_ Git - youtube-dl/blob - youtube_dl/extractor/youtube.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5
   6 import itertools
   7 import json
   8 import os.path
   9 import random
  10 import re
  11 import time
  12 import traceback
  13
  14 from .common import InfoExtractor, SearchInfoExtractor
  15 from ..jsinterp import JSInterpreter
  16 from ..swfinterp import SWFInterpreter
  17 from ..compat import (
  18     compat_chr,
  19     compat_HTTPError,
  20     compat_kwargs,
  21     compat_parse_qs,
  22     compat_urllib_parse_unquote,
  23     compat_urllib_parse_unquote_plus,
  24     compat_urllib_parse_urlencode,
  25     compat_urllib_parse_urlparse,
  26     compat_urlparse,
  27     compat_str,
  28 )
  29 from ..utils import (
  30     clean_html,
  31     dict_get,
  32     error_to_compat_str,
  33     ExtractorError,
  34     float_or_none,
  35     get_element_by_attribute,
  36     get_element_by_id,
  37     int_or_none,
  38     mimetype2ext,
  39     orderedSet,
  40     parse_codecs,
  41     parse_duration,
  42     qualities,
  43     remove_quotes,
  44     remove_start,
  45     smuggle_url,
  46     str_or_none,
  47     str_to_int,
  48     try_get,
  49     unescapeHTML,
  50     unified_strdate,
  51     unsmuggle_url,
  52     uppercase_escape,
  53     url_or_none,
  54     urlencode_postdata,
  55 )
  56
  57
  58 class YoutubeBaseInfoExtractor(InfoExtractor):
  59     """Provide base functions for Youtube extractors"""
  60     _LOGIN_URL = 'https://accounts.google.com/ServiceLogin'
  61     _TWOFACTOR_URL = 'https://accounts.google.com/signin/challenge'
  62
  63     _LOOKUP_URL = 'https://accounts.google.com/_/signin/sl/lookup'
  64     _CHALLENGE_URL = 'https://accounts.google.com/_/signin/sl/challenge'
  65     _TFA_URL = 'https://accounts.google.com/_/signin/challenge?hl=en&TL={0}'
  66
  67     _NETRC_MACHINE = 'youtube'
  68     # If True it will raise an error if no login info is provided
  69     _LOGIN_REQUIRED = False
  70
  71     _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)[0-9A-Za-z-_]{10,}'
  72
  73     def _set_language(self):
  74         self._set_cookie(
  75             '.youtube.com', 'PREF', 'f1=50000000&hl=en',
  76             # YouTube sets the expire time to about two months
  77             expire_time=time.time() + 2 * 30 * 24 * 3600)
  78
  79     def _ids_to_results(self, ids):
  80         return [
  81             self.url_result(vid_id, 'Youtube', video_id=vid_id)
  82             for vid_id in ids]
  83
  84     def _login(self):
  85         """
  86         Attempt to log in to YouTube.
  87         True is returned if successful or skipped.
  88         False is returned if login failed.
  89
  90         If _LOGIN_REQUIRED is set and no authentication was provided, an error is raised.
  91         """
  92         username, password = self._get_login_info()
  93         # No authentication to be performed
  94         if username is None:
  95             if self._LOGIN_REQUIRED and self._downloader.params.get('cookiefile') is None:
  96                 raise ExtractorError('No login info available, needed for using %s.' % self.IE_NAME, expected=True)
  97             return True
  98
  99         login_page = self._download_webpage(
 100             self._LOGIN_URL, None,
 101             note='Downloading login page',
 102             errnote='unable to fetch login page', fatal=False)
 103         if login_page is False:
 104             return
 105
 106         login_form = self._hidden_inputs(login_page)
 107
 108         def req(url, f_req, note, errnote):
 109             data = login_form.copy()
 110             data.update({
 111                 'pstMsg': 1,
 112                 'checkConnection': 'youtube',
 113                 'checkedDomains': 'youtube',
 114                 'hl': 'en',
 115                 'deviceinfo': '[null,null,null,[],null,"US",null,null,[],"GlifWebSignIn",null,[null,null,[]]]',
 116                 'f.req': json.dumps(f_req),
 117                 'flowName': 'GlifWebSignIn',
 118                 'flowEntry': 'ServiceLogin',
 119             })
 120             return self._download_json(
 121                 url, None, note=note, errnote=errnote,
 122                 transform_source=lambda s: re.sub(r'^[^[]*', '', s),
 123                 fatal=False,
 124                 data=urlencode_postdata(data), headers={
 125                     'Content-Type': 'application/x-www-form-urlencoded;charset=utf-8',
 126                     'Google-Accounts-XSRF': 1,
 127                 })
 128
 129         def warn(message):
 130             self._downloader.report_warning(message)
 131
 132         lookup_req = [
 133             username,
 134             None, [], None, 'US', None, None, 2, False, True,
 135             [
 136                 None, None,
 137                 [2, 1, None, 1,
 138                  'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn',
 139                  None, [], 4],
 140                 1, [None, None, []], None, None, None, True
 141             ],
 142             username,
 143         ]
 144
 145         lookup_results = req(
 146             self._LOOKUP_URL, lookup_req,
 147             'Looking up account info', 'Unable to look up account info')
 148
 149         if lookup_results is False:
 150             return False
 151
 152         user_hash = try_get(lookup_results, lambda x: x[0][2], compat_str)
 153         if not user_hash:
 154             warn('Unable to extract user hash')
 155             return False
 156
 157         challenge_req = [
 158             user_hash,
 159             None, 1, None, [1, None, None, None, [password, None, True]],
 160             [
 161                 None, None, [2, 1, None, 1, 'https://accounts.google.com/ServiceLogin?passive=true&continue=https%3A%2F%2Fwww.youtube.com%2Fsignin%3Fnext%3D%252F%26action_handle_signin%3Dtrue%26hl%3Den%26app%3Ddesktop%26feature%3Dsign_in_button&hl=en&service=youtube&uilel=3&requestPath=%2FServiceLogin&Page=PasswordSeparationSignIn', None, [], 4],
 162                 1, [None, None, []], None, None, None, True
 163             ]]
 164
 165         challenge_results = req(
 166             self._CHALLENGE_URL, challenge_req,
 167             'Logging in', 'Unable to log in')
 168
 169         if challenge_results is False:
 170             return
 171
 172         login_res = try_get(challenge_results, lambda x: x[0][5], list)
 173         if login_res:
 174             login_msg = try_get(login_res, lambda x: x[5], compat_str)
 175             warn(
 176                 'Unable to login: %s' % 'Invalid password'
 177                 if login_msg == 'INCORRECT_ANSWER_ENTERED' else login_msg)
 178             return False
 179
 180         res = try_get(challenge_results, lambda x: x[0][-1], list)
 181         if not res:
 182             warn('Unable to extract result entry')
 183             return False
 184
 185         login_challenge = try_get(res, lambda x: x[0][0], list)
 186         if login_challenge:
 187             challenge_str = try_get(login_challenge, lambda x: x[2], compat_str)
 188             if challenge_str == 'TWO_STEP_VERIFICATION':
 189                 # SEND_SUCCESS - TFA code has been successfully sent to phone
 190                 # QUOTA_EXCEEDED - reached the limit of TFA codes
 191                 status = try_get(login_challenge, lambda x: x[5], compat_str)
 192                 if status == 'QUOTA_EXCEEDED':
 193                     warn('Exceeded the limit of TFA codes, try later')
 194                     return False
 195
 196                 tl = try_get(challenge_results, lambda x: x[1][2], compat_str)
 197                 if not tl:
 198                     warn('Unable to extract TL')
 199                     return False
 200
 201                 tfa_code = self._get_tfa_info('2-step verification code')
 202
 203                 if not tfa_code:
 204                     warn(
 205                         'Two-factor authentication required. Provide it either interactively or with --twofactor <code>'
 206                         '(Note that only TOTP (Google Authenticator App) codes work at this time.)')
 207                     return False
 208
 209                 tfa_code = remove_start(tfa_code, 'G-')
 210
 211                 tfa_req = [
 212                     user_hash, None, 2, None,
 213                     [
 214                         9, None, None, None, None, None, None, None,
 215                         [None, tfa_code, True, 2]
 216                     ]]
 217
 218                 tfa_results = req(
 219                     self._TFA_URL.format(tl), tfa_req,
 220                     'Submitting TFA code', 'Unable to submit TFA code')
 221
 222                 if tfa_results is False:
 223                     return False
 224
 225                 tfa_res = try_get(tfa_results, lambda x: x[0][5], list)
 226                 if tfa_res:
 227                     tfa_msg = try_get(tfa_res, lambda x: x[5], compat_str)
 228                     warn(
 229                         'Unable to finish TFA: %s' % 'Invalid TFA code'
 230                         if tfa_msg == 'INCORRECT_ANSWER_ENTERED' else tfa_msg)
 231                     return False
 232
 233                 check_cookie_url = try_get(
 234                     tfa_results, lambda x: x[0][-1][2], compat_str)
 235             else:
 236                 CHALLENGES = {
 237                     'LOGIN_CHALLENGE': "This device isn't recognized. For your security, Google wants to make sure it's really you.",
 238                     'USERNAME_RECOVERY': 'Please provide additional information to aid in the recovery process.',
 239                     'REAUTH': "There is something unusual about your activity. For your security, Google wants to make sure it's really you.",
 240                 }
 241                 challenge = CHALLENGES.get(
 242                     challenge_str,
 243                     '%s returned error %s.' % (self.IE_NAME, challenge_str))
 244                 warn('%s\nGo to https://accounts.google.com/, login and solve a challenge.' % challenge)
 245                 return False
 246         else:
 247             check_cookie_url = try_get(res, lambda x: x[2], compat_str)
 248
 249         if not check_cookie_url:
 250             warn('Unable to extract CheckCookie URL')
 251             return False
 252
 253         check_cookie_results = self._download_webpage(
 254             check_cookie_url, None, 'Checking cookie', fatal=False)
 255
 256         if check_cookie_results is False:
 257             return False
 258
 259         if 'https://myaccount.google.com/' not in check_cookie_results:
 260             warn('Unable to log in')
 261             return False
 262
 263         return True
 264
 265     def _download_webpage_handle(self, *args, **kwargs):
 266         query = kwargs.get('query', {}).copy()
 267         query['disable_polymer'] = 'true'
 268         kwargs['query'] = query
 269         return super(YoutubeBaseInfoExtractor, self)._download_webpage_handle(
 270             *args, **compat_kwargs(kwargs))
 271
 272     def _real_initialize(self):
 273         if self._downloader is None:
 274             return
 275         self._set_language()
 276         if not self._login():
 277             return
 278
 279
 280 class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
 281     # Extract entries from page with "Load more" button
 282     def _entries(self, page, playlist_id):
 283         more_widget_html = content_html = page
 284         for page_num in itertools.count(1):
 285             for entry in self._process_page(content_html):
 286                 yield entry
 287
 288             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
 289             if not mobj:
 290                 break
 291
 292             count = 0
 293             retries = 3
 294             while count <= retries:
 295                 try:
 296                     # Downloading page may result in intermittent 5xx HTTP error
 297                     # that is usually worked around with a retry
 298                     more = self._download_json(
 299                         'https://youtube.com/%s' % mobj.group('more'), playlist_id,
 300                         'Downloading page #%s%s'
 301                         % (page_num, ' (retry #%d)' % count if count else ''),
 302                         transform_source=uppercase_escape)
 303                     break
 304                 except ExtractorError as e:
 305                     if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
 306                         count += 1
 307                         if count <= retries:
 308                             continue
 309                     raise
 310
 311             content_html = more['content_html']
 312             if not content_html.strip():
 313                 # Some webpages show a "Load more" button but they don't
 314                 # have more videos
 315                 break
 316             more_widget_html = more['load_more_widget_html']
 317
 318
 319 class YoutubePlaylistBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
 320     def _process_page(self, content):
 321         for video_id, video_title in self.extract_videos_from_page(content):
 322             yield self.url_result(video_id, 'Youtube', video_id, video_title)
 323
 324     def extract_videos_from_page(self, page):
 325         ids_in_page = []
 326         titles_in_page = []
 327         for mobj in re.finditer(self._VIDEO_RE, page):
 328             # The link with index 0 is not the first video of the playlist (not sure if still actual)
 329             if 'index' in mobj.groupdict() and mobj.group('id') == '0':
 330                 continue
 331             video_id = mobj.group('id')
 332             video_title = unescapeHTML(mobj.group('title'))
 333             if video_title:
 334                 video_title = video_title.strip()
 335             try:
 336                 idx = ids_in_page.index(video_id)
 337                 if video_title and not titles_in_page[idx]:
 338                     titles_in_page[idx] = video_title
 339             except ValueError:
 340                 ids_in_page.append(video_id)
 341                 titles_in_page.append(video_title)
 342         return zip(ids_in_page, titles_in_page)
 343
 344
 345 class YoutubePlaylistsBaseInfoExtractor(YoutubeEntryListBaseInfoExtractor):
 346     def _process_page(self, content):
 347         for playlist_id in orderedSet(re.findall(
 348                 r'<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*><a[^>]+href="/?playlist\?list=([0-9A-Za-z-_]{10,})"',
 349                 content)):
 350             yield self.url_result(
 351                 'https://www.youtube.com/playlist?list=%s' % playlist_id, 'YoutubePlaylist')
 352
 353     def _real_extract(self, url):
 354         playlist_id = self._match_id(url)
 355         webpage = self._download_webpage(url, playlist_id)
 356         title = self._og_search_title(webpage, fatal=False)
 357         return self.playlist_result(self._entries(webpage, playlist_id), playlist_id, title)
 358
 359
 360 class YoutubeIE(YoutubeBaseInfoExtractor):
 361     IE_DESC = 'YouTube.com'
 362     _VALID_URL = r"""(?x)^
 363                      (
 364                          (?:https?://|//)                                    # http(s):// or protocol-independent URL
 365                          (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/|
 366                             (?:www\.)?deturl\.com/www\.youtube\.com/|
 367                             (?:www\.)?pwnyoutube\.com/|
 368                             (?:www\.)?hooktube\.com/|
 369                             (?:www\.)?yourepeat\.com/|
 370                             tube\.majestyc\.net/|
 371                             (?:(?:www|dev)\.)?invidio\.us/|
 372                             (?:www\.)?invidiou\.sh/|
 373                             (?:www\.)?invidious\.snopyta\.org/|
 374                             (?:www\.)?invidious\.kabi\.tk/|
 375                             (?:www\.)?vid\.wxzm\.sx/|
 376                             youtube\.googleapis\.com/)                        # the various hostnames, with wildcard subdomains
 377                          (?:.*?\#/)?                                          # handle anchor (#/) redirect urls
 378                          (?:                                                  # the various things that can precede the ID:
 379                              (?:(?:v|embed|e)/(?!videoseries))                # v/ or embed/ or e/
 380                              |(?:                                             # or the v= param in all its forms
 381                                  (?:(?:watch|movie)(?:_popup)?(?:\.php)?/?)?  # preceding watch(_popup|.php) or nothing (like /?v=xxxx)
 382                                  (?:\?|\#!?)                                  # the params delimiter ? or # or #!
 383                                  (?:.*?[&;])??                                # any other preceding param (like /?s=tuff&v=xxxx or ?s=tuff&amp;v=V36LpHqtcDY)
 384                                  v=
 385                              )
 386                          ))
 387                          |(?:
 388                             youtu\.be|                                        # just youtu.be/xxxx
 389                             vid\.plus|                                        # or vid.plus/xxxx
 390                             zwearz\.com/watch|                                # or zwearz.com/watch/xxxx
 391                          )/
 392                          |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId=
 393                          )
 394                      )?                                                       # all until now is optional -> you can pass the naked ID
 395                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
 396                      (?!.*?\blist=
 397                         (?:
 398                             %(playlist_id)s|                                  # combined list/video URLs are handled by the playlist IE
 399                             WL                                                # WL are handled by the watch later IE
 400                         )
 401                      )
 402                      (?(1).+)?                                                # if we found the ID, everything can follow
 403                      $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
 404     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
 405     _formats = {
 406         '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
 407         '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
 408         '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'},
 409         '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'},
 410         '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'},
 411         '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 412         '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 413         '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 414         # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well
 415         '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'},
 416         '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 417         '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'},
 418         '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
 419         '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'},
 420         '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
 421         '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'},
 422         '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 423         '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'},
 424
 425
 426         # 3D videos
 427         '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
 428         '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20},
 429         '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
 430         '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20},
 431         '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20},
 432         '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
 433         '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20},
 434
 435         # Apple HTTP Live Streaming
 436         '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 437         '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 438         '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
 439         '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10},
 440         '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
 441         '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10},
 442         '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10},
 443         '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10},
 444
 445         # DASH mp4 video
 446         '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'},
 447         '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'},
 448         '135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
 449         '136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
 450         '137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
 451         '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'},  # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
 452         '160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
 453         '212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
 454         '264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
 455         '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
 456         '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60},
 457         '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'},
 458
 459         # Dash mp4 audio
 460         '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'},
 461         '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'},
 462         '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'},
 463         '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
 464         '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'},
 465         '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'},
 466         '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'},
 467
 468         # Dash webm
 469         '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 470         '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 471         '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 472         '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 473         '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 474         '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'},
 475         '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'},
 476         '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 477         '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 478         '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 479         '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 480         '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 481         '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 482         '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 483         '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 484         # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug)
 485         '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 486         '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 487         '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 488         '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 489         '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'},
 490         '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60},
 491
 492         # Dash webm audio
 493         '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128},
 494         '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256},
 495
 496         # Dash webm audio with opus inside
 497         '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50},
 498         '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70},
 499         '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160},
 500
 501         # RTMP (unnamed)
 502         '_rtmp': {'protocol': 'rtmp'},
 503
 504         # av01 video only formats sometimes served with "unknown" codecs
 505         '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 506         '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 507         '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 508         '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
 509     }
 510     _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
 511
 512     _GEO_BYPASS = False
 513
 514     IE_NAME = 'youtube'
 515     _TESTS = [
 516         {
 517             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&t=1s&end=9',
 518             'info_dict': {
 519                 'id': 'BaW_jenozKc',
 520                 'ext': 'mp4',
 521                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
 522                 'uploader': 'Philipp Hagemeister',
 523                 'uploader_id': 'phihag',
 524                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
 525                 'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
 526                 'channel_url': r're:https?://(?:www\.)?youtube\.com/channel/UCLqxVugv74EIW3VWh2NOa3Q',
 527                 'upload_date': '20121002',
 528                 '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 .',
 529                 'categories': ['Science & Technology'],
 530                 'tags': ['youtube-dl'],
 531                 'duration': 10,
 532                 'view_count': int,
 533                 'like_count': int,
 534                 'dislike_count': int,
 535                 'start_time': 1,
 536                 'end_time': 9,
 537             }
 538         },
 539         {
 540             'url': 'https://www.youtube.com/watch?v=UxxajLWwzqY',
 541             'note': 'Test generic use_cipher_signature video (#897)',
 542             'info_dict': {
 543                 'id': 'UxxajLWwzqY',
 544                 'ext': 'mp4',
 545                 'upload_date': '20120506',
 546                 'title': 'Icona Pop - I Love It (feat. Charli XCX) [OFFICIAL VIDEO]',
 547                 'alt_title': 'I Love It (feat. Charli XCX)',
 548                 'description': 'md5:f3ceb5ef83a08d95b9d146f973157cc8',
 549                 'tags': ['Icona Pop i love it', 'sweden', 'pop music', 'big beat records', 'big beat', 'charli',
 550                          'xcx', 'charli xcx', 'girls', 'hbo', 'i love it', "i don't care", 'icona', 'pop',
 551                          'iconic ep', 'iconic', 'love', 'it'],
 552                 'duration': 180,
 553                 'uploader': 'Icona Pop',
 554                 'uploader_id': 'IconaPop',
 555                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IconaPop',
 556                 'creator': 'Icona Pop',
 557                 'track': 'I Love It (feat. Charli XCX)',
 558                 'artist': 'Icona Pop',
 559             }
 560         },
 561         {
 562             'url': 'https://www.youtube.com/watch?v=07FYdnEawAQ',
 563             'note': 'Test VEVO video with age protection (#956)',
 564             'info_dict': {
 565                 'id': '07FYdnEawAQ',
 566                 'ext': 'mp4',
 567                 'upload_date': '20130703',
 568                 'title': 'Justin Timberlake - Tunnel Vision (Official Music Video) (Explicit)',
 569                 'alt_title': 'Tunnel Vision',
 570                 'description': 'md5:07dab3356cde4199048e4c7cd93471e1',
 571                 'duration': 419,
 572                 'uploader': 'justintimberlakeVEVO',
 573                 'uploader_id': 'justintimberlakeVEVO',
 574                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/justintimberlakeVEVO',
 575                 'creator': 'Justin Timberlake',
 576                 'track': 'Tunnel Vision',
 577                 'artist': 'Justin Timberlake',
 578                 'age_limit': 18,
 579             }
 580         },
 581         {
 582             'url': '//www.YouTube.com/watch?v=yZIXLfi8CZQ',
 583             'note': 'Embed-only video (#1746)',
 584             'info_dict': {
 585                 'id': 'yZIXLfi8CZQ',
 586                 'ext': 'mp4',
 587                 'upload_date': '20120608',
 588                 'title': 'Principal Sexually Assaults A Teacher - Episode 117 - 8th June 2012',
 589                 'description': 'md5:09b78bd971f1e3e289601dfba15ca4f7',
 590                 'uploader': 'SET India',
 591                 'uploader_id': 'setindia',
 592                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia',
 593                 'age_limit': 18,
 594             }
 595         },
 596         {
 597             'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=UxxajLWwzqY',
 598             'note': 'Use the first video ID in the URL',
 599             'info_dict': {
 600                 'id': 'BaW_jenozKc',
 601                 'ext': 'mp4',
 602                 'title': 'youtube-dl test video "\'/\\ä↭𝕐',
 603                 'uploader': 'Philipp Hagemeister',
 604                 'uploader_id': 'phihag',
 605                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/phihag',
 606                 'upload_date': '20121002',
 607                 '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 .',
 608                 'categories': ['Science & Technology'],
 609                 'tags': ['youtube-dl'],
 610                 'duration': 10,
 611                 'view_count': int,
 612                 'like_count': int,
 613                 'dislike_count': int,
 614             },
 615             'params': {
 616                 'skip_download': True,
 617             },
 618         },
 619         {
 620             'url': 'https://www.youtube.com/watch?v=a9LDPn-MO4I',
 621             'note': '256k DASH audio (format 141) via DASH manifest',
 622             'info_dict': {
 623                 'id': 'a9LDPn-MO4I',
 624                 'ext': 'm4a',
 625                 'upload_date': '20121002',
 626                 'uploader_id': '8KVIDEO',
 627                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/8KVIDEO',
 628                 'description': '',
 629                 'uploader': '8KVIDEO',
 630                 'title': 'UHDTV TEST 8K VIDEO.mp4'
 631             },
 632             'params': {
 633                 'youtube_include_dash_manifest': True,
 634                 'format': '141',
 635             },
 636             'skip': 'format 141 not served anymore',
 637         },
 638         # DASH manifest with encrypted signature
 639         {
 640             'url': 'https://www.youtube.com/watch?v=IB3lcPjvWLA',
 641             'info_dict': {
 642                 'id': 'IB3lcPjvWLA',
 643                 'ext': 'm4a',
 644                 'title': 'Afrojack, Spree Wilson - The Spark (Official Music Video) ft. Spree Wilson',
 645                 'description': 'md5:8f5e2b82460520b619ccac1f509d43bf',
 646                 'duration': 244,
 647                 'uploader': 'AfrojackVEVO',
 648                 'uploader_id': 'AfrojackVEVO',
 649                 'upload_date': '20131011',
 650             },
 651             'params': {
 652                 'youtube_include_dash_manifest': True,
 653                 'format': '141/bestaudio[ext=m4a]',
 654             },
 655         },
 656         # JS player signature function name containing $
 657         {
 658             'url': 'https://www.youtube.com/watch?v=nfWlot6h_JM',
 659             'info_dict': {
 660                 'id': 'nfWlot6h_JM',
 661                 'ext': 'm4a',
 662                 'title': 'Taylor Swift - Shake It Off',
 663                 'description': 'md5:bec2185232c05479482cb5a9b82719bf',
 664                 'duration': 242,
 665                 'uploader': 'TaylorSwiftVEVO',
 666                 'uploader_id': 'TaylorSwiftVEVO',
 667                 'upload_date': '20140818',
 668                 'creator': 'Taylor Swift',
 669             },
 670             'params': {
 671                 'youtube_include_dash_manifest': True,
 672                 'format': '141/bestaudio[ext=m4a]',
 673             },
 674         },
 675         # Controversy video
 676         {
 677             'url': 'https://www.youtube.com/watch?v=T4XJQO3qol8',
 678             'info_dict': {
 679                 'id': 'T4XJQO3qol8',
 680                 'ext': 'mp4',
 681                 'duration': 219,
 682                 'upload_date': '20100909',
 683                 'uploader': 'Amazing Atheist',
 684                 'uploader_id': 'TheAmazingAtheist',
 685                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist',
 686                 'title': 'Burning Everyone\'s Koran',
 687                 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html',
 688             }
 689         },
 690         # Normal age-gate video (No vevo, embed allowed)
 691         {
 692             'url': 'https://youtube.com/watch?v=HtVdAasjOgU',
 693             'info_dict': {
 694                 'id': 'HtVdAasjOgU',
 695                 'ext': 'mp4',
 696                 'title': 'The Witcher 3: Wild Hunt - The Sword Of Destiny Trailer',
 697                 'description': r're:(?s).{100,}About the Game\n.*?The Witcher 3: Wild Hunt.{100,}',
 698                 'duration': 142,
 699                 'uploader': 'The Witcher',
 700                 'uploader_id': 'WitcherGame',
 701                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/WitcherGame',
 702                 'upload_date': '20140605',
 703                 'age_limit': 18,
 704             },
 705         },
 706         # Age-gate video with encrypted signature
 707         {
 708             'url': 'https://www.youtube.com/watch?v=6kLq3WMV1nU',
 709             'info_dict': {
 710                 'id': '6kLq3WMV1nU',
 711                 'ext': 'mp4',
 712                 'title': 'Dedication To My Ex (Miss That) (Lyric Video)',
 713                 'description': 'md5:33765bb339e1b47e7e72b5490139bb41',
 714                 'duration': 246,
 715                 'uploader': 'LloydVEVO',
 716                 'uploader_id': 'LloydVEVO',
 717                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/LloydVEVO',
 718                 'upload_date': '20110629',
 719                 'age_limit': 18,
 720             },
 721         },
 722         # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
 723         # YouTube Red ad is not captured for creator
 724         {
 725             'url': '__2ABJjxzNo',
 726             'info_dict': {
 727                 'id': '__2ABJjxzNo',
 728                 'ext': 'mp4',
 729                 'duration': 266,
 730                 'upload_date': '20100430',
 731                 'uploader_id': 'deadmau5',
 732                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5',
 733                 'creator': 'deadmau5',
 734                 'description': 'md5:12c56784b8032162bb936a5f76d55360',
 735                 'uploader': 'deadmau5',
 736                 'title': 'Deadmau5 - Some Chords (HD)',
 737                 'alt_title': 'Some Chords',
 738             },
 739             'expected_warnings': [
 740                 'DASH manifest missing',
 741             ]
 742         },
 743         # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
 744         {
 745             'url': 'lqQg6PlCWgI',
 746             'info_dict': {
 747                 'id': 'lqQg6PlCWgI',
 748                 'ext': 'mp4',
 749                 'duration': 6085,
 750                 'upload_date': '20150827',
 751                 'uploader_id': 'olympic',
 752                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/olympic',
 753                 'description': 'HO09  - Women -  GER-AUS - Hockey - 31 July 2012 - London 2012 Olympic Games',
 754                 'uploader': 'Olympic',
 755                 'title': 'Hockey - Women -  GER-AUS - London 2012 Olympic Games',
 756             },
 757             'params': {
 758                 'skip_download': 'requires avconv',
 759             }
 760         },
 761         # Non-square pixels
 762         {
 763             'url': 'https://www.youtube.com/watch?v=_b-2C3KPAM0',
 764             'info_dict': {
 765                 'id': '_b-2C3KPAM0',
 766                 'ext': 'mp4',
 767                 'stretched_ratio': 16 / 9.,
 768                 'duration': 85,
 769                 'upload_date': '20110310',
 770                 'uploader_id': 'AllenMeow',
 771                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/AllenMeow',
 772                 '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',
 773                 'uploader': '孫ᄋᄅ',
 774                 'title': '[A-made] 變態妍字幕版 太妍 我就是這樣的人',
 775             },
 776         },
 777         # url_encoded_fmt_stream_map is empty string
 778         {
 779             'url': 'qEJwOuvDf7I',
 780             'info_dict': {
 781                 'id': 'qEJwOuvDf7I',
 782                 'ext': 'webm',
 783                 'title': 'Обсуждение судебной практики по выборам 14 сентября 2014 года в Санкт-Петербурге',
 784                 'description': '',
 785                 'upload_date': '20150404',
 786                 'uploader_id': 'spbelect',
 787                 'uploader': 'Наблюдатели Петербурга',
 788             },
 789             'params': {
 790                 'skip_download': 'requires avconv',
 791             },
 792             'skip': 'This live event has ended.',
 793         },
 794         # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
 795         {
 796             'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
 797             'info_dict': {
 798                 'id': 'FIl7x6_3R5Y',
 799                 'ext': 'webm',
 800                 'title': 'md5:7b81415841e02ecd4313668cde88737a',
 801                 'description': 'md5:116377fd2963b81ec4ce64b542173306',
 802                 'duration': 220,
 803                 'upload_date': '20150625',
 804                 'uploader_id': 'dorappi2000',
 805                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/dorappi2000',
 806                 'uploader': 'dorappi2000',
 807                 'formats': 'mincount:31',
 808             },
 809             'skip': 'not actual anymore',
 810         },
 811         # DASH manifest with segment_list
 812         {
 813             'url': 'https://www.youtube.com/embed/CsmdDsKjzN8',
 814             'md5': '8ce563a1d667b599d21064e982ab9e31',
 815             'info_dict': {
 816                 'id': 'CsmdDsKjzN8',
 817                 'ext': 'mp4',
 818                 'upload_date': '20150501',  # According to '<meta itemprop="datePublished"', but in other places it's 20150510
 819                 'uploader': 'Airtek',
 820                 'description': 'Retransmisión en directo de la XVIII media maratón de Zaragoza.',
 821                 'uploader_id': 'UCzTzUmjXxxacNnL8I3m4LnQ',
 822                 'title': 'Retransmisión XVIII Media maratón Zaragoza 2015',
 823             },
 824             'params': {
 825                 'youtube_include_dash_manifest': True,
 826                 'format': '135',  # bestvideo
 827             },
 828             'skip': 'This live event has ended.',
 829         },
 830         {
 831             # Multifeed videos (multiple cameras), URL is for Main Camera
 832             'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs',
 833             'info_dict': {
 834                 'id': 'jqWvoWXjCVs',
 835                 'title': 'teamPGP: Rocket League Noob Stream',
 836                 'description': 'md5:dc7872fb300e143831327f1bae3af010',
 837             },
 838             'playlist': [{
 839                 'info_dict': {
 840                     'id': 'jqWvoWXjCVs',
 841                     'ext': 'mp4',
 842                     'title': 'teamPGP: Rocket League Noob Stream (Main Camera)',
 843                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 844                     'duration': 7335,
 845                     'upload_date': '20150721',
 846                     'uploader': 'Beer Games Beer',
 847                     'uploader_id': 'beergamesbeer',
 848                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 849                     'license': 'Standard YouTube License',
 850                 },
 851             }, {
 852                 'info_dict': {
 853                     'id': '6h8e8xoXJzg',
 854                     'ext': 'mp4',
 855                     'title': 'teamPGP: Rocket League Noob Stream (kreestuh)',
 856                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 857                     'duration': 7337,
 858                     'upload_date': '20150721',
 859                     'uploader': 'Beer Games Beer',
 860                     'uploader_id': 'beergamesbeer',
 861                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 862                     'license': 'Standard YouTube License',
 863                 },
 864             }, {
 865                 'info_dict': {
 866                     'id': 'PUOgX5z9xZw',
 867                     'ext': 'mp4',
 868                     'title': 'teamPGP: Rocket League Noob Stream (grizzle)',
 869                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 870                     'duration': 7337,
 871                     'upload_date': '20150721',
 872                     'uploader': 'Beer Games Beer',
 873                     'uploader_id': 'beergamesbeer',
 874                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 875                     'license': 'Standard YouTube License',
 876                 },
 877             }, {
 878                 'info_dict': {
 879                     'id': 'teuwxikvS5k',
 880                     'ext': 'mp4',
 881                     'title': 'teamPGP: Rocket League Noob Stream (zim)',
 882                     'description': 'md5:dc7872fb300e143831327f1bae3af010',
 883                     'duration': 7334,
 884                     'upload_date': '20150721',
 885                     'uploader': 'Beer Games Beer',
 886                     'uploader_id': 'beergamesbeer',
 887                     'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer',
 888                     'license': 'Standard YouTube License',
 889                 },
 890             }],
 891             'params': {
 892                 'skip_download': True,
 893             },
 894             'skip': 'This video is not available.',
 895         },
 896         {
 897             # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
 898             'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
 899             'info_dict': {
 900                 'id': 'gVfLd0zydlo',
 901                 'title': 'DevConf.cz 2016 Day 2 Workshops 1 14:00 - 15:30',
 902             },
 903             'playlist_count': 2,
 904             'skip': 'Not multifeed anymore',
 905         },
 906         {
 907             'url': 'https://vid.plus/FlRa-iH7PGw',
 908             'only_matching': True,
 909         },
 910         {
 911             'url': 'https://zwearz.com/watch/9lWxNJF-ufM/electra-woman-dyna-girl-official-trailer-grace-helbig.html',
 912             'only_matching': True,
 913         },
 914         {
 915             # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
 916             # Also tests cut-off URL expansion in video description (see
 917             # https://github.com/ytdl-org/youtube-dl/issues/1892,
 918             # https://github.com/ytdl-org/youtube-dl/issues/8164)
 919             'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
 920             'info_dict': {
 921                 'id': 'lsguqyKfVQg',
 922                 'ext': 'mp4',
 923                 'title': '{dark walk}; Loki/AC/Dishonored; collab w/Elflover21',
 924                 'alt_title': 'Dark Walk - Position Music',
 925                 'description': 'md5:8085699c11dc3f597ce0410b0dcbb34a',
 926                 'duration': 133,
 927                 'upload_date': '20151119',
 928                 'uploader_id': 'IronSoulElf',
 929                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/IronSoulElf',
 930                 'uploader': 'IronSoulElf',
 931                 'creator': 'Todd Haberman,  Daniel Law Heath and Aaron Kaplan',
 932                 'track': 'Dark Walk - Position Music',
 933                 'artist': 'Todd Haberman,  Daniel Law Heath and Aaron Kaplan',
 934                 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
 935             },
 936             'params': {
 937                 'skip_download': True,
 938             },
 939         },
 940         {
 941             # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
 942             'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
 943             'only_matching': True,
 944         },
 945         {
 946             # Video with yt:stretch=17:0
 947             'url': 'https://www.youtube.com/watch?v=Q39EVAstoRM',
 948             'info_dict': {
 949                 'id': 'Q39EVAstoRM',
 950                 'ext': 'mp4',
 951                 'title': 'Clash Of Clans#14 Dicas De Ataque Para CV 4',
 952                 'description': 'md5:ee18a25c350637c8faff806845bddee9',
 953                 'upload_date': '20151107',
 954                 'uploader_id': 'UCCr7TALkRbo3EtFzETQF1LA',
 955                 'uploader': 'CH GAMER DROID',
 956             },
 957             'params': {
 958                 'skip_download': True,
 959             },
 960             'skip': 'This video does not exist.',
 961         },
 962         {
 963             # Video licensed under Creative Commons
 964             'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA',
 965             'info_dict': {
 966                 'id': 'M4gD1WSo5mA',
 967                 'ext': 'mp4',
 968                 'title': 'md5:e41008789470fc2533a3252216f1c1d1',
 969                 'description': 'md5:a677553cf0840649b731a3024aeff4cc',
 970                 'duration': 721,
 971                 'upload_date': '20150127',
 972                 'uploader_id': 'BerkmanCenter',
 973                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/BerkmanCenter',
 974                 'uploader': 'The Berkman Klein Center for Internet & Society',
 975                 'license': 'Creative Commons Attribution license (reuse allowed)',
 976             },
 977             'params': {
 978                 'skip_download': True,
 979             },
 980         },
 981         {
 982             # Channel-like uploader_url
 983             'url': 'https://www.youtube.com/watch?v=eQcmzGIKrzg',
 984             'info_dict': {
 985                 'id': 'eQcmzGIKrzg',
 986                 'ext': 'mp4',
 987                 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders',
 988                 'description': 'md5:dda0d780d5a6e120758d1711d062a867',
 989                 'duration': 4060,
 990                 'upload_date': '20151119',
 991                 'uploader': 'Bernie Sanders',
 992                 'uploader_id': 'UCH1dpzjCEiGAt8CXkryhkZg',
 993                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCH1dpzjCEiGAt8CXkryhkZg',
 994                 'license': 'Creative Commons Attribution license (reuse allowed)',
 995             },
 996             'params': {
 997                 'skip_download': True,
 998             },
 999         },
1000         {
1001             'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
1002             'only_matching': True,
1003         },
1004         {
1005             # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
1006             'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
1007             'only_matching': True,
1008         },
1009         {
1010             # Rental video preview
1011             'url': 'https://www.youtube.com/watch?v=yYr8q0y5Jfg',
1012             'info_dict': {
1013                 'id': 'uGpuVWrhIzE',
1014                 'ext': 'mp4',
1015                 'title': 'Piku - Trailer',
1016                 'description': 'md5:c36bd60c3fd6f1954086c083c72092eb',
1017                 'upload_date': '20150811',
1018                 'uploader': 'FlixMatrix',
1019                 'uploader_id': 'FlixMatrixKaravan',
1020                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/FlixMatrixKaravan',
1021                 'license': 'Standard YouTube License',
1022             },
1023             'params': {
1024                 'skip_download': True,
1025             },
1026             'skip': 'This video is not available.',
1027         },
1028         {
1029             # YouTube Red video with episode data
1030             'url': 'https://www.youtube.com/watch?v=iqKdEhx-dD4',
1031             'info_dict': {
1032                 'id': 'iqKdEhx-dD4',
1033                 'ext': 'mp4',
1034                 'title': 'Isolation - Mind Field (Ep 1)',
1035                 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f',
1036                 'duration': 2085,
1037                 'upload_date': '20170118',
1038                 'uploader': 'Vsauce',
1039                 'uploader_id': 'Vsauce',
1040                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/Vsauce',
1041                 'series': 'Mind Field',
1042                 'season_number': 1,
1043                 'episode_number': 1,
1044             },
1045             'params': {
1046                 'skip_download': True,
1047             },
1048             'expected_warnings': [
1049                 'Skipping DASH manifest',
1050             ],
1051         },
1052         {
1053             # The following content has been identified by the YouTube community
1054             # as inappropriate or offensive to some audiences.
1055             'url': 'https://www.youtube.com/watch?v=6SJNVb0GnPI',
1056             'info_dict': {
1057                 'id': '6SJNVb0GnPI',
1058                 'ext': 'mp4',
1059                 'title': 'Race Differences in Intelligence',
1060                 'description': 'md5:5d161533167390427a1f8ee89a1fc6f1',
1061                 'duration': 965,
1062                 'upload_date': '20140124',
1063                 'uploader': 'New Century Foundation',
1064                 'uploader_id': 'UCEJYpZGqgUob0zVVEaLhvVg',
1065                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCEJYpZGqgUob0zVVEaLhvVg',
1066             },
1067             'params': {
1068                 'skip_download': True,
1069             },
1070         },
1071         {
1072             # itag 212
1073             'url': '1t24XAntNCY',
1074             'only_matching': True,
1075         },
1076         {
1077             # geo restricted to JP
1078             'url': 'sJL6WA-aGkQ',
1079             'only_matching': True,
1080         },
1081         {
1082             'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
1083             'only_matching': True,
1084         },
1085         {
1086             'url': 'https://invidio.us/watch?v=BaW_jenozKc',
1087             'only_matching': True,
1088         },
1089         {
1090             # DRM protected
1091             'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc',
1092             'only_matching': True,
1093         },
1094         {
1095             # Video with unsupported adaptive stream type formats
1096             'url': 'https://www.youtube.com/watch?v=Z4Vy8R84T1U',
1097             'info_dict': {
1098                 'id': 'Z4Vy8R84T1U',
1099                 'ext': 'mp4',
1100                 'title': 'saman SMAN 53 Jakarta(Sancety) opening COFFEE4th at SMAN 53 Jakarta',
1101                 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
1102                 'duration': 433,
1103                 'upload_date': '20130923',
1104                 'uploader': 'Amelia Putri Harwita',
1105                 'uploader_id': 'UCpOxM49HJxmC1qCalXyB3_Q',
1106                 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCpOxM49HJxmC1qCalXyB3_Q',
1107                 'formats': 'maxcount:10',
1108             },
1109             'params': {
1110                 'skip_download': True,
1111                 'youtube_include_dash_manifest': False,
1112             },
1113         },
1114         {
1115             # Youtube Music Auto-generated description
1116             'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
1117             'info_dict': {
1118                 'id': 'MgNrAu2pzNs',
1119                 'ext': 'mp4',
1120                 'title': 'Voyeur Girl',
1121                 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
1122                 'upload_date': '20190312',
1123                 'uploader': 'Various Artists - Topic',
1124                 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
1125                 'artist': 'Stephen',
1126                 'track': 'Voyeur Girl',
1127                 'album': 'it\'s too much love to know my dear',
1128                 'release_date': '20190313',
1129                 'release_year': 2019,
1130             },
1131             'params': {
1132                 'skip_download': True,
1133             },
1134         },
1135         {
1136             # Youtube Music Auto-generated description
1137             # Retrieve 'artist' field from 'Artist:' in video description
1138             # when it is present on youtube music video
1139             'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
1140             'info_dict': {
1141                 'id': 'k0jLE7tTwjY',
1142                 'ext': 'mp4',
1143                 'title': 'Latch Feat. Sam Smith',
1144                 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
1145                 'upload_date': '20150110',
1146                 'uploader': 'Various Artists - Topic',
1147                 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
1148                 'artist': 'Disclosure',
1149                 'track': 'Latch Feat. Sam Smith',
1150                 'album': 'Latch Featuring Sam Smith',
1151                 'release_date': '20121008',
1152                 'release_year': 2012,
1153             },
1154             'params': {
1155                 'skip_download': True,
1156             },
1157         },
1158         {
1159             # Youtube Music Auto-generated description
1160             # handle multiple artists on youtube music video
1161             'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
1162             'info_dict': {
1163                 'id': '74qn0eJSjpA',
1164                 'ext': 'mp4',
1165                 'title': 'Eastside',
1166                 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
1167                 'upload_date': '20180710',
1168                 'uploader': 'Benny Blanco - Topic',
1169                 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
1170                 'artist': 'benny blanco, Halsey, Khalid',
1171                 'track': 'Eastside',
1172                 'album': 'Eastside',
1173                 'release_date': '20180713',
1174                 'release_year': 2018,
1175             },
1176             'params': {
1177                 'skip_download': True,
1178             },
1179         },
1180         {
1181             # Youtube Music Auto-generated description
1182             # handle youtube music video with release_year and no release_date
1183             'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
1184             'info_dict': {
1185                 'id': '-hcAI0g-f5M',
1186                 'ext': 'mp4',
1187                 'title': 'Put It On Me',
1188                 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
1189                 'upload_date': '20180426',
1190                 'uploader': 'Matt Maeson - Topic',
1191                 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
1192                 'artist': 'Matt Maeson',
1193                 'track': 'Put It On Me',
1194                 'album': 'The Hearse',
1195                 'release_date': None,
1196                 'release_year': 2018,
1197             },
1198             'params': {
1199                 'skip_download': True,
1200             },
1201         },
1202     ]
1203
1204     def __init__(self, *args, **kwargs):
1205         super(YoutubeIE, self).__init__(*args, **kwargs)
1206         self._player_cache = {}
1207
1208     def report_video_info_webpage_download(self, video_id):
1209         """Report attempt to download video info webpage."""
1210         self.to_screen('%s: Downloading video info webpage' % video_id)
1211
1212     def report_information_extraction(self, video_id):
1213         """Report attempt to extract video information."""
1214         self.to_screen('%s: Extracting video information' % video_id)
1215
1216     def report_unavailable_format(self, video_id, format):
1217         """Report extracted video URL."""
1218         self.to_screen('%s: Format %s not available' % (video_id, format))
1219
1220     def report_rtmp_download(self):
1221         """Indicate the download will use the RTMP protocol."""
1222         self.to_screen('RTMP download detected')
1223
1224     def _signature_cache_id(self, example_sig):
1225         """ Return a string representation of a signature """
1226         return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
1227
1228     def _extract_signature_function(self, video_id, player_url, example_sig):
1229         id_m = re.match(
1230             r'.*?-(?P<id>[a-zA-Z0-9_-]+)(?:/watch_as3|/html5player(?:-new)?|(?:/[a-z]{2,3}_[A-Z]{2})?/base)?\.(?P<ext>[a-z]+)$',
1231             player_url)
1232         if not id_m:
1233             raise ExtractorError('Cannot identify player %r' % player_url)
1234         player_type = id_m.group('ext')
1235         player_id = id_m.group('id')
1236
1237         # Read from filesystem cache
1238         func_id = '%s_%s_%s' % (
1239             player_type, player_id, self._signature_cache_id(example_sig))
1240         assert os.path.basename(func_id) == func_id
1241
1242         cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id)
1243         if cache_spec is not None:
1244             return lambda s: ''.join(s[i] for i in cache_spec)
1245
1246         download_note = (
1247             'Downloading player %s' % player_url
1248             if self._downloader.params.get('verbose') else
1249             'Downloading %s player %s' % (player_type, player_id)
1250         )
1251         if player_type == 'js':
1252             code = self._download_webpage(
1253                 player_url, video_id,
1254                 note=download_note,
1255                 errnote='Download of %s failed' % player_url)
1256             res = self._parse_sig_js(code)
1257         elif player_type == 'swf':
1258             urlh = self._request_webpage(
1259                 player_url, video_id,
1260                 note=download_note,
1261                 errnote='Download of %s failed' % player_url)
1262             code = urlh.read()
1263             res = self._parse_sig_swf(code)
1264         else:
1265             assert False, 'Invalid player type %r' % player_type
1266
1267         test_string = ''.join(map(compat_chr, range(len(example_sig))))
1268         cache_res = res(test_string)
1269         cache_spec = [ord(c) for c in cache_res]
1270
1271         self._downloader.cache.store('youtube-sigfuncs', func_id, cache_spec)
1272         return res
1273
1274     def _print_sig_code(self, func, example_sig):
1275         def gen_sig_code(idxs):
1276             def _genslice(start, end, step):
1277                 starts = '' if start == 0 else str(start)
1278                 ends = (':%d' % (end + step)) if end + step >= 0 else ':'
1279                 steps = '' if step == 1 else (':%d' % step)
1280                 return 's[%s%s%s]' % (starts, ends, steps)
1281
1282             step = None
1283             # Quelch pyflakes warnings - start will be set when step is set
1284             start = '(Never used)'
1285             for i, prev in zip(idxs[1:], idxs[:-1]):
1286                 if step is not None:
1287                     if i - prev == step:
1288                         continue
1289                     yield _genslice(start, prev, step)
1290                     step = None
1291                     continue
1292                 if i - prev in [-1, 1]:
1293                     step = i - prev
1294                     start = prev
1295                     continue
1296                 else:
1297                     yield 's[%d]' % prev
1298             if step is None:
1299                 yield 's[%d]' % i
1300             else:
1301                 yield _genslice(start, i, step)
1302
1303         test_string = ''.join(map(compat_chr, range(len(example_sig))))
1304         cache_res = func(test_string)
1305         cache_spec = [ord(c) for c in cache_res]
1306         expr_code = ' + '.join(gen_sig_code(cache_spec))
1307         signature_id_tuple = '(%s)' % (
1308             ', '.join(compat_str(len(p)) for p in example_sig.split('.')))
1309         code = ('if tuple(len(p) for p in s.split(\'.\')) == %s:\n'
1310                 '    return %s\n') % (signature_id_tuple, expr_code)
1311         self.to_screen('Extracted signature function:\n' + code)
1312
1313     def _parse_sig_js(self, jscode):
1314         funcname = self._search_regex(
1315             (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1316              r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1317              # Obsolete patterns
1318              r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1319              r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
1320              r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1321              r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1322              r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1323              r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1324              r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
1325              r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
1326             jscode, 'Initial JS player signature function name', group='sig')
1327
1328         jsi = JSInterpreter(jscode)
1329         initial_function = jsi.extract_function(funcname)
1330         return lambda s: initial_function([s])
1331
1332     def _parse_sig_swf(self, file_contents):
1333         swfi = SWFInterpreter(file_contents)
1334         TARGET_CLASSNAME = 'SignatureDecipher'
1335         searched_class = swfi.extract_class(TARGET_CLASSNAME)
1336         initial_function = swfi.extract_function(searched_class, 'decipher')
1337         return lambda s: initial_function([s])
1338
1339     def _decrypt_signature(self, s, video_id, player_url, age_gate=False):
1340         """Turn the encrypted s field into a working signature"""
1341
1342         if player_url is None:
1343             raise ExtractorError('Cannot decrypt signature without player_url')
1344
1345         if player_url.startswith('//'):
1346             player_url = 'https:' + player_url
1347         elif not re.match(r'https?://', player_url):
1348             player_url = compat_urlparse.urljoin(
1349                 'https://www.youtube.com', player_url)
1350         try:
1351             player_id = (player_url, self._signature_cache_id(s))
1352             if player_id not in self._player_cache:
1353                 func = self._extract_signature_function(
1354                     video_id, player_url, s
1355                 )
1356                 self._player_cache[player_id] = func
1357             func = self._player_cache[player_id]
1358             if self._downloader.params.get('youtube_print_sig_code'):
1359                 self._print_sig_code(func, s)
1360             return func(s)
1361         except Exception as e:
1362             tb = traceback.format_exc()
1363             raise ExtractorError(
1364                 'Signature extraction failed: ' + tb, cause=e)
1365
1366     def _get_subtitles(self, video_id, webpage):
1367         try:
1368             subs_doc = self._download_xml(
1369                 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id,
1370                 video_id, note=False)
1371         except ExtractorError as err:
1372             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
1373             return {}
1374
1375         sub_lang_list = {}
1376         for track in subs_doc.findall('track'):
1377             lang = track.attrib['lang_code']
1378             if lang in sub_lang_list:
1379                 continue
1380             sub_formats = []
1381             for ext in self._SUBTITLE_FORMATS:
1382                 params = compat_urllib_parse_urlencode({
1383                     'lang': lang,
1384                     'v': video_id,
1385                     'fmt': ext,
1386                     'name': track.attrib['name'].encode('utf-8'),
1387                 })
1388                 sub_formats.append({
1389                     'url': 'https://www.youtube.com/api/timedtext?' + params,
1390                     'ext': ext,
1391                 })
1392             sub_lang_list[lang] = sub_formats
1393         if not sub_lang_list:
1394             self._downloader.report_warning('video doesn\'t have subtitles')
1395             return {}
1396         return sub_lang_list
1397
1398     def _get_ytplayer_config(self, video_id, webpage):
1399         patterns = (
1400             # User data may contain arbitrary character sequences that may affect
1401             # JSON extraction with regex, e.g. when '};' is contained the second
1402             # regex won't capture the whole JSON. Yet working around by trying more
1403             # concrete regex first keeping in mind proper quoted string handling
1404             # to be implemented in future that will replace this workaround (see
1405             # https://github.com/ytdl-org/youtube-dl/issues/7468,
1406             # https://github.com/ytdl-org/youtube-dl/pull/7599)
1407             r';ytplayer\.config\s*=\s*({.+?});ytplayer',
1408             r';ytplayer\.config\s*=\s*({.+?});',
1409         )
1410         config = self._search_regex(
1411             patterns, webpage, 'ytplayer.config', default=None)
1412         if config:
1413             return self._parse_json(
1414                 uppercase_escape(config), video_id, fatal=False)
1415
1416     def _get_automatic_captions(self, video_id, webpage):
1417         """We need the webpage for getting the captions url, pass it as an
1418            argument to speed up the process."""
1419         self.to_screen('%s: Looking for automatic captions' % video_id)
1420         player_config = self._get_ytplayer_config(video_id, webpage)
1421         err_msg = 'Couldn\'t find automatic captions for %s' % video_id
1422         if not player_config:
1423             self._downloader.report_warning(err_msg)
1424             return {}
1425         try:
1426             args = player_config['args']
1427             caption_url = args.get('ttsurl')
1428             if caption_url:
1429                 timestamp = args['timestamp']
1430                 # We get the available subtitles
1431                 list_params = compat_urllib_parse_urlencode({
1432                     'type': 'list',
1433                     'tlangs': 1,
1434                     'asrs': 1,
1435                 })
1436                 list_url = caption_url + '&' + list_params
1437                 caption_list = self._download_xml(list_url, video_id)
1438                 original_lang_node = caption_list.find('track')
1439                 if original_lang_node is None:
1440                     self._downloader.report_warning('Video doesn\'t have automatic captions')
1441                     return {}
1442                 original_lang = original_lang_node.attrib['lang_code']
1443                 caption_kind = original_lang_node.attrib.get('kind', '')
1444
1445                 sub_lang_list = {}
1446                 for lang_node in caption_list.findall('target'):
1447                     sub_lang = lang_node.attrib['lang_code']
1448                     sub_formats = []
1449                     for ext in self._SUBTITLE_FORMATS:
1450                         params = compat_urllib_parse_urlencode({
1451                             'lang': original_lang,
1452                             'tlang': sub_lang,
1453                             'fmt': ext,
1454                             'ts': timestamp,
1455                             'kind': caption_kind,
1456                         })
1457                         sub_formats.append({
1458                             'url': caption_url + '&' + params,
1459                             'ext': ext,
1460                         })
1461                     sub_lang_list[sub_lang] = sub_formats
1462                 return sub_lang_list
1463
1464             def make_captions(sub_url, sub_langs):
1465                 parsed_sub_url = compat_urllib_parse_urlparse(sub_url)
1466                 caption_qs = compat_parse_qs(parsed_sub_url.query)
1467                 captions = {}
1468                 for sub_lang in sub_langs:
1469                     sub_formats = []
1470                     for ext in self._SUBTITLE_FORMATS:
1471                         caption_qs.update({
1472                             'tlang': [sub_lang],
1473                             'fmt': [ext],
1474                         })
1475                         sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace(
1476                             query=compat_urllib_parse_urlencode(caption_qs, True)))
1477                         sub_formats.append({
1478                             'url': sub_url,
1479                             'ext': ext,
1480                         })
1481                     captions[sub_lang] = sub_formats
1482                 return captions
1483
1484             # New captions format as of 22.06.2017
1485             player_response = args.get('player_response')
1486             if player_response and isinstance(player_response, compat_str):
1487                 player_response = self._parse_json(
1488                     player_response, video_id, fatal=False)
1489                 if player_response:
1490                     renderer = player_response['captions']['playerCaptionsTracklistRenderer']
1491                     base_url = renderer['captionTracks'][0]['baseUrl']
1492                     sub_lang_list = []
1493                     for lang in renderer['translationLanguages']:
1494                         lang_code = lang.get('languageCode')
1495                         if lang_code:
1496                             sub_lang_list.append(lang_code)
1497                     return make_captions(base_url, sub_lang_list)
1498
1499             # Some videos don't provide ttsurl but rather caption_tracks and
1500             # caption_translation_languages (e.g. 20LmZk1hakA)
1501             # Does not used anymore as of 22.06.2017
1502             caption_tracks = args['caption_tracks']
1503             caption_translation_languages = args['caption_translation_languages']
1504             caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0]
1505             sub_lang_list = []
1506             for lang in caption_translation_languages.split(','):
1507                 lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang))
1508                 sub_lang = lang_qs.get('lc', [None])[0]
1509                 if sub_lang:
1510                     sub_lang_list.append(sub_lang)
1511             return make_captions(caption_url, sub_lang_list)
1512         # An extractor error can be raise by the download process if there are
1513         # no automatic captions but there are subtitles
1514         except (KeyError, IndexError, ExtractorError):
1515             self._downloader.report_warning(err_msg)
1516             return {}
1517
1518     def _mark_watched(self, video_id, video_info, player_response):
1519         playback_url = url_or_none(try_get(
1520             player_response,
1521             lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get(
1522             video_info, lambda x: x['videostats_playback_base_url'][0]))
1523         if not playback_url:
1524             return
1525         parsed_playback_url = compat_urlparse.urlparse(playback_url)
1526         qs = compat_urlparse.parse_qs(parsed_playback_url.query)
1527
1528         # cpn generation algorithm is reverse engineered from base.js.
1529         # In fact it works even with dummy cpn.
1530         CPN_ALPHABET = 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789-_'
1531         cpn = ''.join((CPN_ALPHABET[random.randint(0, 256) & 63] for _ in range(0, 16)))
1532
1533         qs.update({
1534             'ver': ['2'],
1535             'cpn': [cpn],
1536         })
1537         playback_url = compat_urlparse.urlunparse(
1538             parsed_playback_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
1539
1540         self._download_webpage(
1541             playback_url, video_id, 'Marking watched',
1542             'Unable to mark watched', fatal=False)
1543
1544     @staticmethod
1545     def _extract_urls(webpage):
1546         # Embedded YouTube player
1547         entries = [
1548             unescapeHTML(mobj.group('url'))
1549             for mobj in re.finditer(r'''(?x)
1550             (?:
1551                 <iframe[^>]+?src=|
1552                 data-video-url=|
1553                 <embed[^>]+?src=|
1554                 embedSWF\(?:\s*|
1555                 <object[^>]+data=|
1556                 new\s+SWFObject\(
1557             )
1558             (["\'])
1559                 (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
1560                 (?:embed|v|p)/[0-9A-Za-z_-]{11}.*?)
1561             \1''', webpage)]
1562
1563         # lazyYT YouTube embed
1564         entries.extend(list(map(
1565             unescapeHTML,
1566             re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage))))
1567
1568         # Wordpress "YouTube Video Importer" plugin
1569         matches = re.findall(r'''(?x)<div[^>]+
1570             class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
1571             data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
1572         entries.extend(m[-1] for m in matches)
1573
1574         return entries
1575
1576     @staticmethod
1577     def _extract_url(webpage):
1578         urls = YoutubeIE._extract_urls(webpage)
1579         return urls[0] if urls else None
1580
1581     @classmethod
1582     def extract_id(cls, url):
1583         mobj = re.match(cls._VALID_URL, url, re.VERBOSE)
1584         if mobj is None:
1585             raise ExtractorError('Invalid URL: %s' % url)
1586         video_id = mobj.group(2)
1587         return video_id
1588
1589     def _extract_annotations(self, video_id):
1590         return self._download_webpage(
1591             'https://www.youtube.com/annotations_invideo', video_id,
1592             note='Downloading annotations',
1593             errnote='Unable to download video annotations', fatal=False,
1594             query={
1595                 'features': 1,
1596                 'legacy': 1,
1597                 'video_id': video_id,
1598             })
1599
1600     @staticmethod
1601     def _extract_chapters(description, duration):
1602         if not description:
1603             return None
1604         chapter_lines = re.findall(
1605             r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)',
1606             description)
1607         if not chapter_lines:
1608             return None
1609         chapters = []
1610         for next_num, (chapter_line, time_point) in enumerate(
1611                 chapter_lines, start=1):
1612             start_time = parse_duration(time_point)
1613             if start_time is None:
1614                 continue
1615             if start_time > duration:
1616                 break
1617             end_time = (duration if next_num == len(chapter_lines)
1618                         else parse_duration(chapter_lines[next_num][1]))
1619             if end_time is None:
1620                 continue
1621             if end_time > duration:
1622                 end_time = duration
1623             if start_time > end_time:
1624                 break
1625             chapter_title = re.sub(
1626                 r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-')
1627             chapter_title = re.sub(r'\s+', ' ', chapter_title)
1628             chapters.append({
1629                 'start_time': start_time,
1630                 'end_time': end_time,
1631                 'title': chapter_title,
1632             })
1633         return chapters
1634
1635     def _real_extract(self, url):
1636         url, smuggled_data = unsmuggle_url(url, {})
1637
1638         proto = (
1639             'http' if self._downloader.params.get('prefer_insecure', False)
1640             else 'https')
1641
1642         start_time = None
1643         end_time = None
1644         parsed_url = compat_urllib_parse_urlparse(url)
1645         for component in [parsed_url.fragment, parsed_url.query]:
1646             query = compat_parse_qs(component)
1647             if start_time is None and 't' in query:
1648                 start_time = parse_duration(query['t'][0])
1649             if start_time is None and 'start' in query:
1650                 start_time = parse_duration(query['start'][0])
1651             if end_time is None and 'end' in query:
1652                 end_time = parse_duration(query['end'][0])
1653
1654         # Extract original video URL from URL with redirection, like age verification, using next_url parameter
1655         mobj = re.search(self._NEXT_URL_RE, url)
1656         if mobj:
1657             url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/')
1658         video_id = self.extract_id(url)
1659
1660         # Get video webpage
1661         url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id
1662         video_webpage = self._download_webpage(url, video_id)
1663
1664         # Attempt to extract SWF player URL
1665         mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage)
1666         if mobj is not None:
1667             player_url = re.sub(r'\\(.)', r'\1', mobj.group(1))
1668         else:
1669             player_url = None
1670
1671         dash_mpds = []
1672
1673         def add_dash_mpd(video_info):
1674             dash_mpd = video_info.get('dashmpd')
1675             if dash_mpd and dash_mpd[0] not in dash_mpds:
1676                 dash_mpds.append(dash_mpd[0])
1677
1678         def add_dash_mpd_pr(pl_response):
1679             dash_mpd = url_or_none(try_get(
1680                 pl_response, lambda x: x['streamingData']['dashManifestUrl'],
1681                 compat_str))
1682             if dash_mpd and dash_mpd not in dash_mpds:
1683                 dash_mpds.append(dash_mpd)
1684
1685         is_live = None
1686         view_count = None
1687
1688         def extract_view_count(v_info):
1689             return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
1690
1691         def extract_token(v_info):
1692             return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
1693
1694         player_response = {}
1695
1696         # Get video info
1697         embed_webpage = None
1698         if re.search(r'player-age-gate-content">', video_webpage) is not None:
1699             age_gate = True
1700             # We simulate the access to the video from www.youtube.com/v/{video_id}
1701             # this can be viewed without login into Youtube
1702             url = proto + '://www.youtube.com/embed/%s' % video_id
1703             embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage')
1704             data = compat_urllib_parse_urlencode({
1705                 'video_id': video_id,
1706                 'eurl': 'https://youtube.googleapis.com/v/' + video_id,
1707                 'sts': self._search_regex(
1708                     r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''),
1709             })
1710             video_info_url = proto + '://www.youtube.com/get_video_info?' + data
1711             video_info_webpage = self._download_webpage(
1712                 video_info_url, video_id,
1713                 note='Refetching age-gated info webpage',
1714                 errnote='unable to download video info webpage')
1715             video_info = compat_parse_qs(video_info_webpage)
1716             add_dash_mpd(video_info)
1717         else:
1718             age_gate = False
1719             video_info = None
1720             sts = None
1721             # Try looking directly into the video webpage
1722             ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
1723             if ytplayer_config:
1724                 args = ytplayer_config['args']
1725                 if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'):
1726                     # Convert to the same format returned by compat_parse_qs
1727                     video_info = dict((k, [v]) for k, v in args.items())
1728                     add_dash_mpd(video_info)
1729                 # Rental video is not rented but preview is available (e.g.
1730                 # https://www.youtube.com/watch?v=yYr8q0y5Jfg,
1731                 # https://github.com/ytdl-org/youtube-dl/issues/10532)
1732                 if not video_info and args.get('ypc_vid'):
1733                     return self.url_result(
1734                         args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
1735                 if args.get('livestream') == '1' or args.get('live_playback') == 1:
1736                     is_live = True
1737                 sts = ytplayer_config.get('sts')
1738                 if not player_response:
1739                     pl_response = str_or_none(args.get('player_response'))
1740                     if pl_response:
1741                         pl_response = self._parse_json(pl_response, video_id, fatal=False)
1742                         if isinstance(pl_response, dict):
1743                             player_response = pl_response
1744             if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True):
1745                 add_dash_mpd_pr(player_response)
1746                 # We also try looking in get_video_info since it may contain different dashmpd
1747                 # URL that points to a DASH manifest with possibly different itag set (some itags
1748                 # are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
1749                 # manifest pointed by get_video_info's dashmpd).
1750                 # The general idea is to take a union of itags of both DASH manifests (for example
1751                 # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
1752                 self.report_video_info_webpage_download(video_id)
1753                 for el in ('embedded', 'detailpage', 'vevo', ''):
1754                     query = {
1755                         'video_id': video_id,
1756                         'ps': 'default',
1757                         'eurl': '',
1758                         'gl': 'US',
1759                         'hl': 'en',
1760                     }
1761                     if el:
1762                         query['el'] = el
1763                     if sts:
1764                         query['sts'] = sts
1765                     video_info_webpage = self._download_webpage(
1766                         '%s://www.youtube.com/get_video_info' % proto,
1767                         video_id, note=False,
1768                         errnote='unable to download video info webpage',
1769                         fatal=False, query=query)
1770                     if not video_info_webpage:
1771                         continue
1772                     get_video_info = compat_parse_qs(video_info_webpage)
1773                     if not player_response:
1774                         pl_response = get_video_info.get('player_response', [None])[0]
1775                         if isinstance(pl_response, dict):
1776                             player_response = pl_response
1777                             add_dash_mpd_pr(player_response)
1778                     add_dash_mpd(get_video_info)
1779                     if view_count is None:
1780                         view_count = extract_view_count(get_video_info)
1781                     if not video_info:
1782                         video_info = get_video_info
1783                     get_token = extract_token(get_video_info)
1784                     if get_token:
1785                         # Different get_video_info requests may report different results, e.g.
1786                         # some may report video unavailability, but some may serve it without
1787                         # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
1788                         # the original webpage as well as el=info and el=embedded get_video_info
1789                         # requests report video unavailability due to geo restriction while
1790                         # el=detailpage succeeds and returns valid data). This is probably
1791                         # due to YouTube measures against IP ranges of hosting providers.
1792                         # Working around by preferring the first succeeded video_info containing
1793                         # the token if no such video_info yet was found.
1794                         token = extract_token(video_info)
1795                         if not token:
1796                             video_info = get_video_info
1797                         break
1798
1799         def extract_unavailable_message():
1800             return self._html_search_regex(
1801                 r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
1802                 video_webpage, 'unavailable message', default=None)
1803
1804         if not video_info:
1805             unavailable_message = extract_unavailable_message()
1806             if not unavailable_message:
1807                 unavailable_message = 'Unable to extract video data'
1808             raise ExtractorError(
1809                 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
1810
1811         video_details = try_get(
1812             player_response, lambda x: x['videoDetails'], dict) or {}
1813
1814         # title
1815         if 'title' in video_info:
1816             video_title = video_info['title'][0]
1817         elif 'title' in player_response:
1818             video_title = video_details['title']
1819         else:
1820             self._downloader.report_warning('Unable to extract video title')
1821             video_title = '_'
1822
1823         # description
1824         description_original = video_description = get_element_by_id("eow-description", video_webpage)
1825         if video_description:
1826
1827             def replace_url(m):
1828                 redir_url = compat_urlparse.urljoin(url, m.group(1))
1829                 parsed_redir_url = compat_urllib_parse_urlparse(redir_url)
1830                 if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect':
1831                     qs = compat_parse_qs(parsed_redir_url.query)
1832                     q = qs.get('q')
1833                     if q and q[0]:
1834                         return q[0]
1835                 return redir_url
1836
1837             description_original = video_description = re.sub(r'''(?x)
1838                 <a\s+
1839                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
1840                     (?:title|href)="([^"]+)"\s+
1841                     (?:[a-zA-Z-]+="[^"]*"\s+)*?
1842                     class="[^"]*"[^>]*>
1843                 [^<]+\.{3}\s*
1844                 </a>
1845             ''', replace_url, video_description)
1846             video_description = clean_html(video_description)
1847         else:
1848             fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
1849             if fd_mobj:
1850                 video_description = unescapeHTML(fd_mobj.group(1))
1851             else:
1852                 video_description = ''
1853
1854         if not smuggled_data.get('force_singlefeed', False):
1855             if not self._downloader.params.get('noplaylist'):
1856                 multifeed_metadata_list = try_get(
1857                     player_response,
1858                     lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'],
1859                     compat_str) or try_get(
1860                     video_info, lambda x: x['multifeed_metadata_list'][0], compat_str)
1861                 if multifeed_metadata_list:
1862                     entries = []
1863                     feed_ids = []
1864                     for feed in multifeed_metadata_list.split(','):
1865                         # Unquote should take place before split on comma (,) since textual
1866                         # fields may contain comma as well (see
1867                         # https://github.com/ytdl-org/youtube-dl/issues/8536)
1868                         feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
1869                         entries.append({
1870                             '_type': 'url_transparent',
1871                             'ie_key': 'Youtube',
1872                             'url': smuggle_url(
1873                                 '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]),
1874                                 {'force_singlefeed': True}),
1875                             'title': '%s (%s)' % (video_title, feed_data['title'][0]),
1876                         })
1877                         feed_ids.append(feed_data['id'][0])
1878                     self.to_screen(
1879                         'Downloading multifeed video (%s) - add --no-playlist to just download video %s'
1880                         % (', '.join(feed_ids), video_id))
1881                     return self.playlist_result(entries, video_id, video_title, video_description)
1882             else:
1883                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
1884
1885         if view_count is None:
1886             view_count = extract_view_count(video_info)
1887         if view_count is None and video_details:
1888             view_count = int_or_none(video_details.get('viewCount'))
1889
1890         # Check for "rental" videos
1891         if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
1892             raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
1893
1894         def _extract_filesize(media_url):
1895             return int_or_none(self._search_regex(
1896                 r'\bclen[=/](\d+)', media_url, 'filesize', default=None))
1897
1898         if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'):
1899             self.report_rtmp_download()
1900             formats = [{
1901                 'format_id': '_rtmp',
1902                 'protocol': 'rtmp',
1903                 'url': video_info['conn'][0],
1904                 'player_url': player_url,
1905             }]
1906         elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
1907             encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
1908             if 'rtmpe%3Dyes' in encoded_url_map:
1909                 raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
1910             formats_spec = {}
1911             fmt_list = video_info.get('fmt_list', [''])[0]
1912             if fmt_list:
1913                 for fmt in fmt_list.split(','):
1914                     spec = fmt.split('/')
1915                     if len(spec) > 1:
1916                         width_height = spec[1].split('x')
1917                         if len(width_height) == 2:
1918                             formats_spec[spec[0]] = {
1919                                 'resolution': spec[1],
1920                                 'width': int_or_none(width_height[0]),
1921                                 'height': int_or_none(width_height[1]),
1922                             }
1923             q = qualities(['small', 'medium', 'hd720'])
1924             streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list)
1925             if streaming_formats:
1926                 for fmt in streaming_formats:
1927                     itag = str_or_none(fmt.get('itag'))
1928                     if not itag:
1929                         continue
1930                     quality = fmt.get('quality')
1931                     quality_label = fmt.get('qualityLabel') or quality
1932                     formats_spec[itag] = {
1933                         'asr': int_or_none(fmt.get('audioSampleRate')),
1934                         'filesize': int_or_none(fmt.get('contentLength')),
1935                         'format_note': quality_label,
1936                         'fps': int_or_none(fmt.get('fps')),
1937                         'height': int_or_none(fmt.get('height')),
1938                         'quality': q(quality),
1939                         # bitrate for itag 43 is always 2147483647
1940                         'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None,
1941                         'width': int_or_none(fmt.get('width')),
1942                     }
1943             formats = []
1944             for url_data_str in encoded_url_map.split(','):
1945                 url_data = compat_parse_qs(url_data_str)
1946                 if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
1947                     continue
1948                 stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
1949                 # Unsupported FORMAT_STREAM_TYPE_OTF
1950                 if stream_type == 3:
1951                     continue
1952                 format_id = url_data['itag'][0]
1953                 url = url_data['url'][0]
1954
1955                 if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True):
1956                     ASSETS_RE = r'"assets":.+?"js":\s*("[^"]+")'
1957                     jsplayer_url_json = self._search_regex(
1958                         ASSETS_RE,
1959                         embed_webpage if age_gate else video_webpage,
1960                         'JS player URL (1)', default=None)
1961                     if not jsplayer_url_json and not age_gate:
1962                         # We need the embed website after all
1963                         if embed_webpage is None:
1964                             embed_url = proto + '://www.youtube.com/embed/%s' % video_id
1965                             embed_webpage = self._download_webpage(
1966                                 embed_url, video_id, 'Downloading embed webpage')
1967                         jsplayer_url_json = self._search_regex(
1968                             ASSETS_RE, embed_webpage, 'JS player URL')
1969
1970                     player_url = json.loads(jsplayer_url_json)
1971                     if player_url is None:
1972                         player_url_json = self._search_regex(
1973                             r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")',
1974                             video_webpage, 'age gate player URL')
1975                         player_url = json.loads(player_url_json)
1976
1977                 if 'sig' in url_data:
1978                     url += '&signature=' + url_data['sig'][0]
1979                 elif 's' in url_data:
1980                     encrypted_sig = url_data['s'][0]
1981
1982                     if self._downloader.params.get('verbose'):
1983                         if player_url is None:
1984                             player_version = 'unknown'
1985                             player_desc = 'unknown'
1986                         else:
1987                             if player_url.endswith('swf'):
1988                                 player_version = self._search_regex(
1989                                     r'-(.+?)(?:/watch_as3)?\.swf$', player_url,
1990                                     'flash player', fatal=False)
1991                                 player_desc = 'flash player %s' % player_version
1992                             else:
1993                                 player_version = self._search_regex(
1994                                     [r'html5player-([^/]+?)(?:/html5player(?:-new)?)?\.js',
1995                                      r'(?:www|player(?:_ias)?)-([^/]+)(?:/[a-z]{2,3}_[A-Z]{2})?/base\.js'],
1996                                     player_url,
1997                                     'html5 player', fatal=False)
1998                                 player_desc = 'html5 player %s' % player_version
1999
2000                         parts_sizes = self._signature_cache_id(encrypted_sig)
2001                         self.to_screen('{%s} signature length %s, %s' %
2002                                        (format_id, parts_sizes, player_desc))
2003
2004                     signature = self._decrypt_signature(
2005                         encrypted_sig, video_id, player_url, age_gate)
2006                     sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
2007                     url += '&%s=%s' % (sp, signature)
2008                 if 'ratebypass' not in url:
2009                     url += '&ratebypass=yes'
2010
2011                 dct = {
2012                     'format_id': format_id,
2013                     'url': url,
2014                     'player_url': player_url,
2015                 }
2016                 if format_id in self._formats:
2017                     dct.update(self._formats[format_id])
2018                 if format_id in formats_spec:
2019                     dct.update(formats_spec[format_id])
2020
2021                 # Some itags are not included in DASH manifest thus corresponding formats will
2022                 # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
2023                 # Trying to extract metadata from url_encoded_fmt_stream_map entry.
2024                 mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
2025                 width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
2026
2027                 filesize = int_or_none(url_data.get(
2028                     'clen', [None])[0]) or _extract_filesize(url)
2029
2030                 quality = url_data.get('quality', [None])[0]
2031
2032                 more_fields = {
2033                     'filesize': filesize,
2034                     'tbr': float_or_none(url_data.get('bitrate', [None])[0], 1000),
2035                     'width': width,
2036                     'height': height,
2037                     'fps': int_or_none(url_data.get('fps', [None])[0]),
2038                     'format_note': url_data.get('quality_label', [None])[0] or quality,
2039                     'quality': q(quality),
2040                 }
2041                 for key, value in more_fields.items():
2042                     if value:
2043                         dct[key] = value
2044                 type_ = url_data.get('type', [None])[0]
2045                 if type_:
2046                     type_split = type_.split(';')
2047                     kind_ext = type_split[0].split('/')
2048                     if len(kind_ext) == 2:
2049                         kind, _ = kind_ext
2050                         dct['ext'] = mimetype2ext(type_split[0])
2051                         if kind in ('audio', 'video'):
2052                             codecs = None
2053                             for mobj in re.finditer(
2054                                     r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_):
2055                                 if mobj.group('key') == 'codecs':
2056                                     codecs = mobj.group('val')
2057                                     break
2058                             if codecs:
2059                                 dct.update(parse_codecs(codecs))
2060                 if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
2061                     dct['downloader_options'] = {
2062                         # Youtube throttles chunks >~10M
2063                         'http_chunk_size': 10485760,
2064                     }
2065                 formats.append(dct)
2066         else:
2067             manifest_url = (
2068                 url_or_none(try_get(
2069                     player_response,
2070                     lambda x: x['streamingData']['hlsManifestUrl'],
2071                     compat_str))
2072                 or url_or_none(try_get(
2073                     video_info, lambda x: x['hlsvp'][0], compat_str)))
2074             if manifest_url:
2075                 formats = []
2076                 m3u8_formats = self._extract_m3u8_formats(
2077                     manifest_url, video_id, 'mp4', fatal=False)
2078                 for a_format in m3u8_formats:
2079                     itag = self._search_regex(
2080                         r'/itag/(\d+)/', a_format['url'], 'itag', default=None)
2081                     if itag:
2082                         a_format['format_id'] = itag
2083                         if itag in self._formats:
2084                             dct = self._formats[itag].copy()
2085                             dct.update(a_format)
2086                             a_format = dct
2087                     a_format['player_url'] = player_url
2088                     # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming
2089                     a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
2090                     formats.append(a_format)
2091             else:
2092                 error_message = clean_html(video_info.get('reason', [None])[0])
2093                 if not error_message:
2094                     error_message = extract_unavailable_message()
2095                 if error_message:
2096                     raise ExtractorError(error_message, expected=True)
2097                 raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info')
2098
2099         # uploader
2100         video_uploader = try_get(
2101             video_info, lambda x: x['author'][0],
2102             compat_str) or str_or_none(video_details.get('author'))
2103         if video_uploader:
2104             video_uploader = compat_urllib_parse_unquote_plus(video_uploader)
2105         else:
2106             self._downloader.report_warning('unable to extract uploader name')
2107
2108         # uploader_id
2109         video_uploader_id = None
2110         video_uploader_url = None
2111         mobj = re.search(
2112             r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">',
2113             video_webpage)
2114         if mobj is not None:
2115             video_uploader_id = mobj.group('uploader_id')
2116             video_uploader_url = mobj.group('uploader_url')
2117         else:
2118             self._downloader.report_warning('unable to extract uploader nickname')
2119
2120         channel_id = (
2121             str_or_none(video_details.get('channelId'))
2122             or self._html_search_meta(
2123                 'channelId', video_webpage, 'channel id', default=None)
2124             or self._search_regex(
2125                 r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
2126                 video_webpage, 'channel id', default=None, group='id'))
2127         channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
2128
2129         # thumbnail image
2130         # We try first to get a high quality image:
2131         m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">',
2132                             video_webpage, re.DOTALL)
2133         if m_thumb is not None:
2134             video_thumbnail = m_thumb.group(1)
2135         elif 'thumbnail_url' not in video_info:
2136             self._downloader.report_warning('unable to extract video thumbnail')
2137             video_thumbnail = None
2138         else:   # don't panic if we can't find it
2139             video_thumbnail = compat_urllib_parse_unquote_plus(video_info['thumbnail_url'][0])
2140
2141         # upload date
2142         upload_date = self._html_search_meta(
2143             'datePublished', video_webpage, 'upload date', default=None)
2144         if not upload_date:
2145             upload_date = self._search_regex(
2146                 [r'(?s)id="eow-date.*?>(.*?)</span>',
2147                  r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'],
2148                 video_webpage, 'upload date', default=None)
2149         upload_date = unified_strdate(upload_date)
2150
2151         video_license = self._html_search_regex(
2152             r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li',
2153             video_webpage, 'license', default=None)
2154
2155         m_music = re.search(
2156             r'''(?x)
2157                 <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s*
2158                 <ul[^>]*>\s*
2159                 <li>(?P<title>.+?)
2160                 by (?P<creator>.+?)
2161                 (?:
2162                     \(.+?\)|
2163                     <a[^>]*
2164                         (?:
2165                             \bhref=["\']/red[^>]*>|             # drop possible
2166                             >\s*Listen ad-free with YouTube Red # YouTube Red ad
2167                         )
2168                     .*?
2169                 )?</li
2170             ''',
2171             video_webpage)
2172         if m_music:
2173             video_alt_title = remove_quotes(unescapeHTML(m_music.group('title')))
2174             video_creator = clean_html(m_music.group('creator'))
2175         else:
2176             video_alt_title = video_creator = None
2177
2178         def extract_meta(field):
2179             return self._html_search_regex(
2180                 r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field,
2181                 video_webpage, field, default=None)
2182
2183         track = extract_meta('Song')
2184         artist = extract_meta('Artist')
2185         album = extract_meta('Album')
2186
2187         # Youtube Music Auto-generated description
2188         release_date = release_year = None
2189         if video_description:
2190             mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^·]+)·(?P<artist>[^\n]+)\n+(?P<album>[^\n]+)(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?(.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?', video_description)
2191             if mobj:
2192                 if not track:
2193                     track = mobj.group('track').strip()
2194                 if not artist:
2195                     artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
2196                 if not album:
2197                     album = mobj.group('album'.strip())
2198                 release_year = mobj.group('release_year')
2199                 release_date = mobj.group('release_date')
2200                 if release_date:
2201                     release_date = release_date.replace('-', '')
2202                     if not release_year:
2203                         release_year = int(release_date[:4])
2204                 if release_year:
2205                     release_year = int(release_year)
2206
2207         m_episode = re.search(
2208             r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
2209             video_webpage)
2210         if m_episode:
2211             series = unescapeHTML(m_episode.group('series'))
2212             season_number = int(m_episode.group('season'))
2213             episode_number = int(m_episode.group('episode'))
2214         else:
2215             series = season_number = episode_number = None
2216
2217         m_cat_container = self._search_regex(
2218             r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>',
2219             video_webpage, 'categories', default=None)
2220         if m_cat_container:
2221             category = self._html_search_regex(
2222                 r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category',
2223                 default=None)
2224             video_categories = None if category is None else [category]
2225         else:
2226             video_categories = None
2227
2228         video_tags = [
2229             unescapeHTML(m.group('content'))
2230             for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)]
2231
2232         def _extract_count(count_name):
2233             return str_to_int(self._search_regex(
2234                 r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>'
2235                 % re.escape(count_name),
2236                 video_webpage, count_name, default=None))
2237
2238         like_count = _extract_count('like')
2239         dislike_count = _extract_count('dislike')
2240
2241         if view_count is None:
2242             view_count = str_to_int(self._search_regex(
2243                 r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
2244                 'view count', default=None))
2245
2246         average_rating = (
2247             float_or_none(video_details.get('averageRating'))
2248             or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
2249
2250         # subtitles
2251         video_subtitles = self.extract_subtitles(video_id, video_webpage)
2252         automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
2253
2254         video_duration = try_get(
2255             video_info, lambda x: int_or_none(x['length_seconds'][0]))
2256         if not video_duration:
2257             video_duration = int_or_none(video_details.get('lengthSeconds'))
2258         if not video_duration:
2259             video_duration = parse_duration(self._html_search_meta(
2260                 'duration', video_webpage, 'video duration'))
2261
2262         # annotations
2263         video_annotations = None
2264         if self._downloader.params.get('writeannotations', False):
2265             video_annotations = self._extract_annotations(video_id)
2266
2267         chapters = self._extract_chapters(description_original, video_duration)
2268
2269         # Look for the DASH manifest
2270         if self._downloader.params.get('youtube_include_dash_manifest', True):
2271             dash_mpd_fatal = True
2272             for mpd_url in dash_mpds:
2273                 dash_formats = {}
2274                 try:
2275                     def decrypt_sig(mobj):
2276                         s = mobj.group(1)
2277                         dec_s = self._decrypt_signature(s, video_id, player_url, age_gate)
2278                         return '/signature/%s' % dec_s
2279
2280                     mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url)
2281
2282                     for df in self._extract_mpd_formats(
2283                             mpd_url, video_id, fatal=dash_mpd_fatal,
2284                             formats_dict=self._formats):
2285                         if not df.get('filesize'):
2286                             df['filesize'] = _extract_filesize(df['url'])
2287                         # Do not overwrite DASH format found in some previous DASH manifest
2288                         if df['format_id'] not in dash_formats:
2289                             dash_formats[df['format_id']] = df
2290                         # Additional DASH manifests may end up in HTTP Error 403 therefore
2291                         # allow them to fail without bug report message if we already have
2292                         # some DASH manifest succeeded. This is temporary workaround to reduce
2293                         # burst of bug reports until we figure out the reason and whether it
2294                         # can be fixed at all.
2295                         dash_mpd_fatal = False
2296                 except (ExtractorError, KeyError) as e:
2297                     self.report_warning(
2298                         'Skipping DASH manifest: %r' % e, video_id)
2299                 if dash_formats:
2300                     # Remove the formats we found through non-DASH, they
2301                     # contain less info and it can be wrong, because we use
2302                     # fixed values (for example the resolution). See
2303                     # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
2304                     # example.
2305                     formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
2306                     formats.extend(dash_formats.values())
2307
2308         # Check for malformed aspect ratio
2309         stretched_m = re.search(
2310             r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">',
2311             video_webpage)
2312         if stretched_m:
2313             w = float(stretched_m.group('w'))
2314             h = float(stretched_m.group('h'))
2315             # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0).
2316             # We will only process correct ratios.
2317             if w > 0 and h > 0:
2318                 ratio = w / h
2319                 for f in formats:
2320                     if f.get('vcodec') != 'none':
2321                         f['stretched_ratio'] = ratio
2322
2323         if not formats:
2324             token = extract_token(video_info)
2325             if not token:
2326                 if 'reason' in video_info:
2327                     if 'The uploader has not made this video available in your country.' in video_info['reason']:
2328                         regions_allowed = self._html_search_meta(
2329                             'regionsAllowed', video_webpage, default=None)
2330                         countries = regions_allowed.split(',') if regions_allowed else None
2331                         self.raise_geo_restricted(
2332                             msg=video_info['reason'][0], countries=countries)
2333                     reason = video_info['reason'][0]
2334                     if 'Invalid parameters' in reason:
2335                         unavailable_message = extract_unavailable_message()
2336                         if unavailable_message:
2337                             reason = unavailable_message
2338                     raise ExtractorError(
2339                         'YouTube said: %s' % reason,
2340                         expected=True, video_id=video_id)
2341                 else:
2342                     raise ExtractorError(
2343                         '"token" parameter not in video info for unknown reason',
2344                         video_id=video_id)
2345
2346         if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
2347             raise ExtractorError('This video is DRM protected.', expected=True)
2348
2349         self._sort_formats(formats)
2350
2351         self.mark_watched(video_id, video_info, player_response)
2352
2353         return {
2354             'id': video_id,
2355             'uploader': video_uploader,
2356             'uploader_id': video_uploader_id,
2357             'uploader_url': video_uploader_url,
2358             'channel_id': channel_id,
2359             'channel_url': channel_url,
2360             'upload_date': upload_date,
2361             'license': video_license,
2362             'creator': video_creator or artist,
2363             'title': video_title,
2364             'alt_title': video_alt_title or track,
2365             'thumbnail': video_thumbnail,
2366             'description': video_description,
2367             'categories': video_categories,
2368             'tags': video_tags,
2369             'subtitles': video_subtitles,
2370             'automatic_captions': automatic_captions,
2371             'duration': video_duration,
2372             'age_limit': 18 if age_gate else 0,
2373             'annotations': video_annotations,
2374             'chapters': chapters,
2375             'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id,
2376             'view_count': view_count,
2377             'like_count': like_count,
2378             'dislike_count': dislike_count,
2379             'average_rating': average_rating,
2380             'formats': formats,
2381             'is_live': is_live,
2382             'start_time': start_time,
2383             'end_time': end_time,
2384             'series': series,
2385             'season_number': season_number,
2386             'episode_number': episode_number,
2387             'track': track,
2388             'artist': artist,
2389             'album': album,
2390             'release_date': release_date,
2391             'release_year': release_year,
2392         }
2393
2394
2395 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
2396     IE_DESC = 'YouTube.com playlists'
2397     _VALID_URL = r"""(?x)(?:
2398                         (?:https?://)?
2399                         (?:\w+\.)?
2400                         (?:
2401                             (?:
2402                                 youtube\.com|
2403                                 invidio\.us
2404                             )
2405                             /
2406                             (?:
2407                                (?:course|view_play_list|my_playlists|artist|playlist|watch|embed/(?:videoseries|[0-9A-Za-z_-]{11}))
2408                                \? (?:.*?[&;])*? (?:p|a|list)=
2409                             |  p/
2410                             )|
2411                             youtu\.be/[0-9A-Za-z_-]{11}\?.*?\blist=
2412                         )
2413                         (
2414                             (?:PL|LL|EC|UU|FL|RD|UL|TL|OLAK5uy_)?[0-9A-Za-z-_]{10,}
2415                             # Top tracks, they can also include dots
2416                             |(?:MC)[\w\.]*
2417                         )
2418                         .*
2419                      |
2420                         (%(playlist_id)s)
2421                      )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
2422     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s'
2423     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
2424     IE_NAME = 'youtube:playlist'
2425     _TESTS = [{
2426         'url': 'https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2427         'info_dict': {
2428             'title': 'ytdl test PL',
2429             'id': 'PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re',
2430         },
2431         'playlist_count': 3,
2432     }, {
2433         'url': 'https://www.youtube.com/playlist?list=PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2434         'info_dict': {
2435             'id': 'PLtPgu7CB4gbZDA7i_euNxn75ISqxwZPYx',
2436             'title': 'YDL_Empty_List',
2437         },
2438         'playlist_count': 0,
2439         'skip': 'This playlist is private',
2440     }, {
2441         'note': 'Playlist with deleted videos (#651). As a bonus, the video #51 is also twice in this list.',
2442         'url': 'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2443         'info_dict': {
2444             'title': '29C3: Not my department',
2445             'id': 'PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC',
2446         },
2447         'playlist_count': 95,
2448     }, {
2449         'note': 'issue #673',
2450         'url': 'PLBB231211A4F62143',
2451         'info_dict': {
2452             'title': '[OLD]Team Fortress 2 (Class-based LP)',
2453             'id': 'PLBB231211A4F62143',
2454         },
2455         'playlist_mincount': 26,
2456     }, {
2457         'note': 'Large playlist',
2458         'url': 'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q',
2459         'info_dict': {
2460             'title': 'Uploads from Cauchemar',
2461             'id': 'UUBABnxM4Ar9ten8Mdjj1j0Q',
2462         },
2463         'playlist_mincount': 799,
2464     }, {
2465         'url': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2466         'info_dict': {
2467             'title': 'YDL_safe_search',
2468             'id': 'PLtPgu7CB4gbY9oDN3drwC3cMbJggS7dKl',
2469         },
2470         'playlist_count': 2,
2471         'skip': 'This playlist is private',
2472     }, {
2473         'note': 'embedded',
2474         'url': 'https://www.youtube.com/embed/videoseries?list=PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2475         'playlist_count': 4,
2476         'info_dict': {
2477             'title': 'JODA15',
2478             'id': 'PL6IaIsEjSbf96XFRuNccS_RuEXwNdsoEu',
2479         }
2480     }, {
2481         'url': 'http://www.youtube.com/embed/_xDOZElKyNU?list=PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2482         'playlist_mincount': 485,
2483         'info_dict': {
2484             'title': '2017 華語最新單曲 (2/24更新)',
2485             'id': 'PLsyOSbh5bs16vubvKePAQ1x3PhKavfBIl',
2486         }
2487     }, {
2488         'note': 'Embedded SWF player',
2489         'url': 'https://www.youtube.com/p/YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ?hl=en_US&fs=1&rel=0',
2490         'playlist_count': 4,
2491         'info_dict': {
2492             'title': 'JODA7',
2493             'id': 'YN5VISEtHet5D4NEvfTd0zcgFk84NqFZ',
2494         }
2495     }, {
2496         'note': 'Buggy playlist: the webpage has a "Load more" button but it doesn\'t have more videos',
2497         'url': 'https://www.youtube.com/playlist?list=UUXw-G3eDE9trcvY2sBMM_aA',
2498         'info_dict': {
2499             'title': 'Uploads from Interstellar Movie',
2500             'id': 'UUXw-G3eDE9trcvY2sBMM_aA',
2501         },
2502         'playlist_mincount': 21,
2503     }, {
2504         # Playlist URL that does not actually serve a playlist
2505         'url': 'https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4',
2506         'info_dict': {
2507             'id': 'FqZTN594JQw',
2508             'ext': 'webm',
2509             'title': "Smiley's People 01 detective, Adventure Series, Action",
2510             'uploader': 'STREEM',
2511             'uploader_id': 'UCyPhqAZgwYWZfxElWVbVJng',
2512             'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCyPhqAZgwYWZfxElWVbVJng',
2513             'upload_date': '20150526',
2514             'license': 'Standard YouTube License',
2515             'description': 'md5:507cdcb5a49ac0da37a920ece610be80',
2516             'categories': ['People & Blogs'],
2517             'tags': list,
2518             'view_count': int,
2519             'like_count': int,
2520             'dislike_count': int,
2521         },
2522         'params': {
2523             'skip_download': True,
2524         },
2525         'add_ie': [YoutubeIE.ie_key()],
2526     }, {
2527         'url': 'https://youtu.be/yeWKywCrFtk?list=PL2qgrgXsNUG5ig9cat4ohreBjYLAPC0J5',
2528         'info_dict': {
2529             'id': 'yeWKywCrFtk',
2530             'ext': 'mp4',
2531             'title': 'Small Scale Baler and Braiding Rugs',
2532             'uploader': 'Backus-Page House Museum',
2533             'uploader_id': 'backuspagemuseum',
2534             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/backuspagemuseum',
2535             'upload_date': '20161008',
2536             'license': 'Standard YouTube License',
2537             'description': 'md5:800c0c78d5eb128500bffd4f0b4f2e8a',
2538             'categories': ['Nonprofits & Activism'],
2539             'tags': list,
2540             'like_count': int,
2541             'dislike_count': int,
2542         },
2543         'params': {
2544             'noplaylist': True,
2545             'skip_download': True,
2546         },
2547     }, {
2548         'url': 'https://youtu.be/uWyaPkt-VOI?list=PL9D9FC436B881BA21',
2549         'only_matching': True,
2550     }, {
2551         'url': 'TLGGrESM50VT6acwMjAyMjAxNw',
2552         'only_matching': True,
2553     }, {
2554         # music album playlist
2555         'url': 'OLAK5uy_m4xAFdmMC5rX3Ji3g93pQe3hqLZw_9LhM',
2556         'only_matching': True,
2557     }, {
2558         'url': 'https://invidio.us/playlist?list=PLDIoUOhQQPlXr63I_vwF9GD8sAKh77dWU',
2559         'only_matching': True,
2560     }]
2561
2562     def _real_initialize(self):
2563         self._login()
2564
2565     def _extract_mix(self, playlist_id):
2566         # The mixes are generated from a single video
2567         # the id of the playlist is just 'RD' + video_id
2568         ids = []
2569         last_id = playlist_id[-11:]
2570         for n in itertools.count(1):
2571             url = 'https://youtube.com/watch?v=%s&list=%s' % (last_id, playlist_id)
2572             webpage = self._download_webpage(
2573                 url, playlist_id, 'Downloading page {0} of Youtube mix'.format(n))
2574             new_ids = orderedSet(re.findall(
2575                 r'''(?xs)data-video-username=".*?".*?
2576                            href="/watch\?v=([0-9A-Za-z_-]{11})&amp;[^"]*?list=%s''' % re.escape(playlist_id),
2577                 webpage))
2578             # Fetch new pages until all the videos are repeated, it seems that
2579             # there are always 51 unique videos.
2580             new_ids = [_id for _id in new_ids if _id not in ids]
2581             if not new_ids:
2582                 break
2583             ids.extend(new_ids)
2584             last_id = ids[-1]
2585
2586         url_results = self._ids_to_results(ids)
2587
2588         search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
2589         title_span = (
2590             search_title('playlist-title')
2591             or search_title('title long-title')
2592             or search_title('title'))
2593         title = clean_html(title_span)
2594
2595         return self.playlist_result(url_results, playlist_id, title)
2596
2597     def _extract_playlist(self, playlist_id):
2598         url = self._TEMPLATE_URL % playlist_id
2599         page = self._download_webpage(url, playlist_id)
2600
2601         # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
2602         for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
2603             match = match.strip()
2604             # Check if the playlist exists or is private
2605             mobj = re.match(r'[^<]*(?:The|This) playlist (?P<reason>does not exist|is private)[^<]*', match)
2606             if mobj:
2607                 reason = mobj.group('reason')
2608                 message = 'This playlist %s' % reason
2609                 if 'private' in reason:
2610                     message += ', use --username or --netrc to access it'
2611                 message += '.'
2612                 raise ExtractorError(message, expected=True)
2613             elif re.match(r'[^<]*Invalid parameters[^<]*', match):
2614                 raise ExtractorError(
2615                     'Invalid parameters. Maybe URL is incorrect.',
2616                     expected=True)
2617             elif re.match(r'[^<]*Choose your language[^<]*', match):
2618                 continue
2619             else:
2620                 self.report_warning('Youtube gives an alert message: ' + match)
2621
2622         playlist_title = self._html_search_regex(
2623             r'(?s)<h1 class="pl-header-title[^"]*"[^>]*>\s*(.*?)\s*</h1>',
2624             page, 'title', default=None)
2625
2626         _UPLOADER_BASE = r'class=["\']pl-header-details[^>]+>\s*<li>\s*<a[^>]+\bhref='
2627         uploader = self._search_regex(
2628             r'%s["\']/(?:user|channel)/[^>]+>([^<]+)' % _UPLOADER_BASE,
2629             page, 'uploader', default=None)
2630         mobj = re.search(
2631             r'%s(["\'])(?P<path>/(?:user|channel)/(?P<uploader_id>.+?))\1' % _UPLOADER_BASE,
2632             page)
2633         if mobj:
2634             uploader_id = mobj.group('uploader_id')
2635             uploader_url = compat_urlparse.urljoin(url, mobj.group('path'))
2636         else:
2637             uploader_id = uploader_url = None
2638
2639         has_videos = True
2640
2641         if not playlist_title:
2642             try:
2643                 # Some playlist URLs don't actually serve a playlist (e.g.
2644                 # https://www.youtube.com/watch?v=FqZTN594JQw&list=PLMYEtVRpaqY00V9W81Cwmzp6N6vZqfUKD4)
2645                 next(self._entries(page, playlist_id))
2646             except StopIteration:
2647                 has_videos = False
2648
2649         playlist = self.playlist_result(
2650             self._entries(page, playlist_id), playlist_id, playlist_title)
2651         playlist.update({
2652             'uploader': uploader,
2653             'uploader_id': uploader_id,
2654             'uploader_url': uploader_url,
2655         })
2656
2657         return has_videos, playlist
2658
2659     def _check_download_just_video(self, url, playlist_id):
2660         # Check if it's a video-specific URL
2661         query_dict = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
2662         video_id = query_dict.get('v', [None])[0] or self._search_regex(
2663             r'(?:(?:^|//)youtu\.be/|youtube\.com/embed/(?!videoseries))([0-9A-Za-z_-]{11})', url,
2664             'video id', default=None)
2665         if video_id:
2666             if self._downloader.params.get('noplaylist'):
2667                 self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
2668                 return video_id, self.url_result(video_id, 'Youtube', video_id=video_id)
2669             else:
2670                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id))
2671                 return video_id, None
2672         return None, None
2673
2674     def _real_extract(self, url):
2675         # Extract playlist id
2676         mobj = re.match(self._VALID_URL, url)
2677         if mobj is None:
2678             raise ExtractorError('Invalid URL: %s' % url)
2679         playlist_id = mobj.group(1) or mobj.group(2)
2680
2681         video_id, video = self._check_download_just_video(url, playlist_id)
2682         if video:
2683             return video
2684
2685         if playlist_id.startswith(('RD', 'UL', 'PU')):
2686             # Mixes require a custom extraction process
2687             return self._extract_mix(playlist_id)
2688
2689         has_videos, playlist = self._extract_playlist(playlist_id)
2690         if has_videos or not video_id:
2691             return playlist
2692
2693         # Some playlist URLs don't actually serve a playlist (see
2694         # https://github.com/ytdl-org/youtube-dl/issues/10537).
2695         # Fallback to plain video extraction if there is a video id
2696         # along with playlist id.
2697         return self.url_result(video_id, 'Youtube', video_id=video_id)
2698
2699
2700 class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
2701     IE_DESC = 'YouTube.com channels'
2702     _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com|(?:www\.)?invidio\.us)/channel/(?P<id>[0-9A-Za-z_-]+)'
2703     _TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
2704     _VIDEO_RE = r'(?:title="(?P<title>[^"]+)"[^>]+)?href="/watch\?v=(?P<id>[0-9A-Za-z_-]+)&?'
2705     IE_NAME = 'youtube:channel'
2706     _TESTS = [{
2707         'note': 'paginated channel',
2708         'url': 'https://www.youtube.com/channel/UCKfVa3S1e4PHvxWcwyMMg8w',
2709         'playlist_mincount': 91,
2710         'info_dict': {
2711             'id': 'UUKfVa3S1e4PHvxWcwyMMg8w',
2712             'title': 'Uploads from lex will',
2713         }
2714     }, {
2715         'note': 'Age restricted channel',
2716         # from https://www.youtube.com/user/DeusExOfficial
2717         'url': 'https://www.youtube.com/channel/UCs0ifCMCm1icqRbqhUINa0w',
2718         'playlist_mincount': 64,
2719         'info_dict': {
2720             'id': 'UUs0ifCMCm1icqRbqhUINa0w',
2721             'title': 'Uploads from Deus Ex',
2722         },
2723     }, {
2724         'url': 'https://invidio.us/channel/UC23qupoDRn9YOAVzeoxjOQA',
2725         'only_matching': True,
2726     }]
2727
2728     @classmethod
2729     def suitable(cls, url):
2730         return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
2731                 else super(YoutubeChannelIE, cls).suitable(url))
2732
2733     def _build_template_url(self, url, channel_id):
2734         return self._TEMPLATE_URL % channel_id
2735
2736     def _real_extract(self, url):
2737         channel_id = self._match_id(url)
2738
2739         url = self._build_template_url(url, channel_id)
2740
2741         # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
2742         # Workaround by extracting as a playlist if managed to obtain channel playlist URL
2743         # otherwise fallback on channel by page extraction
2744         channel_page = self._download_webpage(
2745             url + '?view=57', channel_id,
2746             'Downloading channel page', fatal=False)
2747         if channel_page is False:
2748             channel_playlist_id = False
2749         else:
2750             channel_playlist_id = self._html_search_meta(
2751                 'channelId', channel_page, 'channel id', default=None)
2752             if not channel_playlist_id:
2753                 channel_url = self._html_search_meta(
2754                     ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
2755                     channel_page, 'channel url', default=None)
2756                 if channel_url:
2757                     channel_playlist_id = self._search_regex(
2758                         r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
2759                         channel_url, 'channel id', default=None)
2760         if channel_playlist_id and channel_playlist_id.startswith('UC'):
2761             playlist_id = 'UU' + channel_playlist_id[2:]
2762             return self.url_result(
2763                 compat_urlparse.urljoin(url, '/playlist?list=%s' % playlist_id), 'YoutubePlaylist')
2764
2765         channel_page = self._download_webpage(url, channel_id, 'Downloading page #1')
2766         autogenerated = re.search(r'''(?x)
2767                 class="[^"]*?(?:
2768                     channel-header-autogenerated-label|
2769                     yt-channel-title-autogenerated
2770                 )[^"]*"''', channel_page) is not None
2771
2772         if autogenerated:
2773             # The videos are contained in a single page
2774             # the ajax pages can't be used, they are empty
2775             entries = [
2776                 self.url_result(
2777                     video_id, 'Youtube', video_id=video_id,
2778                     video_title=video_title)
2779                 for video_id, video_title in self.extract_videos_from_page(channel_page)]
2780             return self.playlist_result(entries, channel_id)
2781
2782         try:
2783             next(self._entries(channel_page, channel_id))
2784         except StopIteration:
2785             alert_message = self._html_search_regex(
2786                 r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
2787                 channel_page, 'alert', default=None, group='alert')
2788             if alert_message:
2789                 raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
2790
2791         return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
2792
2793
2794 class YoutubeUserIE(YoutubeChannelIE):
2795     IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
2796     _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results|shared)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
2797     _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
2798     IE_NAME = 'youtube:user'
2799
2800     _TESTS = [{
2801         'url': 'https://www.youtube.com/user/TheLinuxFoundation',
2802         'playlist_mincount': 320,
2803         'info_dict': {
2804             'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
2805             'title': 'Uploads from The Linux Foundation',
2806         }
2807     }, {
2808         # Only available via https://www.youtube.com/c/12minuteathlete/videos
2809         # but not https://www.youtube.com/user/12minuteathlete/videos
2810         'url': 'https://www.youtube.com/c/12minuteathlete/videos',
2811         'playlist_mincount': 249,
2812         'info_dict': {
2813             'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
2814             'title': 'Uploads from 12 Minute Athlete',
2815         }
2816     }, {
2817         'url': 'ytuser:phihag',
2818         'only_matching': True,
2819     }, {
2820         'url': 'https://www.youtube.com/c/gametrailers',
2821         'only_matching': True,
2822     }, {
2823         'url': 'https://www.youtube.com/gametrailers',
2824         'only_matching': True,
2825     }, {
2826         # This channel is not available, geo restricted to JP
2827         'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
2828         'only_matching': True,
2829     }]
2830
2831     @classmethod
2832     def suitable(cls, url):
2833         # Don't return True if the url can be extracted with other youtube
2834         # extractor, the regex would is too permissive and it would match.
2835         other_yt_ies = iter(klass for (name, klass) in globals().items() if name.startswith('Youtube') and name.endswith('IE') and klass is not cls)
2836         if any(ie.suitable(url) for ie in other_yt_ies):
2837             return False
2838         else:
2839             return super(YoutubeUserIE, cls).suitable(url)
2840
2841     def _build_template_url(self, url, channel_id):
2842         mobj = re.match(self._VALID_URL, url)
2843         return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
2844
2845
2846 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
2847     IE_DESC = 'YouTube.com live streams'
2848     _VALID_URL = r'(?P<base_url>https?://(?:\w+\.)?youtube\.com/(?:(?:user|channel|c)/)?(?P<id>[^/]+))/live'
2849     IE_NAME = 'youtube:live'
2850
2851     _TESTS = [{
2852         'url': 'https://www.youtube.com/user/TheYoungTurks/live',
2853         'info_dict': {
2854             'id': 'a48o2S1cPoo',
2855             'ext': 'mp4',
2856             'title': 'The Young Turks - Live Main Show',
2857             'uploader': 'The Young Turks',
2858             'uploader_id': 'TheYoungTurks',
2859             'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheYoungTurks',
2860             'upload_date': '20150715',
2861             'license': 'Standard YouTube License',
2862             'description': 'md5:438179573adcdff3c97ebb1ee632b891',
2863             'categories': ['News & Politics'],
2864             'tags': ['Cenk Uygur (TV Program Creator)', 'The Young Turks (Award-Winning Work)', 'Talk Show (TV Genre)'],
2865             'like_count': int,
2866             'dislike_count': int,
2867         },
2868         'params': {
2869             'skip_download': True,
2870         },
2871     }, {
2872         'url': 'https://www.youtube.com/channel/UC1yBKRuGpC1tSM73A0ZjYjQ/live',
2873         'only_matching': True,
2874     }, {
2875         'url': 'https://www.youtube.com/c/CommanderVideoHq/live',
2876         'only_matching': True,
2877     }, {
2878         'url': 'https://www.youtube.com/TheYoungTurks/live',
2879         'only_matching': True,
2880     }]
2881
2882     def _real_extract(self, url):
2883         mobj = re.match(self._VALID_URL, url)
2884         channel_id = mobj.group('id')
2885         base_url = mobj.group('base_url')
2886         webpage = self._download_webpage(url, channel_id, fatal=False)
2887         if webpage:
2888             page_type = self._og_search_property(
2889                 'type', webpage, 'page type', default='')
2890             video_id = self._html_search_meta(
2891                 'videoId', webpage, 'video id', default=None)
2892             if page_type.startswith('video') and video_id and re.match(
2893                     r'^[0-9A-Za-z_-]{11}$', video_id):
2894                 return self.url_result(video_id, YoutubeIE.ie_key())
2895         return self.url_result(base_url)
2896
2897
2898 class YoutubePlaylistsIE(YoutubePlaylistsBaseInfoExtractor):
2899     IE_DESC = 'YouTube.com user/channel playlists'
2900     _VALID_URL = r'https?://(?:\w+\.)?youtube\.com/(?:user|channel)/(?P<id>[^/]+)/playlists'
2901     IE_NAME = 'youtube:playlists'
2902
2903     _TESTS = [{
2904         'url': 'https://www.youtube.com/user/ThirstForScience/playlists',
2905         'playlist_mincount': 4,
2906         'info_dict': {
2907             'id': 'ThirstForScience',
2908             'title': 'Thirst for Science',
2909         },
2910     }, {
2911         # with "Load more" button
2912         'url': 'https://www.youtube.com/user/igorkle1/playlists?view=1&sort=dd',
2913         'playlist_mincount': 70,
2914         'info_dict': {
2915             'id': 'igorkle1',
2916             'title': 'Игорь Клейнер',
2917         },
2918     }, {
2919         'url': 'https://www.youtube.com/channel/UCiU1dHvZObB2iP6xkJ__Icw/playlists',
2920         'playlist_mincount': 17,
2921         'info_dict': {
2922             'id': 'UCiU1dHvZObB2iP6xkJ__Icw',
2923             'title': 'Chem Player',
2924         },
2925     }]
2926
2927
2928 class YoutubeSearchBaseInfoExtractor(YoutubePlaylistBaseInfoExtractor):
2929     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
2930
2931
2932 class YoutubeSearchIE(SearchInfoExtractor, YoutubeSearchBaseInfoExtractor):
2933     IE_DESC = 'YouTube.com searches'
2934     # there doesn't appear to be a real limit, for example if you search for
2935     # 'python' you get more than 8.000.000 results
2936     _MAX_RESULTS = float('inf')
2937     IE_NAME = 'youtube:search'
2938     _SEARCH_KEY = 'ytsearch'
2939     _EXTRA_QUERY_ARGS = {}
2940     _TESTS = []
2941
2942     def _get_n_results(self, query, n):
2943         """Get a specified number of results for a query"""
2944
2945         videos = []
2946         limit = n
2947
2948         url_query = {
2949             'search_query': query.encode('utf-8'),
2950         }
2951         url_query.update(self._EXTRA_QUERY_ARGS)
2952         result_url = 'https://www.youtube.com/results?' + compat_urllib_parse_urlencode(url_query)
2953
2954         for pagenum in itertools.count(1):
2955             data = self._download_json(
2956                 result_url, video_id='query "%s"' % query,
2957                 note='Downloading page %s' % pagenum,
2958                 errnote='Unable to download API page',
2959                 query={'spf': 'navigate'})
2960             html_content = data[1]['body']['content']
2961
2962             if 'class="search-message' in html_content:
2963                 raise ExtractorError(
2964                     '[youtube] No video results', expected=True)
2965
2966             new_videos = list(self._process_page(html_content))
2967             videos += new_videos
2968             if not new_videos or len(videos) > limit:
2969                 break
2970             next_link = self._html_search_regex(
2971                 r'href="(/results\?[^"]*\bsp=[^"]+)"[^>]*>\s*<span[^>]+class="[^"]*\byt-uix-button-content\b[^"]*"[^>]*>Next',
2972                 html_content, 'next link', default=None)
2973             if next_link is None:
2974                 break
2975             result_url = compat_urlparse.urljoin('https://www.youtube.com/', next_link)
2976
2977         if len(videos) > n:
2978             videos = videos[:n]
2979         return self.playlist_result(videos, query)
2980
2981
2982 class YoutubeSearchDateIE(YoutubeSearchIE):
2983     IE_NAME = YoutubeSearchIE.IE_NAME + ':date'
2984     _SEARCH_KEY = 'ytsearchdate'
2985     IE_DESC = 'YouTube.com searches, newest videos first'
2986     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
2987
2988
2989 class YoutubeSearchURLIE(YoutubeSearchBaseInfoExtractor):
2990     IE_DESC = 'YouTube.com search URLs'
2991     IE_NAME = 'youtube:search_url'
2992     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
2993     _TESTS = [{
2994         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
2995         'playlist_mincount': 5,
2996         'info_dict': {
2997             'title': 'youtube-dl test video',
2998         }
2999     }, {
3000         'url': 'https://www.youtube.com/results?q=test&sp=EgQIBBgB',
3001         'only_matching': True,
3002     }]
3003
3004     def _real_extract(self, url):
3005         mobj = re.match(self._VALID_URL, url)
3006         query = compat_urllib_parse_unquote_plus(mobj.group('query'))
3007         webpage = self._download_webpage(url, query)
3008         return self.playlist_result(self._process_page(webpage), playlist_title=query)
3009
3010
3011 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):
3012     IE_DESC = 'YouTube.com (multi-season) shows'
3013     _VALID_URL = r'https?://(?:www\.)?youtube\.com/show/(?P<id>[^?#]*)'
3014     IE_NAME = 'youtube:show'
3015     _TESTS = [{
3016         'url': 'https://www.youtube.com/show/airdisasters',
3017         'playlist_mincount': 5,
3018         'info_dict': {
3019             'id': 'airdisasters',
3020             'title': 'Air Disasters',
3021         }
3022     }]
3023
3024     def _real_extract(self, url):
3025         playlist_id = self._match_id(url)
3026         return super(YoutubeShowIE, self)._real_extract(
3027             'https://www.youtube.com/show/%s/playlists' % playlist_id)
3028
3029
3030 class YoutubeFeedsInfoExtractor(YoutubeBaseInfoExtractor):
3031     """
3032     Base class for feed extractors
3033     Subclasses must define the _FEED_NAME and _PLAYLIST_TITLE properties.
3034     """
3035     _LOGIN_REQUIRED = True
3036
3037     @property
3038     def IE_NAME(self):
3039         return 'youtube:%s' % self._FEED_NAME
3040
3041     def _real_initialize(self):
3042         self._login()
3043
3044     def _entries(self, page):
3045         # The extraction process is the same as for playlists, but the regex
3046         # for the video ids doesn't contain an index
3047         ids = []
3048         more_widget_html = content_html = page
3049         for page_num in itertools.count(1):
3050             matches = re.findall(r'href="\s*/watch\?v=([0-9A-Za-z_-]{11})', content_html)
3051
3052             # 'recommended' feed has infinite 'load more' and each new portion spins
3053             # the same videos in (sometimes) slightly different order, so we'll check
3054             # for unicity and break when portion has no new videos
3055             new_ids = list(filter(lambda video_id: video_id not in ids, orderedSet(matches)))
3056             if not new_ids:
3057                 break
3058
3059             ids.extend(new_ids)
3060
3061             for entry in self._ids_to_results(new_ids):
3062                 yield entry
3063
3064             mobj = re.search(r'data-uix-load-more-href="/?(?P<more>[^"]+)"', more_widget_html)
3065             if not mobj:
3066                 break
3067
3068             more = self._download_json(
3069                 'https://youtube.com/%s' % mobj.group('more'), self._PLAYLIST_TITLE,
3070                 'Downloading page #%s' % page_num,
3071                 transform_source=uppercase_escape)
3072             content_html = more['content_html']
3073             more_widget_html = more['load_more_widget_html']
3074
3075     def _real_extract(self, url):
3076         page = self._download_webpage(
3077             'https://www.youtube.com/feed/%s' % self._FEED_NAME,
3078             self._PLAYLIST_TITLE)
3079         return self.playlist_result(
3080             self._entries(page), playlist_title=self._PLAYLIST_TITLE)
3081
3082
3083 class YoutubeWatchLaterIE(YoutubePlaylistIE):
3084     IE_NAME = 'youtube:watchlater'
3085     IE_DESC = 'Youtube watch later list, ":ytwatchlater" for short (requires authentication)'
3086     _VALID_URL = r'https?://(?:www\.)?youtube\.com/(?:feed/watch_later|(?:playlist|watch)\?(?:.+&)?list=WL)|:ytwatchlater'
3087
3088     _TESTS = [{
3089         'url': 'https://www.youtube.com/playlist?list=WL',
3090         'only_matching': True,
3091     }, {
3092         'url': 'https://www.youtube.com/watch?v=bCNU9TrbiRk&index=1&list=WL',
3093         'only_matching': True,
3094     }]
3095
3096     def _real_extract(self, url):
3097         _, video = self._check_download_just_video(url, 'WL')
3098         if video:
3099             return video
3100         _, playlist = self._extract_playlist('WL')
3101         return playlist
3102
3103
3104 class YoutubeFavouritesIE(YoutubeBaseInfoExtractor):
3105     IE_NAME = 'youtube:favorites'
3106     IE_DESC = 'YouTube.com favourite videos, ":ytfav" for short (requires authentication)'
3107     _VALID_URL = r'https?://(?:www\.)?youtube\.com/my_favorites|:ytfav(?:ou?rites)?'
3108     _LOGIN_REQUIRED = True
3109
3110     def _real_extract(self, url):
3111         webpage = self._download_webpage('https://www.youtube.com/my_favorites', 'Youtube Favourites videos')
3112         playlist_id = self._search_regex(r'list=(.+?)["&]', webpage, 'favourites playlist id')
3113         return self.url_result(playlist_id, 'YoutubePlaylist')
3114
3115
3116 class YoutubeRecommendedIE(YoutubeFeedsInfoExtractor):
3117     IE_DESC = 'YouTube.com recommended videos, ":ytrec" for short (requires authentication)'
3118     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/recommended|:ytrec(?:ommended)?'
3119     _FEED_NAME = 'recommended'
3120     _PLAYLIST_TITLE = 'Youtube Recommended videos'
3121
3122
3123 class YoutubeSubscriptionsIE(YoutubeFeedsInfoExtractor):
3124     IE_DESC = 'YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)'
3125     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/subscriptions|:ytsubs(?:criptions)?'
3126     _FEED_NAME = 'subscriptions'
3127     _PLAYLIST_TITLE = 'Youtube Subscriptions'
3128
3129
3130 class YoutubeHistoryIE(YoutubeFeedsInfoExtractor):
3131     IE_DESC = 'Youtube watch history, ":ythistory" for short (requires authentication)'
3132     _VALID_URL = r'https?://(?:www\.)?youtube\.com/feed/history|:ythistory'
3133     _FEED_NAME = 'history'
3134     _PLAYLIST_TITLE = 'Youtube History'
3135
3136
3137 class YoutubeTruncatedURLIE(InfoExtractor):
3138     IE_NAME = 'youtube:truncated_url'
3139     IE_DESC = False  # Do not list
3140     _VALID_URL = r'''(?x)
3141         (?:https?://)?
3142         (?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie)?\.com/
3143         (?:watch\?(?:
3144             feature=[a-z_]+|
3145             annotation_id=annotation_[^&]+|
3146             x-yt-cl=[0-9]+|
3147             hl=[^&]*|
3148             t=[0-9]+
3149         )?
3150         |
3151             attribution_link\?a=[^&]+
3152         )
3153         $
3154     '''
3155
3156     _TESTS = [{
3157         'url': 'https://www.youtube.com/watch?annotation_id=annotation_3951667041',
3158         'only_matching': True,
3159     }, {
3160         'url': 'https://www.youtube.com/watch?',
3161         'only_matching': True,
3162     }, {
3163         'url': 'https://www.youtube.com/watch?x-yt-cl=84503534',
3164         'only_matching': True,
3165     }, {
3166         'url': 'https://www.youtube.com/watch?feature=foo',
3167         'only_matching': True,
3168     }, {
3169         'url': 'https://www.youtube.com/watch?hl=en-GB',
3170         'only_matching': True,
3171     }, {
3172         'url': 'https://www.youtube.com/watch?t=2372',
3173         'only_matching': True,
3174     }]
3175
3176     def _real_extract(self, url):
3177         raise ExtractorError(
3178             'Did you forget to quote the URL? Remember that & is a meta '
3179             'character in most shells, so you want to put the URL in quotes, '
3180             'like  youtube-dl '
3181             '"https://www.youtube.com/watch?feature=foo&v=BaW_jenozKc" '
3182             ' or simply  youtube-dl BaW_jenozKc  .',
3183             expected=True)
3184
3185
3186 class YoutubeTruncatedIDIE(InfoExtractor):
3187     IE_NAME = 'youtube:truncated_id'
3188     IE_DESC = False  # Do not list
3189     _VALID_URL = r'https?://(?:www\.)?youtube\.com/watch\?v=(?P<id>[0-9A-Za-z_-]{1,10})$'
3190
3191     _TESTS = [{
3192         'url': 'https://www.youtube.com/watch?v=N_708QY7Ob',
3193         'only_matching': True,
3194     }]
3195
3196     def _real_extract(self, url):
3197         video_id = self._match_id(url)
3198         raise ExtractorError(
3199             'Incomplete YouTube ID %s. URL %s looks truncated.' % (video_id, url),
3200             expected=True)