youtube_dl/extractor/dailymotion.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import json
   6 import itertools
   7
   8 from .common import InfoExtractor
   9
  10 from ..utils import (
  11     determine_ext,
  12     error_to_compat_str,
  13     ExtractorError,
  14     int_or_none,
  15     parse_iso8601,
  16     sanitized_Request,
  17     str_to_int,
  18     unescapeHTML,
  19 )
  20
  21
  22 class DailymotionBaseInfoExtractor(InfoExtractor):
  23     @staticmethod
  24     def _build_request(url):
  25         """Build a request with the family filter disabled"""
  26         request = sanitized_Request(url)
  27         request.add_header('Cookie', 'family_filter=off; ff=off')
  28         return request
  29
  30     def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
  31         request = self._build_request(url)
  32         return self._download_webpage_handle(request, *args, **kwargs)
  33
  34     def _download_webpage_no_ff(self, url, *args, **kwargs):
  35         request = self._build_request(url)
  36         return self._download_webpage(request, *args, **kwargs)
  37
  38
  39 class DailymotionIE(DailymotionBaseInfoExtractor):
  40     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:embed|swf|#)/)?video/(?P<id>[^/?_]+)'
  41     IE_NAME = 'dailymotion'
  42
  43     _FORMATS = [
  44         ('stream_h264_ld_url', 'ld'),
  45         ('stream_h264_url', 'standard'),
  46         ('stream_h264_hq_url', 'hq'),
  47         ('stream_h264_hd_url', 'hd'),
  48         ('stream_h264_hd1080_url', 'hd180'),
  49     ]
  50
  51     _TESTS = [
  52         {
  53             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  54             'md5': '2137c41a8e78554bb09225b8eb322406',
  55             'info_dict': {
  56                 'id': 'x2iuewm',
  57                 'ext': 'mp4',
  58                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  59                 'description': 'Several come bundled with the Steam Controller.',
  60                 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
  61                 'duration': 74,
  62                 'timestamp': 1425657362,
  63                 'upload_date': '20150306',
  64                 'uploader': 'IGN',
  65                 'uploader_id': 'xijv66',
  66                 'age_limit': 0,
  67                 'view_count': int,
  68                 'comment_count': int,
  69             }
  70         },
  71         # Vevo video
  72         {
  73             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  74             'info_dict': {
  75                 'title': 'Roar (Official)',
  76                 'id': 'USUV71301934',
  77                 'ext': 'mp4',
  78                 'uploader': 'Katy Perry',
  79                 'upload_date': '20130905',
  80             },
  81             'params': {
  82                 'skip_download': True,
  83             },
  84             'skip': 'VEVO is only available in some countries',
  85         },
  86         # age-restricted video
  87         {
  88             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  89             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  90             'info_dict': {
  91                 'id': 'xyh2zz',
  92                 'ext': 'mp4',
  93                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  94                 'uploader': 'HotWaves1012',
  95                 'age_limit': 18,
  96             }
  97         },
  98         # geo-restricted, player v5
  99         {
 100             'url': 'http://www.dailymotion.com/video/xhza0o',
 101             'only_matching': True,
 102         },
 103         # with subtitles
 104         {
 105             'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 106             'only_matching': True,
 107         },
 108         {
 109             'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 110             'only_matching': True,
 111         }
 112     ]
 113
 114     def _real_extract(self, url):
 115         video_id = self._match_id(url)
 116
 117         webpage = self._download_webpage_no_ff(
 118             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 119
 120         age_limit = self._rta_search(webpage)
 121
 122         description = self._og_search_description(webpage) or self._html_search_meta(
 123             'description', webpage, 'description')
 124
 125         view_count = str_to_int(self._search_regex(
 126             [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
 127              r'video_views_count[^>]+>\s+([\d\.,]+)'],
 128             webpage, 'view count', fatal=False))
 129         comment_count = int_or_none(self._search_regex(
 130             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 131             webpage, 'comment count', fatal=False))
 132
 133         player_v5 = self._search_regex(
 134             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 135              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 136              r'buildPlayer\(({.+?})\);'],
 137             webpage, 'player v5', default=None)
 138         if player_v5:
 139             player = self._parse_json(player_v5, video_id)
 140             metadata = player['metadata']
 141
 142             self._check_error(metadata)
 143
 144             formats = []
 145             for quality, media_list in metadata['qualities'].items():
 146                 for media in media_list:
 147                     media_url = media.get('url')
 148                     if not media_url:
 149                         continue
 150                     type_ = media.get('type')
 151                     if type_ == 'application/vnd.lumberjack.manifest':
 152                         continue
 153                     ext = determine_ext(media_url)
 154                     if type_ == 'application/x-mpegURL' or ext == 'm3u8':
 155                         formats.extend(self._extract_m3u8_formats(
 156                             media_url, video_id, 'mp4', preference=-1,
 157                             m3u8_id='hls', fatal=False))
 158                     elif type_ == 'application/f4m' or ext == 'f4m':
 159                         formats.extend(self._extract_f4m_formats(
 160                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 161                     else:
 162                         f = {
 163                             'url': media_url,
 164                             'format_id': 'http-%s' % quality,
 165                         }
 166                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 167                         if m:
 168                             f.update({
 169                                 'width': int(m.group('width')),
 170                                 'height': int(m.group('height')),
 171                             })
 172                         formats.append(f)
 173             self._sort_formats(formats)
 174
 175             title = metadata['title']
 176             duration = int_or_none(metadata.get('duration'))
 177             timestamp = int_or_none(metadata.get('created_time'))
 178             thumbnail = metadata.get('poster_url')
 179             uploader = metadata.get('owner', {}).get('screenname')
 180             uploader_id = metadata.get('owner', {}).get('id')
 181
 182             subtitles = {}
 183             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 184             if subtitles_data and isinstance(subtitles_data, dict):
 185                 for subtitle_lang, subtitle in subtitles_data.items():
 186                     subtitles[subtitle_lang] = [{
 187                         'ext': determine_ext(subtitle_url),
 188                         'url': subtitle_url,
 189                     } for subtitle_url in subtitle.get('urls', [])]
 190
 191             return {
 192                 'id': video_id,
 193                 'title': title,
 194                 'description': description,
 195                 'thumbnail': thumbnail,
 196                 'duration': duration,
 197                 'timestamp': timestamp,
 198                 'uploader': uploader,
 199                 'uploader_id': uploader_id,
 200                 'age_limit': age_limit,
 201                 'view_count': view_count,
 202                 'comment_count': comment_count,
 203                 'formats': formats,
 204                 'subtitles': subtitles,
 205             }
 206
 207         # vevo embed
 208         vevo_id = self._search_regex(
 209             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 210             webpage, 'vevo embed', default=None)
 211         if vevo_id:
 212             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 213
 214         # fallback old player
 215         embed_page = self._download_webpage_no_ff(
 216             'https://www.dailymotion.com/embed/video/%s' % video_id,
 217             video_id, 'Downloading embed page')
 218
 219         timestamp = parse_iso8601(self._html_search_meta(
 220             'video:release_date', webpage, 'upload date'))
 221
 222         info = self._parse_json(
 223             self._search_regex(
 224                 r'var info = ({.*?}),$', embed_page,
 225                 'video info', flags=re.MULTILINE),
 226             video_id)
 227
 228         self._check_error(info)
 229
 230         formats = []
 231         for (key, format_id) in self._FORMATS:
 232             video_url = info.get(key)
 233             if video_url is not None:
 234                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 235                 if m_size is not None:
 236                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 237                 else:
 238                     width, height = None, None
 239                 formats.append({
 240                     'url': video_url,
 241                     'ext': 'mp4',
 242                     'format_id': format_id,
 243                     'width': width,
 244                     'height': height,
 245                 })
 246         self._sort_formats(formats)
 247
 248         # subtitles
 249         video_subtitles = self.extract_subtitles(video_id, webpage)
 250
 251         title = self._og_search_title(webpage, default=None)
 252         if title is None:
 253             title = self._html_search_regex(
 254                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 255                 'title')
 256
 257         return {
 258             'id': video_id,
 259             'formats': formats,
 260             'uploader': info['owner.screenname'],
 261             'timestamp': timestamp,
 262             'title': title,
 263             'description': description,
 264             'subtitles': video_subtitles,
 265             'thumbnail': info['thumbnail_url'],
 266             'age_limit': age_limit,
 267             'view_count': view_count,
 268             'duration': info['duration']
 269         }
 270
 271     def _check_error(self, info):
 272         if info.get('error') is not None:
 273             raise ExtractorError(
 274                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 275
 276     def _get_subtitles(self, video_id, webpage):
 277         try:
 278             sub_list = self._download_webpage(
 279                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 280                 video_id, note=False)
 281         except ExtractorError as err:
 282             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 283             return {}
 284         info = json.loads(sub_list)
 285         if (info['total'] > 0):
 286             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 287             return sub_lang_list
 288         self._downloader.report_warning('video doesn\'t have subtitles')
 289         return {}
 290
 291
 292 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 293     IE_NAME = 'dailymotion:playlist'
 294     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 295     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 296     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 297     _TESTS = [{
 298         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 299         'info_dict': {
 300             'title': 'SPORT',
 301             'id': 'xv4bw_nqtv_sport',
 302         },
 303         'playlist_mincount': 20,
 304     }]
 305
 306     def _extract_entries(self, id):
 307         video_ids = set()
 308         processed_urls = set()
 309         for pagenum in itertools.count(1):
 310             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 311             webpage, urlh = self._download_webpage_handle_no_ff(
 312                 page_url, id, 'Downloading page %s' % pagenum)
 313             if urlh.geturl() in processed_urls:
 314                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 315                     page_url, urlh.geturl()), id)
 316                 break
 317
 318             processed_urls.add(urlh.geturl())
 319
 320             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 321                 if video_id not in video_ids:
 322                     yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 323                     video_ids.add(video_id)
 324
 325             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 326                 break
 327
 328     def _real_extract(self, url):
 329         mobj = re.match(self._VALID_URL, url)
 330         playlist_id = mobj.group('id')
 331         webpage = self._download_webpage(url, playlist_id)
 332
 333         return {
 334             '_type': 'playlist',
 335             'id': playlist_id,
 336             'title': self._og_search_title(webpage),
 337             'entries': self._extract_entries(playlist_id),
 338         }
 339
 340
 341 class DailymotionUserIE(DailymotionPlaylistIE):
 342     IE_NAME = 'dailymotion:user'
 343     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 344     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 345     _TESTS = [{
 346         'url': 'https://www.dailymotion.com/user/nqtv',
 347         'info_dict': {
 348             'id': 'nqtv',
 349             'title': 'Rémi Gaillard',
 350         },
 351         'playlist_mincount': 100,
 352     }, {
 353         'url': 'http://www.dailymotion.com/user/UnderProject',
 354         'info_dict': {
 355             'id': 'UnderProject',
 356             'title': 'UnderProject',
 357         },
 358         'playlist_mincount': 1800,
 359         'expected_warnings': [
 360             'Stopped at duplicated page',
 361         ],
 362         'skip': 'Takes too long time',
 363     }]
 364
 365     def _real_extract(self, url):
 366         mobj = re.match(self._VALID_URL, url)
 367         user = mobj.group('user')
 368         webpage = self._download_webpage(
 369             'https://www.dailymotion.com/user/%s' % user, user)
 370         full_user = unescapeHTML(self._html_search_regex(
 371             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 372             webpage, 'user'))
 373
 374         return {
 375             '_type': 'playlist',
 376             'id': user,
 377             'title': full_user,
 378             'entries': self._extract_entries(user),
 379         }
 380
 381
 382 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 383     _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
 384     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 385     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 386
 387     _TESTS = [{
 388         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 389         # Tested at FranceTvInfo_2
 390         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 391         'only_matching': True,
 392     }, {
 393         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 394         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 395         'only_matching': True,
 396     }]
 397
 398     @classmethod
 399     def _extract_dmcloud_url(self, webpage):
 400         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
 401         if mobj:
 402             return mobj.group(1)
 403
 404         mobj = re.search(
 405             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
 406             webpage)
 407         if mobj:
 408             return mobj.group(1)
 409
 410     def _real_extract(self, url):
 411         video_id = self._match_id(url)
 412
 413         webpage = self._download_webpage_no_ff(url, video_id)
 414
 415         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 416
 417         video_info = self._parse_json(self._search_regex(
 418             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 419
 420         # TODO: parse ios_url, which is in fact a manifest
 421         video_url = video_info['mp4_url']
 422
 423         return {
 424             'id': video_id,
 425             'url': video_url,
 426             'title': title,
 427             'thumbnail': video_info.get('thumbnail_url'),
 428         }