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