youtube_dl/extractor/dailymotion.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import json
   6 import re
   7
   8 from .common import InfoExtractor
   9 from ..compat import compat_HTTPError
  10 from ..utils import (
  11     age_restricted,
  12     clean_html,
  13     ExtractorError,
  14     int_or_none,
  15     OnDemandPagedList,
  16     try_get,
  17     unescapeHTML,
  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._downloader.params.get('age_limit'))
  46         self._set_dailymotion_cookie('ff', 'on' if self._FAMILY_FILTER else 'off')
  47
  48     def _call_api(self, object_type, xid, object_fields, note, filter_extra=None):
  49         if not self._HEADERS.get('Authorization'):
  50             cookies = self._get_dailymotion_cookies()
  51             token = self._get_cookie_value(cookies, 'access_token') or self._get_cookie_value(cookies, 'client_token')
  52             if not token:
  53                 data = {
  54                     'client_id': 'f1a362d288c1b98099c7',
  55                     'client_secret': 'eea605b96e01c796ff369935357eca920c5da4c5',
  56                 }
  57                 username, password = self._get_login_info()
  58                 if username:
  59                     data.update({
  60                         'grant_type': 'password',
  61                         'password': password,
  62                         'username': username,
  63                     })
  64                 else:
  65                     data['grant_type'] = 'client_credentials'
  66                 try:
  67                     token = self._download_json(
  68                         'https://graphql.api.dailymotion.com/oauth/token',
  69                         None, 'Downloading Access Token',
  70                         data=urlencode_postdata(data))['access_token']
  71                 except ExtractorError as e:
  72                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
  73                         raise ExtractorError(self._parse_json(
  74                             e.cause.read().decode(), xid)['error_description'], expected=True)
  75                     raise
  76                 self._set_dailymotion_cookie('access_token' if username else 'client_token', token)
  77             self._HEADERS['Authorization'] = 'Bearer ' + token
  78
  79         resp = self._download_json(
  80             'https://graphql.api.dailymotion.com/', xid, note, data=json.dumps({
  81                 'query': '''{
  82   %s(xid: "%s"%s) {
  83     %s
  84   }
  85 }''' % (object_type, xid, ', ' + filter_extra if filter_extra else '', object_fields),
  86             }).encode(), headers=self._HEADERS)
  87         obj = resp['data'][object_type]
  88         if not obj:
  89             raise ExtractorError(resp['errors'][0]['message'], expected=True)
  90         return obj
  91
  92
  93 class DailymotionIE(DailymotionBaseInfoExtractor):
  94     _VALID_URL = r'''(?ix)
  95                     https?://
  96                         (?:
  97                             (?:(?:www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:embed|swf|\#)/)?video|swf)|
  98                             (?:www\.)?lequipe\.fr/video
  99                         )
 100                         /(?P<id>[^/?_]+)(?:.+?\bplaylist=(?P<playlist_id>x[0-9a-z]+))?
 101                     '''
 102     IE_NAME = 'dailymotion'
 103     _TESTS = [{
 104         'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
 105         'md5': '074b95bdee76b9e3654137aee9c79dfe',
 106         'info_dict': {
 107             'id': 'x5kesuj',
 108             'ext': 'mp4',
 109             'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
 110             'description': 'Office Christmas Party Review -  Jason Bateman, Olivia Munn, T.J. Miller',
 111             'duration': 187,
 112             'timestamp': 1493651285,
 113             'upload_date': '20170501',
 114             'uploader': 'Deadline',
 115             'uploader_id': 'x1xm8ri',
 116             'age_limit': 0,
 117         },
 118     }, {
 119         'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
 120         'md5': '2137c41a8e78554bb09225b8eb322406',
 121         'info_dict': {
 122             'id': 'x2iuewm',
 123             'ext': 'mp4',
 124             'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
 125             'description': 'Several come bundled with the Steam Controller.',
 126             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
 127             'duration': 74,
 128             'timestamp': 1425657362,
 129             'upload_date': '20150306',
 130             'uploader': 'IGN',
 131             'uploader_id': 'xijv66',
 132             'age_limit': 0,
 133             'view_count': int,
 134         },
 135         'skip': 'video gone',
 136     }, {
 137         # Vevo video
 138         'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
 139         'info_dict': {
 140             'title': 'Roar (Official)',
 141             'id': 'USUV71301934',
 142             'ext': 'mp4',
 143             'uploader': 'Katy Perry',
 144             'upload_date': '20130905',
 145         },
 146         'params': {
 147             'skip_download': True,
 148         },
 149         'skip': 'VEVO is only available in some countries',
 150     }, {
 151         # age-restricted video
 152         'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
 153         'md5': '0d667a7b9cebecc3c89ee93099c4159d',
 154         'info_dict': {
 155             'id': 'xyh2zz',
 156             'ext': 'mp4',
 157             'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
 158             'uploader': 'HotWaves1012',
 159             'age_limit': 18,
 160         },
 161         'skip': 'video gone',
 162     }, {
 163         # geo-restricted, player v5
 164         'url': 'http://www.dailymotion.com/video/xhza0o',
 165         'only_matching': True,
 166     }, {
 167         # with subtitles
 168         'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 169         'only_matching': True,
 170     }, {
 171         'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 172         'only_matching': True,
 173     }, {
 174         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
 175         'only_matching': True,
 176     }, {
 177         'url': 'https://www.lequipe.fr/video/x791mem',
 178         'only_matching': True,
 179     }, {
 180         'url': 'https://www.lequipe.fr/video/k7MtHciueyTcrFtFKA2',
 181         'only_matching': True,
 182     }, {
 183         'url': 'https://www.dailymotion.com/video/x3z49k?playlist=xv4bw',
 184         'only_matching': True,
 185     }]
 186     _GEO_BYPASS = False
 187     _COMMON_MEDIA_FIELDS = '''description
 188       geoblockedCountries {
 189         allowed
 190       }
 191       xid'''
 192
 193     @staticmethod
 194     def _extract_urls(webpage):
 195         urls = []
 196         # Look for embedded Dailymotion player
 197         # https://developer.dailymotion.com/player#player-parameters
 198         for mobj in re.finditer(
 199                 r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage):
 200             urls.append(unescapeHTML(mobj.group('url')))
 201         for mobj in re.finditer(
 202                 r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P<id>[0-9a-zA-Z]+).+?}\s*\);', webpage):
 203             urls.append('https://www.dailymotion.com/embed/video/' + mobj.group('id'))
 204         return urls
 205
 206     def _real_extract(self, url):
 207         video_id, playlist_id = re.match(self._VALID_URL, url).groups()
 208
 209         if playlist_id:
 210             if not self._downloader.params.get('noplaylist'):
 211                 self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % playlist_id)
 212                 return self.url_result(
 213                     'http://www.dailymotion.com/playlist/' + playlist_id,
 214                     'DailymotionPlaylist', playlist_id)
 215             self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
 216
 217         password = self._downloader.params.get('videopassword')
 218         media = self._call_api(
 219             'media', video_id, '''... on Video {
 220       %s
 221       stats {
 222         likes {
 223           total
 224         }
 225         views {
 226           total
 227         }
 228       }
 229     }
 230     ... on Live {
 231       %s
 232       audienceCount
 233       isOnAir
 234     }''' % (self._COMMON_MEDIA_FIELDS, self._COMMON_MEDIA_FIELDS), 'Downloading media JSON metadata',
 235             'password: "%s"' % self._downloader.params.get('videopassword') if password else None)
 236         xid = media['xid']
 237
 238         metadata = self._download_json(
 239             'https://www.dailymotion.com/player/metadata/video/' + xid,
 240             xid, 'Downloading metadata JSON',
 241             query={'app': 'com.dailymotion.neon'})
 242
 243         error = metadata.get('error')
 244         if error:
 245             title = error.get('title') or error['raw_message']
 246             # See https://developer.dailymotion.com/api#access-error
 247             if error.get('code') == 'DM007':
 248                 allowed_countries = try_get(media, lambda x: x['geoblockedCountries']['allowed'], list)
 249                 self.raise_geo_restricted(msg=title, countries=allowed_countries)
 250             raise ExtractorError(
 251                 '%s said: %s' % (self.IE_NAME, title), expected=True)
 252
 253         title = metadata['title']
 254         is_live = media.get('isOnAir')
 255         formats = []
 256         for quality, media_list in metadata['qualities'].items():
 257             for m in media_list:
 258                 media_url = m.get('url')
 259                 media_type = m.get('type')
 260                 if not media_url or media_type == 'application/vnd.lumberjack.manifest':
 261                     continue
 262                 if media_type == 'application/x-mpegURL':
 263                     formats.extend(self._extract_m3u8_formats(
 264                         media_url, video_id, 'mp4',
 265                         'm3u8' if is_live else 'm3u8_native',
 266                         m3u8_id='hls', fatal=False))
 267                 else:
 268                     f = {
 269                         'url': media_url,
 270                         'format_id': 'http-' + quality,
 271                     }
 272                     m = re.search(r'/H264-(\d+)x(\d+)(?:-(60)/)?', media_url)
 273                     if m:
 274                         width, height, fps = map(int_or_none, m.groups())
 275                         f.update({
 276                             'fps': fps,
 277                             'height': height,
 278                             'width': width,
 279                         })
 280                     formats.append(f)
 281         for f in formats:
 282             f['url'] = f['url'].split('#')[0]
 283             if not f.get('fps') and f['format_id'].endswith('@60'):
 284                 f['fps'] = 60
 285         self._sort_formats(formats)
 286
 287         subtitles = {}
 288         subtitles_data = try_get(metadata, lambda x: x['subtitles']['data'], dict) or {}
 289         for subtitle_lang, subtitle in subtitles_data.items():
 290             subtitles[subtitle_lang] = [{
 291                 'url': subtitle_url,
 292             } for subtitle_url in subtitle.get('urls', [])]
 293
 294         thumbnails = []
 295         for height, poster_url in metadata.get('posters', {}).items():
 296             thumbnails.append({
 297                 'height': int_or_none(height),
 298                 'id': height,
 299                 'url': poster_url,
 300             })
 301
 302         owner = metadata.get('owner') or {}
 303         stats = media.get('stats') or {}
 304         get_count = lambda x: int_or_none(try_get(stats, lambda y: y[x + 's']['total']))
 305
 306         return {
 307             'id': video_id,
 308             'title': self._live_title(title) if is_live else title,
 309             'description': clean_html(media.get('description')),
 310             'thumbnails': thumbnails,
 311             'duration': int_or_none(metadata.get('duration')) or None,
 312             'timestamp': int_or_none(metadata.get('created_time')),
 313             'uploader': owner.get('screenname'),
 314             'uploader_id': owner.get('id') or metadata.get('screenname'),
 315             'age_limit': 18 if metadata.get('explicit') else 0,
 316             'tags': metadata.get('tags'),
 317             'view_count': get_count('view') or int_or_none(media.get('audienceCount')),
 318             'like_count': get_count('like'),
 319             'formats': formats,
 320             'subtitles': subtitles,
 321             'is_live': is_live,
 322         }
 323
 324
 325 class DailymotionPlaylistBaseIE(DailymotionBaseInfoExtractor):
 326     _PAGE_SIZE = 100
 327
 328     def _fetch_page(self, playlist_id, page):
 329         page += 1
 330         videos = self._call_api(
 331             self._OBJECT_TYPE, playlist_id,
 332             '''videos(allowExplicit: %s, first: %d, page: %d) {
 333       edges {
 334         node {
 335           xid
 336           url
 337         }
 338       }
 339     }''' % ('false' if self._FAMILY_FILTER else 'true', self._PAGE_SIZE, page),
 340             'Downloading page %d' % page)['videos']
 341         for edge in videos['edges']:
 342             node = edge['node']
 343             yield self.url_result(
 344                 node['url'], DailymotionIE.ie_key(), node['xid'])
 345
 346     def _real_extract(self, url):
 347         playlist_id = self._match_id(url)
 348         entries = OnDemandPagedList(functools.partial(
 349             self._fetch_page, playlist_id), self._PAGE_SIZE)
 350         return self.playlist_result(
 351             entries, playlist_id)
 352
 353
 354 class DailymotionPlaylistIE(DailymotionPlaylistBaseIE):
 355     IE_NAME = 'dailymotion:playlist'
 356     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>x[0-9a-z]+)'
 357     _TESTS = [{
 358         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 359         'info_dict': {
 360             'id': 'xv4bw',
 361         },
 362         'playlist_mincount': 20,
 363     }]
 364     _OBJECT_TYPE = 'collection'
 365
 366
 367 class DailymotionUserIE(DailymotionPlaylistBaseIE):
 368     IE_NAME = 'dailymotion:user'
 369     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<id>[^/]+)'
 370     _TESTS = [{
 371         'url': 'https://www.dailymotion.com/user/nqtv',
 372         'info_dict': {
 373             'id': 'nqtv',
 374         },
 375         'playlist_mincount': 152,
 376     }, {
 377         'url': 'http://www.dailymotion.com/user/UnderProject',
 378         'info_dict': {
 379             'id': 'UnderProject',
 380         },
 381         'playlist_mincount': 1000,
 382         'skip': 'Takes too long time',
 383     }, {
 384         'url': 'https://www.dailymotion.com/user/nqtv',
 385         'info_dict': {
 386             'id': 'nqtv',
 387         },
 388         'playlist_mincount': 148,
 389         'params': {
 390             'age_limit': 0,
 391         },
 392     }]
 393     _OBJECT_TYPE = 'channel'