation_list[0]: - continuation_list[0] = self._extract_continuation(parent_renderer) + continuation_list[0] = self._extract_continuation(is_renderer) + + if not continuation_list[0]: + continuation_list[0] = self._extract_continuation(parent_renderer) - continuation_list = [None] # Python 2 does not support nonlocal + def _entries(self, tab, item_id, ytcfg, account_syncid, visitor_data): + continuation_list = [None] + extract_entries = lambda x: self._extract_entries(x, continuation_list) tab_content = try_get(tab, lambda x: x['content'], dict) if not tab_content: return @@ -3770,6 +4072,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): continue known_renderers = { + 'videoRenderer': (self._grid_entries, 'items'), # for membership tab 'gridPlaylistRenderer': (self._grid_entries, 'items'), 'gridVideoRenderer': (self._grid_entries, 'items'), 'gridChannelRenderer': (self._grid_entries, 'items'), @@ -3797,13 +4100,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): break @staticmethod - def _extract_selected_tab(tabs): + def _extract_selected_tab(tabs, fatal=True): for tab in tabs: renderer = dict_get(tab, ('tabRenderer', 'expandableTabRenderer')) or {} if renderer.get('selected') is True: return renderer else: - raise ExtractorError('Unable to find selected tab') + if fatal: + raise ExtractorError('Unable to find selected tab') @classmethod def _extract_uploader(cls, data): @@ -3822,10 +4126,10 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): def _extract_from_tabs(self, item_id, ytcfg, data, tabs): playlist_id = title = description = channel_url = channel_name = channel_id = None - thumbnails_list = [] tags = [] selected_tab = self._extract_selected_tab(tabs) + primary_sidebar_renderer = self._extract_sidebar_info_renderer(data, 'playlistSidebarPrimaryInfoRenderer') renderer = try_get( data, lambda x: x['metadata']['channelMetadataRenderer'], dict) if renderer: @@ -3841,34 +4145,49 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 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')), - }) + # We can get the uncropped banner/avatar by replacing the crop params with '=s0' + # See: https://github.com/hypervideo/hypervideo/issues/2237#issuecomment-1013694714 + def _get_uncropped(url): + return url_or_none((url or '').split('=')[0] + '=s0') + + avatar_thumbnails = self._extract_thumbnails(renderer, 'avatar') + if avatar_thumbnails: + uncropped_avatar = _get_uncropped(avatar_thumbnails[0]['url']) + if uncropped_avatar: + avatar_thumbnails.append({ + 'url': uncropped_avatar, + 'id': 'avatar_uncropped', + 'preference': 1 + }) + + channel_banners = self._extract_thumbnails( + data, ('header', ..., ['banner', 'mobileBanner', 'tvBanner'])) + for banner in channel_banners: + banner['preference'] = -10 + + if channel_banners: + uncropped_banner = _get_uncropped(channel_banners[0]['url']) + if uncropped_banner: + channel_banners.append({ + 'url': uncropped_banner, + 'id': 'banner_uncropped', + 'preference': -5 + }) + + primary_thumbnails = self._extract_thumbnails( + primary_sidebar_renderer, ('thumbnailRenderer', ('playlistVideoThumbnailRenderer', 'playlistCustomThumbnailRenderer'), 'thumbnail')) + if playlist_id is None: playlist_id = item_id + + playlist_stats = traverse_obj(primary_sidebar_renderer, 'stats') + last_updated_unix, _ = self._extract_time_text(playlist_stats, 2) if title is None: - title = ( - try_get(data, lambda x: x['header']['hashtagHeaderRenderer']['hashtag']['simpleText']) - or playlist_id) + title = self._get_text(data, ('header', 'hashtagHeaderRenderer', 'hashtag')) 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, @@ -3876,12 +4195,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'uploader': channel_name, 'uploader_id': channel_id, 'uploader_url': channel_url, - 'thumbnails': thumbnails, + 'thumbnails': primary_thumbnails + avatar_thumbnails + channel_banners, 'tags': tags, + 'view_count': self._get_count(playlist_stats, 1), + 'availability': self._extract_availability(data), + 'modified_date': strftime_or_none(last_updated_unix, '%Y%m%d'), + 'playlist_count': self._get_count(playlist_stats, 0), + 'channel_follower_count': self._get_count(data, ('header', ..., 'subscriberCountText')), } - availability = self._extract_availability(data) - if availability: - metadata['availability'] = availability if not channel_id: metadata.update(self._extract_uploader(data)) metadata.update({ @@ -4059,7 +4380,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): self.report_warning(error_to_compat_str(e)) break - if dict_get(data, ('contents', 'currentVideoEndpoint')): + if dict_get(data, ('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')): break last_error = 'Incomplete yt initial data received' @@ -4076,6 +4397,16 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 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) + # Reject webpage data if redirected to home page without explicitly requesting + selected_tab = self._extract_selected_tab(traverse_obj( + data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list, default=[]), fatal=False) or {} + if (url != 'https://www.youtube.com/feed/recommended' + and selected_tab.get('tabIdentifier') == 'FEwhat_to_watch' # Home page + and 'no-youtube-channel-redirect' not in self.get_param('compat_opts', [])): + msg = 'The channel/playlist does not exist and the URL redirected to youtube.com home page' + if fatal: + raise ExtractorError(msg, expected=True) + self.report_warning(msg, only_once=True) if not data: if not ytcfg and self.is_authenticated: msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.' @@ -4100,67 +4431,756 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 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')) + check_get_keys=('contents', 'currentVideoEndpoint', 'onResponseReceivedActions')) 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 + _SEARCH_PARAMS = None + + def _search_results(self, query, params=NO_DEFAULT, default_client='web'): + data = {'query': query} + if params is NO_DEFAULT: + params = self._SEARCH_PARAMS + if params: + data['params'] = params + + content_keys = ( + ('contents', 'twoColumnSearchResultsRenderer', 'primaryContents', 'sectionListRenderer', 'contents'), + ('onResponseReceivedCommands', 0, 'appendContinuationItemsAction', 'continuationItems'), + # ytmusic search + ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'), + ('continuationContents', ), + ) + check_get_keys = tuple(set(keys[0] for keys in content_keys)) + + continuation_list = [None] + for page_num in itertools.count(1): + data.update(continuation_list[0] or {}) + search = self._extract_response( + item_id='query "%s" page %s' % (query, page_num), ep='search', query=data, + default_client=default_client, check_get_keys=check_get_keys) + slr_contents = traverse_obj(search, *content_keys) + yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list) + if not continuation_list[0]: + break - 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) +class YoutubeTabIE(YoutubeTabBaseInfoExtractor): + IE_DESC = 'YouTube Tabs' + _VALID_URL = r'''(?x: + https?:// + (?:\w+\.)? + (?: + youtube(?:kids)?\.com| + %(invidious)s + )/ + (?: + (?P<channel_type>channel|c|user|browse)/| + (?P<not_channel> + feed/|hashtag/| + (?:playlist|watch)\?.*?\blist= + )| + (?!(?:%(reserved_names)s)\b) # Direct URLs + ) + (?P<id>[^/?\#&]+) + )''' % { + 'reserved_names': YoutubeBaseInfoExtractor._RESERVED_NAMES, + 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES), + } + IE_NAME = 'youtube:tab' - def __real_extract(self, url, smuggled_data): + _TESTS = [{ + 'note': 'playlists, multipage', + 'url': 'https://www.youtube.com/c/ИгорьКлейнер/playlists?view=1&flow=grid', + 'playlist_mincount': 94, + 'info_dict': { + 'id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'title': 'Igor Kleiner - Playlists', + 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2', + 'uploader': 'Igor Kleiner', + 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'channel': 'Igor Kleiner', + 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'], + 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg', + 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg', + 'channel_follower_count': int + }, + }, { + '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': 'Igor Kleiner - Playlists', + 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2', + 'uploader_id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'uploader': 'Igor Kleiner', + 'uploader_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg', + 'tags': ['"критическое', 'мышление"', '"наука', 'просто"', 'математика', '"анализ', 'данных"'], + 'channel_id': 'UCqj7Cz7revf5maW9g5pgNcg', + 'channel': 'Igor Kleiner', + 'channel_url': 'https://www.youtube.com/channel/UCqj7Cz7revf5maW9g5pgNcg', + 'channel_follower_count': int + }, + }, { + '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', + 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw', + 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw', + 'channel': '3Blue1Brown', + 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw', + 'tags': ['Mathematics'], + 'channel_follower_count': int + }, + }, { + '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', + 'uploader_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ', + 'channel_url': 'https://www.youtube.com/channel/UCAEtajcuhQ6an9WEzY9LEMQ', + 'channel_id': 'UCAEtajcuhQ6an9WEzY9LEMQ', + 'tags': 'count:13', + 'channel': 'ThirstForScience', + 'channel_follower_count': int + } + }, { + 'url': 'https://www.youtube.com/c/ChristophLaimer/playlists', + 'only_matching': True, + }, { + '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': 'youtube-dl public playlist', + 'description': '', + 'tags': [], + 'view_count': int, + 'modified_date': '20201130', + 'channel': 'Sergey M.', + 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA', + 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + }, + 'playlist_count': 1, + }, { + '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': 'youtube-dl empty playlist', + 'tags': [], + 'channel': 'Sergey M.', + 'description': '', + 'modified_date': '20160902', + 'channel_id': 'UCmlqkdCBesrv2Lak1mF_MxA', + 'channel_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + 'uploader_url': 'https://www.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + }, + 'playlist_count': 0, + }, { + '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', + 'channel': 'lex will', + 'tags': ['bible', 'history', 'prophesy'], + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_follower_count': int + }, + 'playlist_mincount': 2, + }, { + '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', + 'tags': ['bible', 'history', 'prophesy'], + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel': 'lex will', + 'channel_follower_count': int + }, + 'playlist_mincount': 975, + }, { + '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', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel': 'lex will', + 'tags': ['bible', 'history', 'prophesy'], + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_follower_count': int + }, + 'playlist_mincount': 199, + }, { + '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', + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel': 'lex will', + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'tags': ['bible', 'history', 'prophesy'], + 'channel_follower_count': int + }, + 'playlist_mincount': 17, + }, { + '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', + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel': 'lex will', + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'tags': ['bible', 'history', 'prophesy'], + 'channel_follower_count': int + }, + 'playlist_mincount': 18, + }, { + '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', + 'uploader_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel': 'lex will', + 'channel_url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w', + 'channel_id': 'UCKfVa3S1e4PHvxWcwyMMg8w', + 'tags': ['bible', 'history', 'prophesy'], + 'channel_follower_count': int + }, + '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', + 'channel_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw', + 'uploader_url': 'https://www.youtube.com/channel/UCYO_jab_esuFRV4b17AJtAw', + 'tags': ['Mathematics'], + 'channel': '3Blue1Brown', + 'channel_id': 'UCYO_jab_esuFRV4b17AJtAw', + 'channel_follower_count': int + }, + }, { + 'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA', + 'only_matching': True, + }, { + 'url': 'https://www.youtubekids.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + 'only_matching': True, + }, { + 'url': 'https://music.youtube.com/channel/UCmlqkdCBesrv2Lak1mF_MxA', + 'only_matching': True, + }, { + 'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.', + 'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', + 'info_dict': { + 'title': '29C3: Not my department', + 'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC', + 'uploader': 'Christiaan008', + 'uploader_id': 'UCEPzS1rYsrkqzSLNp76nrcg', + 'description': 'md5:a14dc1a8ef8307a9807fe136a0660268', + 'tags': [], + 'uploader_url': 'https://www.youtube.com/c/ChRiStIaAn008', + 'view_count': int, + 'modified_date': '20150605', + 'channel_id': 'UCEPzS1rYsrkqzSLNp76nrcg', + 'channel_url': 'https://www.youtube.com/c/ChRiStIaAn008', + 'channel': 'Christiaan008', + }, + 'playlist_count': 96, + }, { + 'note': 'Large playlist', + 'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q', + 'info_dict': { + 'title': 'Uploads from Cauchemar', + 'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q', + 'uploader': 'Cauchemar', + 'uploader_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q', + 'channel_url': 'https://www.youtube.com/c/Cauchemar89', + 'tags': [], + 'modified_date': r're:\d{8}', + 'channel': 'Cauchemar', + 'uploader_url': 'https://www.youtube.com/c/Cauchemar89', + 'view_count': int, + 'description': '', + 'channel_id': 'UCBABnxM4Ar9ten8Mdjj1j0Q', + }, + 'playlist_mincount': 1123, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], + }, { + 'note': 'even larger playlist, 8832 videos', + 'url': 'http://www.youtube.com/user/NASAgovVideo/videos', + 'only_matching': True, + }, { + 'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos', + 'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA', + 'info_dict': { + 'title': 'Uploads from Interstellar Movie', + 'id': 'UUXw-G3eDE9trcvY2sBMM_aA', + 'uploader': 'Interstellar Movie', + 'uploader_id': 'UCXw-G3eDE9trcvY2sBMM_aA', + 'uploader_url': 'https://www.youtube.com/c/InterstellarMovie', + 'tags': [], + 'view_count': int, + 'channel_id': 'UCXw-G3eDE9trcvY2sBMM_aA', + 'channel_url': 'https://www.youtube.com/c/InterstellarMovie', + 'channel': 'Interstellar Movie', + 'description': '', + 'modified_date': r're:\d{8}', + }, + 'playlist_mincount': 21, + }, { + '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', + 'view_count': int, + 'channel': 'Phim Siêu Nhân Nhật Bản', + 'tags': [], + 'uploader_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q', + 'description': '', + 'channel_url': 'https://www.youtube.com/channel/UCTYLiWFZy8xtPwxFwX9rV7Q', + 'channel_id': 'UCTYLiWFZy8xtPwxFwX9rV7Q', + 'modified_date': r're:\d{8}', + }, + 'playlist_mincount': 200, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], + }, { + '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', + 'channel': 'BlankTV', + 'channel_url': 'https://www.youtube.com/c/blanktv', + 'channel_id': 'UC8l9frL61Yl5KFOl87nIm2w', + 'view_count': int, + 'tags': [], + 'uploader_url': 'https://www.youtube.com/c/blanktv', + 'modified_date': r're:\d{8}', + 'description': '', + }, + 'playlist_mincount': 1000, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], + }, { + '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', + 'uploader_url': 'https://www.youtube.com/user/Computerphile', + 'tags': [], + 'view_count': int, + 'channel_id': 'UC9-y-6csu5WGm29I7JiwpnA', + 'channel_url': 'https://www.youtube.com/user/Computerphile', + 'channel': 'Computerphile', + }, + 'playlist_mincount': 11, + }, { + 'url': 'https://invidio.us/playlist?list=PL4lCao7KL_QFVb7Iudeipvc2BCavECqzc', + 'only_matching': True, + }, { + 'note': 'Playlist URL that does not actually serve a playlist', + 'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4', + 'info_dict': { + 'id': 'FqZTN594JQw', + 'ext': 'webm', + 'title': "Smiley's People 01 detective, Adventure Series, Action", + 'uploader': 'STREEM', + 'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng', + 'upload_date': '20150526', + 'license': 'Standard YouTube License', + 'description': 'md5:507cdcb5a49ac0da37a920ece610be80', + 'categories': ['People & Blogs'], + 'tags': list, + 'view_count': int, + 'like_count': int, + }, + 'params': { + 'skip_download': True, + }, + 'skip': 'This video is not available.', + 'add_ie': [YoutubeIE.ie_key()], + }, { + 'url': 'https://www.youtubekids.com/watch?v=Agk7R8I8o5U&list=PUZ6jURNr1WQZCNHF0ao-c0g', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live', + 'info_dict': { + 'id': 'GgL890LIznQ', # This will keep changing + 'ext': 'mp4', + 'title': str, + 'uploader': 'Sky News', + 'uploader_id': 'skynews', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/skynews', + 'upload_date': r're:\d{8}', + 'description': str, + 'categories': ['News & Politics'], + 'tags': list, + 'like_count': int, + 'release_timestamp': 1642502819, + 'channel': 'Sky News', + 'channel_id': 'UCoMdktPbSTixAyNGwb-UYkQ', + 'age_limit': 0, + 'view_count': int, + 'thumbnail': 'https://i.ytimg.com/vi/GgL890LIznQ/maxresdefault_live.jpg', + 'playable_in_embed': True, + 'release_date': '20220118', + 'availability': 'public', + 'live_status': 'is_live', + 'channel_url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ', + 'channel_follower_count': int + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Ignoring subtitle tracks found in '], + }, { + 'url': 'https://www.youtube.com/user/TheYoungTurks/live', + 'info_dict': { + 'id': 'a48o2S1cPoo', + 'ext': 'mp4', + 'title': 'The Young Turks - Live Main Show', + 'uploader': 'The Young Turks', + 'uploader_id': 'TheYoungTurks', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks', + 'upload_date': '20150715', + 'license': 'Standard YouTube License', + 'description': 'md5:438179573adcdff3c97ebb1ee632b891', + 'categories': ['News & Politics'], + 'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'], + 'like_count': int, + }, + 'params': { + 'skip_download': True, + }, + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live', + 'only_matching': True, + }, { + '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, + }, { + 'url': 'https://www.youtube.com/feed/library', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/feed/history', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/feed/subscriptions', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/feed/watch_later', + 'only_matching': True, + }, { + 'note': 'Recommended - redirects to home page.', + 'url': 'https://www.youtube.com/feed/recommended', + 'only_matching': True, + }, { + 'note': 'inline playlist with not always working continuations', + 'url': 'https://www.youtube.com/watch?v=UC6u0Tct-Fo&list=PL36D642111D65BE7C', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/course', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/zsecurity', + 'only_matching': True, + }, { + 'url': 'http://www.youtube.com/NASAgovVideo/videos', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/TheYoungTurks/live', + 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/hashtag/cctv9', + 'info_dict': { + 'id': 'cctv9', + 'title': '#cctv9', + 'tags': [], + }, + 'playlist_mincount': 350, + }, { + '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', + 'uploader_url': 'https://www.youtube.com/c/NoCopyrightSounds', + 'channel_url': 'https://www.youtube.com/c/NoCopyrightSounds', + 'modified_date': r're:\d{8}', + 'view_count': int, + 'channel_id': 'UC_aEa8K-EOJ3D6gOs7HcyNg', + 'tags': [], + 'channel': 'NoCopyrightSounds', + }, + 'playlist_mincount': 166, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], + }, { + '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', + 'tags': [], + 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw', + 'channel': 'Royalty Free Music - Topic', + 'view_count': int, + 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw', + 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw', + 'modified_date': r're:\d{8}', + 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw', + 'description': '', + }, + 'expected_warnings': [ + 'The URL does not have a videos tab', + r'[Uu]navailable videos (are|will be) hidden', + ], + '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', + 'tags': [], + }, + 'expected_warnings': [ + 'the playlist redirect gave error', + ], + '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)', + 'tags': [], + 'view_count': int, + 'description': '', + 'availability': 'unlisted', + 'modified_date': r're:\d{8}', + }, + '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', + 'tags': [], + 'modified_date': '20211208', + 'channel': 'colethedj', + 'view_count': int, + 'description': '', + 'uploader_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q', + 'channel_id': 'UC9zHu_mHU96r19o-wV5Qs1Q', + 'channel_url': 'https://www.youtube.com/channel/UC9zHu_mHU96r19o-wV5Qs1Q', + }, + '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', + 'tags': [], + }, + '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', + 'channel': 'Cody\'sLab', + 'channel_id': 'UCu6mSoMNzHQiBIOCkHUa2Aw', + 'tags': [], + 'channel_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw', + 'uploader_url': 'https://www.youtube.com/channel/UCu6mSoMNzHQiBIOCkHUa2Aw', + 'channel_follower_count': int + }, + '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', + 'modified_date': r're:\d{8}', + 'channel_id': 'UC9ALqqC4aIeG5iDs7i90Bfw', + 'description': '', + 'channel_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw', + 'tags': [], + 'channel': 'Royalty Free Music - Topic', + 'view_count': int, + 'uploader_url': 'https://www.youtube.com/channel/UC9ALqqC4aIeG5iDs7i90Bfw', + }, + 'expected_warnings': [ + 'does not have a videos tab', + r'[Uu]navailable videos (are|will be) hidden', + ], + 'playlist_mincount': 101, + 'params': { + 'skip_download': True, + 'extractor_args': {'youtubetab': {'skip': ['webpage']}} + }, + }, { + 'note': 'non-standard redirect to regional channel', + 'url': 'https://www.youtube.com/channel/UCwVVpHQ2Cs9iGJfpdFngePQ', + 'only_matching': True + }] + + @classmethod + def suitable(cls, url): + return False if YoutubeIE.suitable(url) else super( + YoutubeTabIE, cls).suitable(url) + + _URL_RE = re.compile(rf'(?P<pre>{_VALID_URL})(?(not_channel)|(?P<tab>/\w+))?(?P<post>.*)$') + + @YoutubeTabBaseInfoExtractor.passthrough_smuggled_data + 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 = 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) + mobj, redirect_warning = get_mobj(url), None # 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 + 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 + pre, tab, post, is_channel = f'https://www.youtube.com/playlist?list={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) + f'https://music.youtube.com/channel/{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.') + 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 + elif mobj['channel_type'] == 'browse': # Youtube music /browse/ should be changed to /channel/ + pre = f'https://www.youtube.com/channel/{item_id}' + + original_tab_name = tab 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') + redirect_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)) @@ -4168,89 +5188,111 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): # Handle both video/playlist URLs qs = parse_qs(url) - video_id = qs.get('v', [None])[0] - playlist_id = qs.get('list', [None])[0] + video_id, playlist_id = [qs.get(key, [None])[0] for key in ('v', 'list')] 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 + self.report_warning(f'A video URL was given without video ID. Trying to download playlist {playlist_id}') + url = f'https://www.youtube.com/playlist?list={playlist_id}' mobj = get_mobj(url) if video_id and playlist_id: if self.get_param('noplaylist'): - self.to_screen('Downloading just video %s because of --no-playlist' % video_id) - 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)) + self.to_screen(f'Downloading just video {video_id} because of --no-playlist') + return self.url_result(f'https://www.youtube.com/watch?v={video_id}', + ie=YoutubeIE.ie_key(), video_id=video_id) + self.to_screen(f'Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}') data, ytcfg = self._extract_data(url, item_id) - tabs = try_get( - data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list) + # YouTube may provide a non-standard redirect to the regional channel + # See: https://github.com/hypervideo/hypervideo/issues/2694 + redirect_url = traverse_obj( + data, ('onResponseReceivedActions', ..., 'navigateAction', 'endpoint', 'commandMetadata', 'webCommandMetadata', 'url'), get_all=False) + if redirect_url and 'no-youtube-channel-redirect' not in compat_opts: + redirect_url = ''.join(( + urljoin('https://www.youtube.com', redirect_url), mobj['tab'], mobj['post'])) + self.to_screen(f'This playlist is likely not available in your region. Following redirect to regional playlist {redirect_url}') + return self.url_result(redirect_url, ie=YoutubeTabIE.ie_key()) + + tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list) if tabs: selected_tab = self._extract_selected_tab(tabs) - tab_name = selected_tab.get('title', '') + selected_tab_name = selected_tab.get('title', '').lower() + if selected_tab_name == 'home': + selected_tab_name = 'featured' + requested_tab_name = mobj['tab'][1:] if 'no-youtube-channel-redirect' not in compat_opts: - if mobj['tab'] == '/live': + if requested_tab_name == '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') + if requested_tab_name not in ('', selected_tab_name): + redirect_warning = f'The channel does not have a {requested_tab_name} tab' + if not original_tab_name: + if item_id[:2] == 'UC': + # Topic channels don't have /videos. Use the equivalent playlist instead + pl_id = f'UU{item_id[2:]}' + pl_url = f'https://www.youtube.com/playlist?list={pl_id}' + try: + data, ytcfg = self._extract_data(pl_url, pl_id, ytcfg=ytcfg, fatal=True, webpage_fatal=True) + except ExtractorError: + redirect_warning += ' and the playlist redirect gave error' + else: + item_id, url, selected_tab_name = pl_id, pl_url, requested_tab_name + redirect_warning += f'. Redirecting to playlist {pl_id} instead' + if selected_tab_name and selected_tab_name != requested_tab_name: + redirect_warning += f'. {selected_tab_name} tab is being downloaded instead' else: - self.report_warning('The URL does not have a %s tab. %s is being downloaded instead' % (mobj['tab'][1:], tab_name)) + raise ExtractorError(redirect_warning, expected=True) - self.write_debug('Final URL: %s' % url) + if redirect_warning: + self.to_screen(redirect_warning) + self.write_debug(f'Final URL: {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) + tabs = traverse_obj(data, ('contents', 'twoColumnBrowseResultsRenderer', 'tabs'), expected_type=list) if tabs: return self._extract_from_tabs(item_id, ytcfg, data, tabs) - playlist = try_get( - data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict) + playlist = traverse_obj( + data, ('contents', 'twoColumnWatchNextResults', 'playlist', 'playlist'), expected_type=dict) if playlist: 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 + video_id = traverse_obj( + data, ('currentVideoEndpoint', 'watchEndpoint', 'videoId'), expected_type=str) or video_id if video_id: 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) + self.report_warning(f'Unable to recognize playlist. Downloading just video {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') class YoutubePlaylistIE(InfoExtractor): - IE_DESC = 'YouTube.com playlists' + IE_DESC = 'YouTube playlists' _VALID_URL = r'''(?x)(?: (?:https?://)? (?:\w+\.)? (?: (?: youtube(?:kids)?\.com| - invidio\.us + %(invidious)s ) /.*?\?.*?\blist= )? (?P<id>%(playlist_id)s) - )''' % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE} + )''' % { + 'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE, + 'invidious': '|'.join(YoutubeBaseInfoExtractor._INVIDIOUS_SITES), + } IE_NAME = 'youtube:playlist' _TESTS = [{ 'note': 'issue #673', @@ -4258,9 +5300,16 @@ class YoutubePlaylistIE(InfoExtractor): 'info_dict': { 'title': '[OLD]Team Fortress 2 (Class-based LP)', 'id': 'PLBB231211A4F62143', - 'uploader': 'Wickydoo', + 'uploader': 'Wickman', 'uploader_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q', 'description': 'md5:8fa6f52abb47a9552002fa3ddfc57fc2', + 'view_count': int, + 'uploader_url': 'https://www.youtube.com/user/Wickydoo', + 'modified_date': r're:\d{8}', + 'channel_id': 'UCKSpbfbl5kRQpTdL7kMc-1Q', + 'channel': 'Wickman', + 'tags': [], + 'channel_url': 'https://www.youtube.com/user/Wickydoo', }, 'playlist_mincount': 29, }, { @@ -4280,7 +5329,16 @@ class YoutubePlaylistIE(InfoExtractor): 'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu', 'uploader': 'milan', 'uploader_id': 'UCEI1-PVPcYXjB73Hfelbmaw', - } + 'description': '', + 'channel_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw', + 'tags': [], + 'modified_date': '20140919', + 'view_count': int, + 'channel': 'milan', + 'channel_id': 'UCEI1-PVPcYXjB73Hfelbmaw', + 'uploader_url': 'https://www.youtube.com/channel/UCEI1-PVPcYXjB73Hfelbmaw', + }, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], }, { 'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl', 'playlist_mincount': 654, @@ -4290,7 +5348,15 @@ class YoutubePlaylistIE(InfoExtractor): 'uploader': 'LBK', 'uploader_id': 'UC21nz3_MesPLqtDqwdvnoxA', 'description': 'md5:da521864744d60a198e3a88af4db0d9d', - } + 'channel': 'LBK', + 'view_count': int, + 'channel_url': 'https://www.youtube.com/c/愛低音的國王', + 'tags': [], + 'uploader_url': 'https://www.youtube.com/c/愛低音的國王', + 'channel_id': 'UC21nz3_MesPLqtDqwdvnoxA', + 'modified_date': r're:\d{8}', + }, + 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'], }, { 'url': 'TLGGrESM50VT6acwMjAyMjAxNw', 'only_matching': True, @@ -4304,9 +5370,7 @@ class YoutubePlaylistIE(InfoExtractor): def suitable(cls, url): if YoutubeTabIE.suitable(url): return False - # 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('v', [None])[0]: return False @@ -4340,7 +5404,16 @@ class YoutubeYtBeIE(InfoExtractor): 'categories': ['Nonprofits & Activism'], 'tags': list, 'like_count': int, - 'dislike_count': int, + 'age_limit': 0, + 'playable_in_embed': True, + 'thumbnail': 'https://i.ytimg.com/vi_webp/yeWKywCrFtk/maxresdefault.webp', + 'channel': 'Backus-Page House Museum', + 'channel_id': 'UCEfMCQ9bs3tjvjy1s451zaw', + 'live_status': 'not_live', + 'view_count': int, + 'channel_url': 'https://www.youtube.com/channel/UCEfMCQ9bs3tjvjy1s451zaw', + 'availability': 'public', + 'duration': 59, }, 'params': { 'noplaylist': True, @@ -4363,8 +5436,24 @@ class YoutubeYtBeIE(InfoExtractor): }), ie=YoutubeTabIE.ie_key(), video_id=playlist_id) +class YoutubeLivestreamEmbedIE(InfoExtractor): + IE_DESC = 'YouTube livestream embeds' + _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/embed/live_stream/?\?(?:[^#]+&)?channel=(?P<id>[^&#]+)' + _TESTS = [{ + 'url': 'https://www.youtube.com/embed/live_stream?channel=UC2_KI6RB__jGdlnK6dvFEZA', + 'only_matching': True, + }] + + def _real_extract(self, url): + channel_id = self._match_id(url) + return self.url_result( + f'https://www.youtube.com/channel/{channel_id}/live', + ie=YoutubeTabIE.ie_key(), video_id=channel_id) + + class YoutubeYtUserIE(InfoExtractor): - IE_DESC = 'YouTube.com user videos, URL or "ytuser" keyword' + IE_DESC = 'YouTube user videos; "ytuser:" prefix' + IE_NAME = 'youtube:user' _VALID_URL = r'ytuser:(?P<id>.+)' _TESTS = [{ 'url': 'ytuser:phihag', @@ -4374,13 +5463,13 @@ class YoutubeYtUserIE(InfoExtractor): def _real_extract(self, url): user_id = self._match_id(url) return self.url_result( - 'https://www.youtube.com/user/%s' % user_id, + 'https://www.youtube.com/user/%s/videos' % user_id, ie=YoutubeTabIE.ie_key(), video_id=user_id) class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): IE_NAME = 'youtube:favorites' - IE_DESC = 'YouTube.com liked videos, ":ytfav" for short (requires authentication)' + IE_DESC = 'YouTube liked videos; ":ytfav" keyword (requires cookies)' _VALID_URL = r':ytfav(?:ou?rite)?s?' _LOGIN_REQUIRED = True _TESTS = [{ @@ -4397,79 +5486,40 @@ class YoutubeFavouritesIE(YoutubeBaseInfoExtractor): ie=YoutubeTabIE.ie_key()) -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') +class YoutubeSearchIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor): + IE_DESC = 'YouTube search' IE_NAME = 'youtube:search' _SEARCH_KEY = 'ytsearch' - _SEARCH_PARAMS = None - _TESTS = [] - - def _search_results(self, query): - data = {'query': query} - if self._SEARCH_PARAMS: - data['params'] = self._SEARCH_PARAMS - continuation = {} - for page_num in itertools.count(1): - 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( - search, - (lambda x: x['contents']['twoColumnSearchResultsRenderer']['primaryContents']['sectionListRenderer']['contents'], - lambda x: x['onResponseReceivedCommands'][0]['appendContinuationItemsAction']['continuationItems']), - 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'], - list) - if not isr_contents: - continue - for content in isr_contents: - if not isinstance(content, dict): - continue - video = content.get('videoRenderer') - if not isinstance(video, dict): - continue - video_id = video.get('videoId') - if not video_id: - continue - - yield self._extract_video(video) - - if not continuation: - break + _SEARCH_PARAMS = 'EgIQAQ%3D%3D' # Videos only + _TESTS = [{ + 'url': 'ytsearch5:youtube-dl test video', + 'playlist_count': 5, + 'info_dict': { + 'id': 'youtube-dl test video', + 'title': 'youtube-dl test video', + } + }] -class YoutubeSearchDateIE(YoutubeSearchIE): +class YoutubeSearchDateIE(YoutubeTabBaseInfoExtractor, SearchInfoExtractor): IE_NAME = YoutubeSearchIE.IE_NAME + ':date' _SEARCH_KEY = 'ytsearchdate' - IE_DESC = 'YouTube.com searches, newest videos first, "ytsearchdate" keyword' - _SEARCH_PARAMS = 'CAI%3D' + IE_DESC = 'YouTube search, newest videos first' + _SEARCH_PARAMS = 'CAISAhAB' # Videos only, sorted by date + _TESTS = [{ + 'url': 'ytsearchdate5:youtube-dl test video', + 'playlist_count': 5, + 'info_dict': { + 'id': 'youtube-dl test video', + 'title': 'youtube-dl test video', + } + }] -class YoutubeSearchURLIE(YoutubeSearchIE): - IE_DESC = 'YouTube.com search URLs' +class YoutubeSearchURLIE(YoutubeTabBaseInfoExtractor): + IE_DESC = 'YouTube search URLs with sorting and filter support' IE_NAME = YoutubeSearchIE.IE_NAME + '_url' - _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?:[^&]+)(?:[&]|$)' - # _MAX_RESULTS = 100 + _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:results|search)\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)' _TESTS = [{ 'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video', 'playlist_mincount': 5, @@ -4478,22 +5528,88 @@ class YoutubeSearchURLIE(YoutubeSearchIE): 'title': 'youtube-dl test video', } }, { + 'url': 'https://www.youtube.com/results?search_query=python&sp=EgIQAg%253D%253D', + 'playlist_mincount': 5, + 'info_dict': { + 'id': 'python', + 'title': 'python', + } + }, { + 'url': 'https://www.youtube.com/results?search_query=%23cats', + 'playlist_mincount': 1, + 'info_dict': { + 'id': '#cats', + 'title': '#cats', + 'entries': [{ + 'url': r're:https://(www\.)?youtube\.com/hashtag/cats', + 'title': '#cats', + }], + }, + }, { 'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB', 'only_matching': True, }] - @classmethod - def _make_valid_url(cls): - return cls._VALID_URL + def _real_extract(self, url): + qs = parse_qs(url) + query = (qs.get('search_query') or qs.get('q'))[0] + return self.playlist_result(self._search_results(query, qs.get('sp', (None,))[0]), query, query) + + +class YoutubeMusicSearchURLIE(YoutubeTabBaseInfoExtractor): + IE_DESC = 'YouTube music search URLs with selectable sections (Eg: #songs)' + IE_NAME = 'youtube:music:search_url' + _VALID_URL = r'https?://music\.youtube\.com/search\?([^#]+&)?(?:search_query|q)=(?:[^&]+)(?:[&#]|$)' + _TESTS = [{ + 'url': 'https://music.youtube.com/search?q=royalty+free+music', + 'playlist_count': 16, + 'info_dict': { + 'id': 'royalty free music', + 'title': 'royalty free music', + } + }, { + 'url': 'https://music.youtube.com/search?q=royalty+free+music&sp=EgWKAQIIAWoKEAoQAxAEEAkQBQ%3D%3D', + 'playlist_mincount': 30, + 'info_dict': { + 'id': 'royalty free music - songs', + 'title': 'royalty free music - songs', + }, + 'params': {'extract_flat': 'in_playlist'} + }, { + 'url': 'https://music.youtube.com/search?q=royalty+free+music#community+playlists', + 'playlist_mincount': 30, + 'info_dict': { + 'id': 'royalty free music - community playlists', + 'title': 'royalty free music - community playlists', + }, + 'params': {'extract_flat': 'in_playlist'} + }] + + _SECTIONS = { + 'albums': 'EgWKAQIYAWoKEAoQAxAEEAkQBQ==', + 'artists': 'EgWKAQIgAWoKEAoQAxAEEAkQBQ==', + 'community playlists': 'EgeKAQQoAEABagoQChADEAQQCRAF', + 'featured playlists': 'EgeKAQQoADgBagwQAxAJEAQQDhAKEAU==', + 'songs': 'EgWKAQIIAWoKEAoQAxAEEAkQBQ==', + 'videos': 'EgWKAQIQAWoKEAoQAxAEEAkQBQ==', + } def _real_extract(self, url): 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) + params = qs.get('sp', (None,))[0] + if params: + section = next((k for k, v in self._SECTIONS.items() if v == params), params) + else: + section = compat_urllib_parse_unquote_plus((url.split('#') + [''])[1]).lower() + params = self._SECTIONS.get(section) + if not params: + section = None + title = join_nonempty(query, section, delim=' - ') + return self.playlist_result(self._search_results(query, params, default_client='web_music'), title, title) -class YoutubeFeedsInfoExtractor(YoutubeTabIE): +class YoutubeFeedsInfoExtractor(InfoExtractor): """ Base class for feed extractors Subclasses must define the _FEED_NAME property. @@ -4507,13 +5623,12 @@ class YoutubeFeedsInfoExtractor(YoutubeTabIE): def _real_extract(self, url): return self.url_result( - 'https://www.youtube.com/feed/%s' % self._FEED_NAME, - ie=YoutubeTabIE.ie_key()) + f'https://www.youtube.com/feed/{self._FEED_NAME}', ie=YoutubeTabIE.ie_key()) class YoutubeWatchLaterIE(InfoExtractor): IE_NAME = 'youtube:watchlater' - IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)' + IE_DESC = 'Youtube watch later list; ":ytwatchlater" keyword (requires cookies)' _VALID_URL = r':ytwatchlater' _TESTS = [{ 'url': ':ytwatchlater', @@ -4526,7 +5641,7 @@ class YoutubeWatchLaterIE(InfoExtractor): class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): - IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)' + IE_DESC = 'YouTube recommended videos; ":ytrec" keyword' _VALID_URL = r'https?://(?:www\.)?youtube\.com/?(?:[?#]|$)|:ytrec(?:ommended)?' _FEED_NAME = 'recommended' _LOGIN_REQUIRED = False @@ -4543,7 +5658,7 @@ class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor): class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): - IE_DESC = 'YouTube.com subscriptions feed, ":ytsubs" for short (requires authentication)' + IE_DESC = 'YouTube subscriptions feed; ":ytsubs" keyword (requires cookies)' _VALID_URL = r':ytsub(?:scription)?s?' _FEED_NAME = 'subscriptions' _TESTS = [{ @@ -4556,7 +5671,7 @@ class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor): class YoutubeHistoryIE(YoutubeFeedsInfoExtractor): - IE_DESC = 'Youtube watch history, ":ythis" for short (requires authentication)' + IE_DESC = 'Youtube watch history; ":ythis" keyword (requires cookies)' _VALID_URL = r':ythis(?:tory)?' _FEED_NAME = 'history' _TESTS = [{ |