_ Git - youtube-dl/blob - 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     mimetype2ext,
  20 )
  21
  22
  23 class DailymotionBaseInfoExtractor(InfoExtractor):
  24     @staticmethod
  25     def _build_request(url):
  26         """Build a request with the family filter disabled"""
  27         request = sanitized_Request(url)
  28         request.add_header('Cookie', 'family_filter=off; ff=off')
  29         return request
  30
  31     def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
  32         request = self._build_request(url)
  33         return self._download_webpage_handle(request, *args, **kwargs)
  34
  35     def _download_webpage_no_ff(self, url, *args, **kwargs):
  36         request = self._build_request(url)
  37         return self._download_webpage(request, *args, **kwargs)
  38
  39
  40 class DailymotionIE(DailymotionBaseInfoExtractor):
  41     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(?:embed|swf|#)/)?video/(?P<id>[^/?_]+)'
  42     IE_NAME = 'dailymotion'
  43
  44     _FORMATS = [
  45         ('stream_h264_ld_url', 'ld'),
  46         ('stream_h264_url', 'standard'),
  47         ('stream_h264_hq_url', 'hq'),
  48         ('stream_h264_hd_url', 'hd'),
  49         ('stream_h264_hd1080_url', 'hd180'),
  50     ]
  51
  52     _TESTS = [
  53         {
  54             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  55             'md5': '2137c41a8e78554bb09225b8eb322406',
  56             'info_dict': {
  57                 'id': 'x2iuewm',
  58                 'ext': 'mp4',
  59                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  60                 'description': 'Several come bundled with the Steam Controller.',
  61                 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
  62                 'duration': 74,
  63                 'timestamp': 1425657362,
  64                 'upload_date': '20150306',
  65                 'uploader': 'IGN',
  66                 'uploader_id': 'xijv66',
  67                 'age_limit': 0,
  68                 'view_count': int,
  69                 'comment_count': int,
  70             }
  71         },
  72         # Vevo video
  73         {
  74             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  75             'info_dict': {
  76                 'title': 'Roar (Official)',
  77                 'id': 'USUV71301934',
  78                 'ext': 'mp4',
  79                 'uploader': 'Katy Perry',
  80                 'upload_date': '20130905',
  81             },
  82             'params': {
  83                 'skip_download': True,
  84             },
  85             'skip': 'VEVO is only available in some countries',
  86         },
  87         # age-restricted video
  88         {
  89             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  90             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  91             'info_dict': {
  92                 'id': 'xyh2zz',
  93                 'ext': 'mp4',
  94                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  95                 'uploader': 'HotWaves1012',
  96                 'age_limit': 18,
  97             }
  98         },
  99         # geo-restricted, player v5
 100         {
 101             'url': 'http://www.dailymotion.com/video/xhza0o',
 102             'only_matching': True,
 103         },
 104         # with subtitles
 105         {
 106             'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 107             'only_matching': True,
 108         },
 109         {
 110             'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 111             'only_matching': True,
 112         }
 113     ]
 114
 115     @staticmethod
 116     def _extract_urls(webpage):
 117         # Look for embedded Dailymotion player
 118         matches = re.findall(
 119             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
 120         return list(map(lambda m: unescapeHTML(m[1]), matches))
 121
 122     def _real_extract(self, url):
 123         video_id = self._match_id(url)
 124
 125         webpage = self._download_webpage_no_ff(
 126             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 127
 128         age_limit = self._rta_search(webpage)
 129
 130         description = self._og_search_description(webpage) or self._html_search_meta(
 131             'description', webpage, 'description')
 132
 133         view_count_str = self._search_regex(
 134             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
 135              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
 136             webpage, 'view count', fatal=False)
 137         if view_count_str:
 138             view_count_str = re.sub(r'\s', '', view_count_str)
 139         view_count = str_to_int(view_count_str)
 140         comment_count = int_or_none(self._search_regex(
 141             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 142             webpage, 'comment count', fatal=False))
 143
 144         player_v5 = self._search_regex(
 145             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 146              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 147              r'buildPlayer\(({.+?})\);'],
 148             webpage, 'player v5', default=None)
 149         if player_v5:
 150             player = self._parse_json(player_v5, video_id)
 151             metadata = player['metadata']
 152
 153             self._check_error(metadata)
 154
 155             formats = []
 156             for quality, media_list in metadata['qualities'].items():
 157                 for media in media_list:
 158                     media_url = media.get('url')
 159                     if not media_url:
 160                         continue
 161                     type_ = media.get('type')
 162                     if type_ == 'application/vnd.lumberjack.manifest':
 163                         continue
 164                     ext = mimetype2ext(type_) or determine_ext(media_url)
 165                     if ext == 'm3u8':
 166                         formats.extend(self._extract_m3u8_formats(
 167                             media_url, video_id, 'mp4', preference=-1,
 168                             m3u8_id='hls', fatal=False))
 169                     elif ext == 'f4m':
 170                         formats.extend(self._extract_f4m_formats(
 171                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 172                     else:
 173                         f = {
 174                             'url': media_url,
 175                             'format_id': 'http-%s' % quality,
 176                             'ext': ext,
 177                         }
 178                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 179                         if m:
 180                             f.update({
 181                                 'width': int(m.group('width')),
 182                                 'height': int(m.group('height')),
 183                             })
 184                         formats.append(f)
 185             self._sort_formats(formats)
 186
 187             title = metadata['title']
 188             duration = int_or_none(metadata.get('duration'))
 189             timestamp = int_or_none(metadata.get('created_time'))
 190             thumbnail = metadata.get('poster_url')
 191             uploader = metadata.get('owner', {}).get('screenname')
 192             uploader_id = metadata.get('owner', {}).get('id')
 193
 194             subtitles = {}
 195             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 196             if subtitles_data and isinstance(subtitles_data, dict):
 197                 for subtitle_lang, subtitle in subtitles_data.items():
 198                     subtitles[subtitle_lang] = [{
 199                         'ext': determine_ext(subtitle_url),
 200                         'url': subtitle_url,
 201                     } for subtitle_url in subtitle.get('urls', [])]
 202
 203             return {
 204                 'id': video_id,
 205                 'title': title,
 206                 'description': description,
 207                 'thumbnail': thumbnail,
 208                 'duration': duration,
 209                 'timestamp': timestamp,
 210                 'uploader': uploader,
 211                 'uploader_id': uploader_id,
 212                 'age_limit': age_limit,
 213                 'view_count': view_count,
 214                 'comment_count': comment_count,
 215                 'formats': formats,
 216                 'subtitles': subtitles,
 217             }
 218
 219         # vevo embed
 220         vevo_id = self._search_regex(
 221             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 222             webpage, 'vevo embed', default=None)
 223         if vevo_id:
 224             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 225
 226         # fallback old player
 227         embed_page = self._download_webpage_no_ff(
 228             'https://www.dailymotion.com/embed/video/%s' % video_id,
 229             video_id, 'Downloading embed page')
 230
 231         timestamp = parse_iso8601(self._html_search_meta(
 232             'video:release_date', webpage, 'upload date'))
 233
 234         info = self._parse_json(
 235             self._search_regex(
 236                 r'var info = ({.*?}),$', embed_page,
 237                 'video info', flags=re.MULTILINE),
 238             video_id)
 239
 240         self._check_error(info)
 241
 242         formats = []
 243         for (key, format_id) in self._FORMATS:
 244             video_url = info.get(key)
 245             if video_url is not None:
 246                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 247                 if m_size is not None:
 248                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 249                 else:
 250                     width, height = None, None
 251                 formats.append({
 252                     'url': video_url,
 253                     'ext': 'mp4',
 254                     'format_id': format_id,
 255                     'width': width,
 256                     'height': height,
 257                 })
 258         self._sort_formats(formats)
 259
 260         # subtitles
 261         video_subtitles = self.extract_subtitles(video_id, webpage)
 262
 263         title = self._og_search_title(webpage, default=None)
 264         if title is None:
 265             title = self._html_search_regex(
 266                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 267                 'title')
 268
 269         return {
 270             'id': video_id,
 271             'formats': formats,
 272             'uploader': info['owner.screenname'],
 273             'timestamp': timestamp,
 274             'title': title,
 275             'description': description,
 276             'subtitles': video_subtitles,
 277             'thumbnail': info['thumbnail_url'],
 278             'age_limit': age_limit,
 279             'view_count': view_count,
 280             'duration': info['duration']
 281         }
 282
 283     def _check_error(self, info):
 284         if info.get('error') is not None:
 285             raise ExtractorError(
 286                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 287
 288     def _get_subtitles(self, video_id, webpage):
 289         try:
 290             sub_list = self._download_webpage(
 291                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 292                 video_id, note=False)
 293         except ExtractorError as err:
 294             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 295             return {}
 296         info = json.loads(sub_list)
 297         if (info['total'] > 0):
 298             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 299             return sub_lang_list
 300         self._downloader.report_warning('video doesn\'t have subtitles')
 301         return {}
 302
 303
 304 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 305     IE_NAME = 'dailymotion:playlist'
 306     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 307     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 308     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 309     _TESTS = [{
 310         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 311         'info_dict': {
 312             'title': 'SPORT',
 313             'id': 'xv4bw_nqtv_sport',
 314         },
 315         'playlist_mincount': 20,
 316     }]
 317
 318     def _extract_entries(self, id):
 319         video_ids = set()
 320         processed_urls = set()
 321         for pagenum in itertools.count(1):
 322             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 323             webpage, urlh = self._download_webpage_handle_no_ff(
 324                 page_url, id, 'Downloading page %s' % pagenum)
 325             if urlh.geturl() in processed_urls:
 326                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 327                     page_url, urlh.geturl()), id)
 328                 break
 329
 330             processed_urls.add(urlh.geturl())
 331
 332             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 333                 if video_id not in video_ids:
 334                     yield self.url_result(
 335                         'http://www.dailymotion.com/video/%s' % video_id,
 336                         DailymotionIE.ie_key(), video_id)
 337                     video_ids.add(video_id)
 338
 339             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 340                 break
 341
 342     def _real_extract(self, url):
 343         mobj = re.match(self._VALID_URL, url)
 344         playlist_id = mobj.group('id')
 345         webpage = self._download_webpage(url, playlist_id)
 346
 347         return {
 348             '_type': 'playlist',
 349             'id': playlist_id,
 350             'title': self._og_search_title(webpage),
 351             'entries': self._extract_entries(playlist_id),
 352         }
 353
 354
 355 class DailymotionUserIE(DailymotionPlaylistIE):
 356     IE_NAME = 'dailymotion:user'
 357     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 358     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 359     _TESTS = [{
 360         'url': 'https://www.dailymotion.com/user/nqtv',
 361         'info_dict': {
 362             'id': 'nqtv',
 363             'title': 'Rémi Gaillard',
 364         },
 365         'playlist_mincount': 100,
 366     }, {
 367         'url': 'http://www.dailymotion.com/user/UnderProject',
 368         'info_dict': {
 369             'id': 'UnderProject',
 370             'title': 'UnderProject',
 371         },
 372         'playlist_mincount': 1800,
 373         'expected_warnings': [
 374             'Stopped at duplicated page',
 375         ],
 376         'skip': 'Takes too long time',
 377     }]
 378
 379     def _real_extract(self, url):
 380         mobj = re.match(self._VALID_URL, url)
 381         user = mobj.group('user')
 382         webpage = self._download_webpage(
 383             'https://www.dailymotion.com/user/%s' % user, user)
 384         full_user = unescapeHTML(self._html_search_regex(
 385             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 386             webpage, 'user'))
 387
 388         return {
 389             '_type': 'playlist',
 390             'id': user,
 391             'title': full_user,
 392             'entries': self._extract_entries(user),
 393         }
 394
 395
 396 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 397     _VALID_URL_PREFIX = r'https?://api\.dmcloud\.net/(?:player/)?embed/'
 398     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 399     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 400
 401     _TESTS = [{
 402         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 403         # Tested at FranceTvInfo_2
 404         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 405         'only_matching': True,
 406     }, {
 407         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 408         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 409         'only_matching': True,
 410     }]
 411
 412     @classmethod
 413     def _extract_dmcloud_url(cls, webpage):
 414         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL, webpage)
 415         if mobj:
 416             return mobj.group(1)
 417
 418         mobj = re.search(
 419             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL,
 420             webpage)
 421         if mobj:
 422             return mobj.group(1)
 423
 424     def _real_extract(self, url):
 425         video_id = self._match_id(url)
 426
 427         webpage = self._download_webpage_no_ff(url, video_id)
 428
 429         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 430
 431         video_info = self._parse_json(self._search_regex(
 432             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 433
 434         # TODO: parse ios_url, which is in fact a manifest
 435         video_url = video_info['mp4_url']
 436
 437         return {
 438             'id': video_id,
 439             'url': video_url,
 440             'title': title,
 441             'thumbnail': video_info.get('thumbnail_url'),
 442         }