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