yt_dlp/extractor/dailymotion.py

   1 import functools
   2 import json
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     OnDemandPagedList,
  10     age_restricted,
  11     clean_html,
  12     int_or_none,
  13     traverse_obj,
  14     try_get,
  15     unescapeHTML,
  16     unsmuggle_url,
  17     urlencode_postdata,
  18 )
  19
  20
  21 class DailymotionBaseInfoExtractor(InfoExtractor):
  22     _FAMILY_FILTER = None
  23     _HEADERS = {
  24         'Content-Type': 'application/json',
  25         'Origin': 'https://www.dailymotion.com',
  26     }
  27     _NETRC_MACHINE = 'dailymotion'
  28
  29     def _get_dailymotion_cookies(self):
  30         return self._get_cookies('https://www.dailymotion.com/')
  31
  32     @staticmethod
  33     def _get_cookie_value(cookies, name):
  34         cookie = cookies.get(name)
  35         if cookie:
  36             return cookie.value
  37
  38     def _set_dailymotion_cookie(self, name, value):
  39         self._set_cookie('www.dailymotion.com', name, value)
  40
  41     def _real_initialize(self):
  42         cookies = self._get_dailymotion_cookies()
  43         ff = self._get_cookie_value(cookies, 'ff')
  44         self._FAMILY_FILTER = ff == 'on' if ff else age_restricted(18, self.get_param('age_limit'))
  45         self._set_dailymotion_cookie('ff', 'on' if self._FAMILY_FILTER else 'off')
  46
  47     def _call_api(self, object_type, xid, object_fields, note, filter_extra=None):
  48         if not self._HEADERS.get('Authorization'):
  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 not token:
  52                 data = {
  53                     'client_id': 'f1a362d288c1b98099c7',
  54                     'client_secret': 'eea605b96e01c796ff369935357eca920c5da4c5',
  55                 }
  56                 username, password = self._get_login_info()
  57                 if username:
  58                     data.update({
  59                         'grant_type': 'password',
  60                         'password': password,
  61                         'username': username,
  62                     })
  63                 else:
  64                     data['grant_type'] = 'client_credentials'
  65                 try:
  66                     token = self._download_json(
  67                         'https://graphql.api.dailymotion.com/oauth/token',
  68                         None, 'Downloading Access Token',
  69                         data=urlencode_postdata(data))['access_token']
  70                 except ExtractorError as e:
  71                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  72                         raise ExtractorError(self._parse_json(
  73                             e.cause.read().decode(), xid)['error_description'], expected=True)
  74                     raise
  75                 self._set_dailymotion_cookie('access_token' if username else 'client_token', token)
  76             self._HEADERS['Authorization'] = 'Bearer ' + token
  77
  78         resp = self._download_json(
  79             'https://graphql.api.dailymotion.com/', xid, note, data=json.dumps({
  80                 'query': '''{
  81   %s(xid: "%s"%s) {
  82     %s
  83   }
  84 }''' % (object_type, xid, ', ' + filter_extra if filter_extra else '', object_fields),
  85             }).encode(), headers=self._HEADERS)
  86         obj = resp['data'][object_type]
  87         if not obj:
  88             raise ExtractorError(resp['errors'][0]['message'], expected=True)
  89         return obj
  90
  91
  92 class DailymotionIE(DailymotionBaseInfoExtractor):
  93     _VALID_URL = r'''(?ix)
  94                     https?://
  95                         (?:
  96                             (?:(?:www|touch|geo)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:(?:embed|swf|\#)/)|player\.html\?)?video|swf)|
  97                             (?:www\.)?lequipe\.fr/video
  98                         )
  99                         [/=](?P<id>[^/?_&]+)(?:.+?\bplaylist=(?P<playlist_id>x[0-9a-z]+))?
 100                     '''
 101     IE_NAME = 'dailymotion'
 102     _TESTS = [{
 103         'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
 104         'md5': '074b95bdee76b9e3654137aee9c79dfe',
 105         'info_dict': {
 106             'id': 'x5kesuj',
 107             'ext': 'mp4',
 108             'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
 109             'description': 'Office Christmas Party Review -  Jason Bateman, Olivia Munn, T.J. Miller',
 110             'duration': 187,
 111             'timestamp': 1493651285,
 112             'upload_date': '20170501',
 113             'uploader': 'Deadline',
 114             'uploader_id': 'x1xm8ri',
 115             'age_limit': 0,
 116         },
 117     }, {
 118         'url': 'https://geo.dailymotion.com/player.html?video=x89eyek&mute=true',
 119         'md5': 'e2f9717c6604773f963f069ca53a07f8',
 120         'info_dict': {
 121             'id': 'x89eyek',
 122             'ext': 'mp4',
 123             'title': "En quête d'esprit du 27/03/2022",
 124             'description': 'md5:66542b9f4df2eb23f314fc097488e553',
 125             'duration': 2756,
 126             'timestamp': 1648383669,
 127             'upload_date': '20220327',
 128             'uploader': 'CNEWS',
 129             'uploader_id': 'x24vth',
 130             'age_limit': 0,
 131             'view_count': int,
 132             'like_count': int,
 133             'tags': ['en_quete_d_esprit'],
 134             'thumbnail': 'https://s2.dmcdn.net/v/Tncwi1YGKdvFbDuDY/x1080',
 135         }
 136     }, {
 137         'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
 138         'md5': '2137c41a8e78554bb09225b8eb322406',
 139         'info_dict': {
 140             'id': 'x2iuewm',
 141             'ext': 'mp4',
 142             'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
 143             'description': 'Several come bundled with the Steam Controller.',
 144             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
 145             'duration': 74,
 146             'timestamp': 1425657362,
 147             'upload_date': '20150306',
 148             'uploader': 'IGN',
 149             'uploader_id': 'xijv66',
 150             'age_limit': 0,
 151             'view_count': int,
 152         },
 153         'skip': 'video gone',
 154     }, {
 155         # Vevo video
 156         'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
 157         'info_dict': {
 158             'title': 'Roar (Official)',
 159             'id': 'USUV71301934',
 160             'ext': 'mp4',
 161             'uploader': 'Katy Perry',
 162             'upload_date': '20130905',
 163         },
 164         'params': {
 165             'skip_download': True,
 166         },
 167         'skip': 'VEVO is only available in some countries',
 168     }, {
 169         # age-restricted video
 170         'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
 171         'md5': '0d667a7b9cebecc3c89ee93099c4159d',
 172         'info_dict': {
 173             'id': 'xyh2zz',
 174             'ext': 'mp4',
 175             'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
 176             'uploader': 'HotWaves1012',
 177             'age_limit': 18,
 178         },
 179         'skip': 'video gone',
 180     }, {
 181         # geo-restricted, player v5
 182         'url': 'http://www.dailymotion.com/video/xhza0o',
 183         'only_matching': True,
 184     }, {
 185         # with subtitles
 186         'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 187         'only_matching': True,
 188     }, {
 189         'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 190         'only_matching': True,
 191     }, {
 192         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
 193         'only_matching': True,
 194     }, {
 195         'url': 'https://www.lequipe.fr/video/x791mem',
 196         'only_matching': True,
 197     }, {
 198         'url': 'https://www.lequipe.fr/video/k7MtHciueyTcrFtFKA2',
 199         'only_matching': True,
 200     }, {
 201         'url': 'https://www.dailymotion.com/video/x3z49k?playlist=xv4bw',
 202         'only_matching': True,
 203     }]
 204     _GEO_BYPASS = False
 205     _COMMON_MEDIA_FIELDS = '''description
 206       geoblockedCountries {
 207         allowed
 208       }
 209       xid'''
 210
 211     @staticmethod
 212     def _extract_urls(webpage):
 213         urls = []
 214         # Look for embedded Dailymotion player
 215         # https://developer.dailymotion.com/player#player-parameters
 216         for mobj in re.finditer(
 217                 r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage):
 218             urls.append(unescapeHTML(mobj.group('url')))
 219         for mobj in re.finditer(
 220                 r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P<id>[0-9a-zA-Z]+).+?}\s*\);', webpage):
 221             urls.append('https://www.dailymotion.com/embed/video/' + mobj.group('id'))
 222         return urls
 223
 224     def _real_extract(self, url):
 225         url, smuggled_data = unsmuggle_url(url)
 226         video_id, playlist_id = self._match_valid_url(url).groups()
 227
 228         if playlist_id:
 229             if self._yes_playlist(playlist_id, video_id):
 230                 return self.url_result(
 231                     'http://www.dailymotion.com/playlist/' + playlist_id,
 232                     'DailymotionPlaylist', playlist_id)
 233
 234         password = self.get_param('videopassword')
 235         media = self._call_api(
 236             'media', video_id, '''... on Video {
 237       %s
 238       stats {
 239         likes {
 240           total
 241         }
 242         views {
 243           total
 244         }
 245       }
 246     }
 247     ... on Live {
 248       %s
 249       audienceCount
 250       isOnAir
 251     }''' % (self._COMMON_MEDIA_FIELDS, self._COMMON_MEDIA_FIELDS), 'Downloading media JSON metadata',
 252             'password: "%s"' % self.get_param('videopassword') if password else None)
 253         xid = media['xid']
 254
 255         metadata = self._download_json(
 256             'https://www.dailymotion.com/player/metadata/video/' + xid,
 257             xid, 'Downloading metadata JSON',
 258             query=traverse_obj(smuggled_data, 'query') or {'app': 'com.dailymotion.neon'})
 259
 260         error = metadata.get('error')
 261         if error:
 262             title = error.get('title') or error['raw_message']
 263             # See https://developer.dailymotion.com/api#access-error
 264             if error.get('code') == 'DM007':
 265                 allowed_countries = try_get(media, lambda x: x['geoblockedCountries']['allowed'], list)
 266                 self.raise_geo_restricted(msg=title, countries=allowed_countries)
 267             raise ExtractorError(
 268                 '%s said: %s' % (self.IE_NAME, title), expected=True)
 269
 270         title = metadata['title']
 271         is_live = media.get('isOnAir')
 272         formats = []
 273         for quality, media_list in metadata['qualities'].items():
 274             for m in media_list:
 275                 media_url = m.get('url')
 276                 media_type = m.get('type')
 277                 if not media_url or media_type == 'application/vnd.lumberjack.manifest':
 278                     continue
 279                 if media_type == 'application/x-mpegURL':
 280                     formats.extend(self._extract_m3u8_formats(
 281                         media_url, video_id, 'mp4', live=is_live, m3u8_id='hls', fatal=False))
 282                 else:
 283                     f = {
 284                         'url': media_url,
 285                         'format_id': 'http-' + quality,
 286                     }
 287                     m = re.search(r'/H264-(\d+)x(\d+)(?:-(60)/)?', media_url)
 288                     if m:
 289                         width, height, fps = map(int_or_none, m.groups())
 290                         f.update({
 291                             'fps': fps,
 292                             'height': height,
 293                             'width': width,
 294                         })
 295                     formats.append(f)
 296         for f in formats:
 297             f['url'] = f['url'].split('#')[0]
 298             if not f.get('fps') and f['format_id'].endswith('@60'):
 299                 f['fps'] = 60
 300         self._sort_formats(formats)
 301
 302         subtitles = {}
 303         subtitles_data = try_get(metadata, lambda x: x['subtitles']['data'], dict) or {}
 304         for subtitle_lang, subtitle in subtitles_data.items():
 305             subtitles[subtitle_lang] = [{
 306                 'url': subtitle_url,
 307             } for subtitle_url in subtitle.get('urls', [])]
 308
 309         thumbnails = []
 310         for height, poster_url in metadata.get('posters', {}).items():
 311             thumbnails.append({
 312                 'height': int_or_none(height),
 313                 'id': height,
 314                 'url': poster_url,
 315             })
 316
 317         owner = metadata.get('owner') or {}
 318         stats = media.get('stats') or {}
 319         get_count = lambda x: int_or_none(try_get(stats, lambda y: y[x + 's']['total']))
 320
 321         return {
 322             'id': video_id,
 323             'title': title,
 324             'description': clean_html(media.get('description')),
 325             'thumbnails': thumbnails,
 326             'duration': int_or_none(metadata.get('duration')) or None,
 327             'timestamp': int_or_none(metadata.get('created_time')),
 328             'uploader': owner.get('screenname'),
 329             'uploader_id': owner.get('id') or metadata.get('screenname'),
 330             'age_limit': 18 if metadata.get('explicit') else 0,
 331             'tags': metadata.get('tags'),
 332             'view_count': get_count('view') or int_or_none(media.get('audienceCount')),
 333             'like_count': get_count('like'),
 334             'formats': formats,
 335             'subtitles': subtitles,
 336             'is_live': is_live,
 337         }
 338
 339
 340 class DailymotionPlaylistBaseIE(DailymotionBaseInfoExtractor):
 341     _PAGE_SIZE = 100
 342
 343     def _fetch_page(self, playlist_id, page):
 344         page += 1
 345         videos = self._call_api(
 346             self._OBJECT_TYPE, playlist_id,
 347             '''videos(allowExplicit: %s, first: %d, page: %d) {
 348       edges {
 349         node {
 350           xid
 351           url
 352         }
 353       }
 354     }''' % ('false' if self._FAMILY_FILTER else 'true', self._PAGE_SIZE, page),
 355             'Downloading page %d' % page)['videos']
 356         for edge in videos['edges']:
 357             node = edge['node']
 358             yield self.url_result(
 359                 node['url'], DailymotionIE.ie_key(), node['xid'])
 360
 361     def _real_extract(self, url):
 362         playlist_id = self._match_id(url)
 363         entries = OnDemandPagedList(functools.partial(
 364             self._fetch_page, playlist_id), self._PAGE_SIZE)
 365         return self.playlist_result(
 366             entries, playlist_id)
 367
 368
 369 class DailymotionPlaylistIE(DailymotionPlaylistBaseIE):
 370     IE_NAME = 'dailymotion:playlist'
 371     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>x[0-9a-z]+)'
 372     _TESTS = [{
 373         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 374         'info_dict': {
 375             'id': 'xv4bw',
 376         },
 377         'playlist_mincount': 20,
 378     }]
 379     _OBJECT_TYPE = 'collection'
 380
 381
 382 class DailymotionUserIE(DailymotionPlaylistBaseIE):
 383     IE_NAME = 'dailymotion:user'
 384     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<id>[^/]+)'
 385     _TESTS = [{
 386         'url': 'https://www.dailymotion.com/user/nqtv',
 387         'info_dict': {
 388             'id': 'nqtv',
 389         },
 390         'playlist_mincount': 152,
 391     }, {
 392         'url': 'http://www.dailymotion.com/user/UnderProject',
 393         'info_dict': {
 394             'id': 'UnderProject',
 395         },
 396         'playlist_mincount': 1000,
 397         'skip': 'Takes too long time',
 398     }, {
 399         'url': 'https://www.dailymotion.com/user/nqtv',
 400         'info_dict': {
 401             'id': 'nqtv',
 402         },
 403         'playlist_mincount': 148,
 404         'params': {
 405             'age_limit': 0,
 406         },
 407     }]
 408     _OBJECT_TYPE = 'channel'