yt_dlp/extractor/dailymotion.py

   1 import functools
   2 import json
   3 import re
   4 import urllib.parse
   5
   6 from .common import InfoExtractor
   7 from ..networking.exceptions import HTTPError
   8 from ..utils import (
   9     ExtractorError,
  10     OnDemandPagedList,
  11     age_restricted,
  12     clean_html,
  13     int_or_none,
  14     traverse_obj,
  15     try_get,
  16     unescapeHTML,
  17     unsmuggle_url,
  18     urlencode_postdata,
  19 )
  20
  21
  22 class DailymotionBaseInfoExtractor(InfoExtractor):
  23     _FAMILY_FILTER = None
  24     _HEADERS = {
  25         'Content-Type': 'application/json',
  26         'Origin': 'https://www.dailymotion.com',
  27     }
  28     _NETRC_MACHINE = 'dailymotion'
  29
  30     def _get_dailymotion_cookies(self):
  31         return self._get_cookies('https://www.dailymotion.com/')
  32
  33     @staticmethod
  34     def _get_cookie_value(cookies, name):
  35         cookie = cookies.get(name)
  36         if cookie:
  37             return cookie.value
  38
  39     def _set_dailymotion_cookie(self, name, value):
  40         self._set_cookie('www.dailymotion.com', name, value)
  41
  42     def _real_initialize(self):
  43         cookies = self._get_dailymotion_cookies()
  44         ff = self._get_cookie_value(cookies, 'ff')
  45         self._FAMILY_FILTER = ff == 'on' if ff else age_restricted(18, self.get_param('age_limit'))
  46         self._set_dailymotion_cookie('ff', 'on' if self._FAMILY_FILTER else 'off')
  47
  48     def _get_token(self, xid):
  49         cookies = self._get_dailymotion_cookies()
  50         token = self._get_cookie_value(cookies, 'access_token') or self._get_cookie_value(cookies, 'client_token')
  51         if token:
  52             return token
  53
  54         data = {
  55             'client_id': 'f1a362d288c1b98099c7',
  56             'client_secret': 'eea605b96e01c796ff369935357eca920c5da4c5',
  57         }
  58         username, password = self._get_login_info()
  59         if username:
  60             data.update({
  61                 'grant_type': 'password',
  62                 'password': password,
  63                 'username': username,
  64             })
  65         else:
  66             data['grant_type'] = 'client_credentials'
  67         try:
  68             token = self._download_json(
  69                 'https://graphql.api.dailymotion.com/oauth/token',
  70                 None, 'Downloading Access Token',
  71                 data=urlencode_postdata(data))['access_token']
  72         except ExtractorError as e:
  73             if isinstance(e.cause, HTTPError) and e.cause.status == 400:
  74                 raise ExtractorError(self._parse_json(
  75                     e.cause.response.read().decode(), xid)['error_description'], expected=True)
  76             raise
  77         self._set_dailymotion_cookie('access_token' if username else 'client_token', token)
  78         return token
  79
  80     def _call_api(self, object_type, xid, object_fields, note, filter_extra=None):
  81         if not self._HEADERS.get('Authorization'):
  82             self._HEADERS['Authorization'] = f'Bearer {self._get_token(xid)}'
  83
  84         resp = self._download_json(
  85             'https://graphql.api.dailymotion.com/', xid, note, data=json.dumps({
  86                 'query': '''{
  87   %s(xid: "%s"%s) {
  88     %s
  89   }
  90 }''' % (object_type, xid, ', ' + filter_extra if filter_extra else '', object_fields),
  91             }).encode(), headers=self._HEADERS)
  92         obj = resp['data'][object_type]
  93         if not obj:
  94             raise ExtractorError(resp['errors'][0]['message'], expected=True)
  95         return obj
  96
  97
  98 class DailymotionIE(DailymotionBaseInfoExtractor):
  99     _VALID_URL = r'''(?ix)
 100                     https?://
 101                         (?:
 102                             (?:(?:www|touch|geo)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:(?:embed|swf|\#)/)|player(?:/\w+)?\.html\?)?video|swf)|
 103                             (?:www\.)?lequipe\.fr/video
 104                         )
 105                         [/=](?P<id>[^/?_&]+)(?:.+?\bplaylist=(?P<playlist_id>x[0-9a-z]+))?
 106                     '''
 107     IE_NAME = 'dailymotion'
 108     _EMBED_REGEX = [r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1']
 109     _TESTS = [{
 110         'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
 111         'md5': '074b95bdee76b9e3654137aee9c79dfe',
 112         'info_dict': {
 113             'id': 'x5kesuj',
 114             'ext': 'mp4',
 115             'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
 116             'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller',
 117             'duration': 187,
 118             'timestamp': 1493651285,
 119             'upload_date': '20170501',
 120             'uploader': 'Deadline',
 121             'uploader_id': 'x1xm8ri',
 122             'age_limit': 0,
 123             'view_count': int,
 124             'like_count': int,
 125             'tags': ['hollywood', 'celeb', 'celebrity', 'movies', 'red carpet'],
 126             'thumbnail': r're:https://(?:s[12]\.)dmcdn\.net/v/K456B1aXqIx58LKWQ/x1080',
 127         },
 128     }, {
 129         'url': 'https://geo.dailymotion.com/player.html?video=x89eyek&mute=true',
 130         'md5': 'e2f9717c6604773f963f069ca53a07f8',
 131         'info_dict': {
 132             'id': 'x89eyek',
 133             'ext': 'mp4',
 134             'title': "En quête d'esprit du 27/03/2022",
 135             'description': 'md5:66542b9f4df2eb23f314fc097488e553',
 136             'duration': 2756,
 137             'timestamp': 1648383669,
 138             'upload_date': '20220327',
 139             'uploader': 'CNEWS',
 140             'uploader_id': 'x24vth',
 141             'age_limit': 0,
 142             'view_count': int,
 143             'like_count': int,
 144             'tags': ['en_quete_d_esprit'],
 145             'thumbnail': r're:https://(?:s[12]\.)dmcdn\.net/v/Tncwi1YNg_RUl7ueu/x1080',
 146         }
 147     }, {
 148         'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
 149         'md5': '2137c41a8e78554bb09225b8eb322406',
 150         'info_dict': {
 151             'id': 'x2iuewm',
 152             'ext': 'mp4',
 153             'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
 154             'description': 'Several come bundled with the Steam Controller.',
 155             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
 156             'duration': 74,
 157             'timestamp': 1425657362,
 158             'upload_date': '20150306',
 159             'uploader': 'IGN',
 160             'uploader_id': 'xijv66',
 161             'age_limit': 0,
 162             'view_count': int,
 163         },
 164         'skip': 'video gone',
 165     }, {
 166         # Vevo video
 167         'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
 168         'info_dict': {
 169             'title': 'Roar (Official)',
 170             'id': 'USUV71301934',
 171             'ext': 'mp4',
 172             'uploader': 'Katy Perry',
 173             'upload_date': '20130905',
 174         },
 175         'params': {
 176             'skip_download': True,
 177         },
 178         'skip': 'VEVO is only available in some countries',
 179     }, {
 180         # age-restricted video
 181         'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
 182         'md5': '0d667a7b9cebecc3c89ee93099c4159d',
 183         'info_dict': {
 184             'id': 'xyh2zz',
 185             'ext': 'mp4',
 186             'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
 187             'uploader': 'HotWaves1012',
 188             'age_limit': 18,
 189         },
 190         'skip': 'video gone',
 191     }, {
 192         # geo-restricted, player v5
 193         'url': 'http://www.dailymotion.com/video/xhza0o',
 194         'only_matching': True,
 195     }, {
 196         # with subtitles
 197         'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 198         'only_matching': True,
 199     }, {
 200         'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 201         'only_matching': True,
 202     }, {
 203         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
 204         'only_matching': True,
 205     }, {
 206         'url': 'https://www.lequipe.fr/video/x791mem',
 207         'only_matching': True,
 208     }, {
 209         'url': 'https://www.lequipe.fr/video/k7MtHciueyTcrFtFKA2',
 210         'only_matching': True,
 211     }, {
 212         'url': 'https://www.dailymotion.com/video/x3z49k?playlist=xv4bw',
 213         'only_matching': True,
 214     }, {
 215         'url': 'https://geo.dailymotion.com/player/x86gw.html?video=k46oCapRs4iikoz9DWy',
 216         'only_matching': True,
 217     }, {
 218         'url': 'https://geo.dailymotion.com/player/xakln.html?video=x8mjju4&customConfig%5BcustomParams%5D=%2Ffr-fr%2Ftennis%2Fwimbledon-mens-singles%2Farticles-video',
 219         'only_matching': True,
 220     }]
 221     _GEO_BYPASS = False
 222     _COMMON_MEDIA_FIELDS = '''description
 223       geoblockedCountries {
 224         allowed
 225       }
 226       xid'''
 227
 228     @classmethod
 229     def _extract_embed_urls(cls, url, webpage):
 230         # https://developer.dailymotion.com/player#player-parameters
 231         yield from super()._extract_embed_urls(url, webpage)
 232         for mobj in re.finditer(
 233                 r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P<id>[0-9a-zA-Z]+).+?}\s*\);', webpage):
 234             yield from 'https://www.dailymotion.com/embed/video/' + mobj.group('id')
 235
 236     def _real_extract(self, url):
 237         url, smuggled_data = unsmuggle_url(url)
 238         video_id, playlist_id = self._match_valid_url(url).groups()
 239
 240         if playlist_id:
 241             if self._yes_playlist(playlist_id, video_id):
 242                 return self.url_result(
 243                     'http://www.dailymotion.com/playlist/' + playlist_id,
 244                     'DailymotionPlaylist', playlist_id)
 245
 246         password = self.get_param('videopassword')
 247         media = self._call_api(
 248             'media', video_id, '''... on Video {
 249       %s
 250       stats {
 251         likes {
 252           total
 253         }
 254         views {
 255           total
 256         }
 257       }
 258     }
 259     ... on Live {
 260       %s
 261       audienceCount
 262       isOnAir
 263     }''' % (self._COMMON_MEDIA_FIELDS, self._COMMON_MEDIA_FIELDS), 'Downloading media JSON metadata',
 264             'password: "%s"' % self.get_param('videopassword') if password else None)
 265         xid = media['xid']
 266
 267         metadata = self._download_json(
 268             'https://www.dailymotion.com/player/metadata/video/' + xid,
 269             xid, 'Downloading metadata JSON',
 270             query=traverse_obj(smuggled_data, 'query') or {'app': 'com.dailymotion.neon'})
 271
 272         error = metadata.get('error')
 273         if error:
 274             title = error.get('title') or error['raw_message']
 275             # See https://developer.dailymotion.com/api#access-error
 276             if error.get('code') == 'DM007':
 277                 allowed_countries = try_get(media, lambda x: x['geoblockedCountries']['allowed'], list)
 278                 self.raise_geo_restricted(msg=title, countries=allowed_countries)
 279             raise ExtractorError(
 280                 '%s said: %s' % (self.IE_NAME, title), expected=True)
 281
 282         title = metadata['title']
 283         is_live = media.get('isOnAir')
 284         formats = []
 285         for quality, media_list in metadata['qualities'].items():
 286             for m in media_list:
 287                 media_url = m.get('url')
 288                 media_type = m.get('type')
 289                 if not media_url or media_type == 'application/vnd.lumberjack.manifest':
 290                     continue
 291                 if media_type == 'application/x-mpegURL':
 292                     formats.extend(self._extract_m3u8_formats(
 293                         media_url, video_id, 'mp4', live=is_live, m3u8_id='hls', fatal=False))
 294                 else:
 295                     f = {
 296                         'url': media_url,
 297                         'format_id': 'http-' + quality,
 298                     }
 299                     m = re.search(r'/H264-(\d+)x(\d+)(?:-(60)/)?', media_url)
 300                     if m:
 301                         width, height, fps = map(int_or_none, m.groups())
 302                         f.update({
 303                             'fps': fps,
 304                             'height': height,
 305                             'width': width,
 306                         })
 307                     formats.append(f)
 308         for f in formats:
 309             f['url'] = f['url'].split('#')[0]
 310             if not f.get('fps') and f['format_id'].endswith('@60'):
 311                 f['fps'] = 60
 312
 313         subtitles = {}
 314         subtitles_data = try_get(metadata, lambda x: x['subtitles']['data'], dict) or {}
 315         for subtitle_lang, subtitle in subtitles_data.items():
 316             subtitles[subtitle_lang] = [{
 317                 'url': subtitle_url,
 318             } for subtitle_url in subtitle.get('urls', [])]
 319
 320         thumbnails = []
 321         for height, poster_url in metadata.get('posters', {}).items():
 322             thumbnails.append({
 323                 'height': int_or_none(height),
 324                 'id': height,
 325                 'url': poster_url,
 326             })
 327
 328         owner = metadata.get('owner') or {}
 329         stats = media.get('stats') or {}
 330         get_count = lambda x: int_or_none(try_get(stats, lambda y: y[x + 's']['total']))
 331
 332         return {
 333             'id': video_id,
 334             'title': title,
 335             'description': clean_html(media.get('description')),
 336             'thumbnails': thumbnails,
 337             'duration': int_or_none(metadata.get('duration')) or None,
 338             'timestamp': int_or_none(metadata.get('created_time')),
 339             'uploader': owner.get('screenname'),
 340             'uploader_id': owner.get('id') or metadata.get('screenname'),
 341             'age_limit': 18 if metadata.get('explicit') else 0,
 342             'tags': metadata.get('tags'),
 343             'view_count': get_count('view') or int_or_none(media.get('audienceCount')),
 344             'like_count': get_count('like'),
 345             'formats': formats,
 346             'subtitles': subtitles,
 347             'is_live': is_live,
 348         }
 349
 350
 351 class DailymotionPlaylistBaseIE(DailymotionBaseInfoExtractor):
 352     _PAGE_SIZE = 100
 353
 354     def _fetch_page(self, playlist_id, page):
 355         page += 1
 356         videos = self._call_api(
 357             self._OBJECT_TYPE, playlist_id,
 358             '''videos(allowExplicit: %s, first: %d, page: %d) {
 359       edges {
 360         node {
 361           xid
 362           url
 363         }
 364       }
 365     }''' % ('false' if self._FAMILY_FILTER else 'true', self._PAGE_SIZE, page),
 366             'Downloading page %d' % page)['videos']
 367         for edge in videos['edges']:
 368             node = edge['node']
 369             yield self.url_result(
 370                 node['url'], DailymotionIE.ie_key(), node['xid'])
 371
 372     def _real_extract(self, url):
 373         playlist_id = self._match_id(url)
 374         entries = OnDemandPagedList(functools.partial(
 375             self._fetch_page, playlist_id), self._PAGE_SIZE)
 376         return self.playlist_result(
 377             entries, playlist_id)
 378
 379
 380 class DailymotionPlaylistIE(DailymotionPlaylistBaseIE):
 381     IE_NAME = 'dailymotion:playlist'
 382     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>x[0-9a-z]+)'
 383     _TESTS = [{
 384         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 385         'info_dict': {
 386             'id': 'xv4bw',
 387         },
 388         'playlist_mincount': 20,
 389     }]
 390     _OBJECT_TYPE = 'collection'
 391
 392     @classmethod
 393     def _extract_embed_urls(cls, url, webpage):
 394         # Look for embedded Dailymotion playlist player (#3822)
 395         for mobj in re.finditer(
 396                 r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1',
 397                 webpage):
 398             for p in re.findall(r'list\[\]=/playlist/([^/]+)/', unescapeHTML(mobj.group('url'))):
 399                 yield '//dailymotion.com/playlist/%s' % p
 400
 401
 402 class DailymotionSearchIE(DailymotionPlaylistBaseIE):
 403     IE_NAME = 'dailymotion:search'
 404     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/search/(?P<id>[^/?#]+)/videos'
 405     _PAGE_SIZE = 20
 406     _TESTS = [{
 407         'url': 'http://www.dailymotion.com/search/king of turtles/videos',
 408         'info_dict': {
 409             'id': 'king of turtles',
 410             'title': 'king of turtles',
 411         },
 412         'playlist_mincount': 90,
 413     }]
 414     _SEARCH_QUERY = 'query SEARCH_QUERY( $query: String! $page: Int $limit: Int ) { search { videos( query: $query first: $limit page: $page ) { edges { node { xid } } } } } '
 415
 416     def _call_search_api(self, term, page, note):
 417         if not self._HEADERS.get('Authorization'):
 418             self._HEADERS['Authorization'] = f'Bearer {self._get_token(term)}'
 419         resp = self._download_json(
 420             'https://graphql.api.dailymotion.com/', None, note, data=json.dumps({
 421                 'operationName': 'SEARCH_QUERY',
 422                 'query': self._SEARCH_QUERY,
 423                 'variables': {
 424                     'limit': 20,
 425                     'page': page,
 426                     'query': term,
 427                 }
 428             }).encode(), headers=self._HEADERS)
 429         obj = traverse_obj(resp, ('data', 'search', {dict}))
 430         if not obj:
 431             raise ExtractorError(
 432                 traverse_obj(resp, ('errors', 0, 'message', {str})) or 'Could not fetch search data')
 433
 434         return obj
 435
 436     def _fetch_page(self, term, page):
 437         page += 1
 438         response = self._call_search_api(term, page, f'Searching "{term}" page {page}')
 439         for xid in traverse_obj(response, ('videos', 'edges', ..., 'node', 'xid')):
 440             yield self.url_result(f'https://www.dailymotion.com/video/{xid}', DailymotionIE, xid)
 441
 442     def _real_extract(self, url):
 443         term = urllib.parse.unquote_plus(self._match_id(url))
 444         return self.playlist_result(
 445             OnDemandPagedList(functools.partial(self._fetch_page, term), self._PAGE_SIZE), term, term)
 446
 447
 448 class DailymotionUserIE(DailymotionPlaylistBaseIE):
 449     IE_NAME = 'dailymotion:user'
 450     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist|search)/)(?:(?:old/)?user/)?(?P<id>[^/?#]+)'
 451     _TESTS = [{
 452         'url': 'https://www.dailymotion.com/user/nqtv',
 453         'info_dict': {
 454             'id': 'nqtv',
 455         },
 456         'playlist_mincount': 152,
 457     }, {
 458         'url': 'http://www.dailymotion.com/user/UnderProject',
 459         'info_dict': {
 460             'id': 'UnderProject',
 461         },
 462         'playlist_mincount': 1000,
 463         'skip': 'Takes too long time',
 464     }, {
 465         'url': 'https://www.dailymotion.com/user/nqtv',
 466         'info_dict': {
 467             'id': 'nqtv',
 468         },
 469         'playlist_mincount': 148,
 470         'params': {
 471             'age_limit': 0,
 472         },
 473     }]
 474     _OBJECT_TYPE = 'channel'