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_handle_no_ff(self, url, *args, **kwargs):
  34         request = self._build_request(url)
  35         return self._download_webpage_handle(request, *args, **kwargs)
  36
  37     def _download_webpage_no_ff(self, url, *args, **kwargs):
  38         request = self._build_request(url)
  39         return self._download_webpage(request, *args, **kwargs)
  40
  41
  42 class DailymotionIE(DailymotionBaseInfoExtractor):
  43     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  44     IE_NAME = 'dailymotion'
  45
  46     _FORMATS = [
  47         ('stream_h264_ld_url', 'ld'),
  48         ('stream_h264_url', 'standard'),
  49         ('stream_h264_hq_url', 'hq'),
  50         ('stream_h264_hd_url', 'hd'),
  51         ('stream_h264_hd1080_url', 'hd180'),
  52     ]
  53
  54     _TESTS = [
  55         {
  56             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  57             'md5': '2137c41a8e78554bb09225b8eb322406',
  58             'info_dict': {
  59                 'id': 'x2iuewm',
  60                 'ext': 'mp4',
  61                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  62                 'description': 'Several come bundled with the Steam Controller.',
  63                 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
  64                 'duration': 74,
  65                 'timestamp': 1425657362,
  66                 'upload_date': '20150306',
  67                 'uploader': 'IGN',
  68                 'uploader_id': 'xijv66',
  69                 'age_limit': 0,
  70                 'view_count': int,
  71                 'comment_count': int,
  72             }
  73         },
  74         # Vevo video
  75         {
  76             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  77             'info_dict': {
  78                 'title': 'Roar (Official)',
  79                 'id': 'USUV71301934',
  80                 'ext': 'mp4',
  81                 'uploader': 'Katy Perry',
  82                 'upload_date': '20130905',
  83             },
  84             'params': {
  85                 'skip_download': True,
  86             },
  87             'skip': 'VEVO is only available in some countries',
  88         },
  89         # age-restricted video
  90         {
  91             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  92             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  93             'info_dict': {
  94                 'id': 'xyh2zz',
  95                 'ext': 'mp4',
  96                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  97                 'uploader': 'HotWaves1012',
  98                 'age_limit': 18,
  99             }
 100         }
 101     ]
 102
 103     def _real_extract(self, url):
 104         video_id = self._match_id(url)
 105
 106         webpage = self._download_webpage_no_ff(
 107             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 108
 109         age_limit = self._rta_search(webpage)
 110
 111         description = self._og_search_description(webpage) or self._html_search_meta(
 112             'description', webpage, 'description')
 113
 114         view_count = str_to_int(self._search_regex(
 115             [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
 116              r'video_views_count[^>]+>\s+([\d\.,]+)'],
 117             webpage, 'view count', fatal=False))
 118         comment_count = int_or_none(self._search_regex(
 119             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 120             webpage, 'comment count', fatal=False))
 121
 122         player_v5 = self._search_regex(
 123             r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 124             webpage, 'player v5', default=None)
 125         if player_v5:
 126             player = self._parse_json(player_v5, video_id)
 127             metadata = player['metadata']
 128             formats = []
 129             for quality, media_list in metadata['qualities'].items():
 130                 for media in media_list:
 131                     media_url = media.get('url')
 132                     if not media_url:
 133                         continue
 134                     type_ = media.get('type')
 135                     if type_ == 'application/vnd.lumberjack.manifest':
 136                         continue
 137                     if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
 138                         formats.extend(self._extract_m3u8_formats(
 139                             media_url, video_id, 'mp4', m3u8_id='hls'))
 140                     else:
 141                         f = {
 142                             'url': media_url,
 143                             'format_id': quality,
 144                         }
 145                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 146                         if m:
 147                             f.update({
 148                                 'width': int(m.group('width')),
 149                                 'height': int(m.group('height')),
 150                             })
 151                         formats.append(f)
 152             self._sort_formats(formats)
 153
 154             title = metadata['title']
 155             duration = int_or_none(metadata.get('duration'))
 156             timestamp = int_or_none(metadata.get('created_time'))
 157             thumbnail = metadata.get('poster_url')
 158             uploader = metadata.get('owner', {}).get('screenname')
 159             uploader_id = metadata.get('owner', {}).get('id')
 160
 161             subtitles = {}
 162             for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items():
 163                 subtitles[subtitle_lang] = [{
 164                     'ext': determine_ext(subtitle_url),
 165                     'url': subtitle_url,
 166                 } for subtitle_url in subtitle.get('urls', [])]
 167
 168             return {
 169                 'id': video_id,
 170                 'title': title,
 171                 'description': description,
 172                 'thumbnail': thumbnail,
 173                 'duration': duration,
 174                 'timestamp': timestamp,
 175                 'uploader': uploader,
 176                 'uploader_id': uploader_id,
 177                 'age_limit': age_limit,
 178                 'view_count': view_count,
 179                 'comment_count': comment_count,
 180                 'formats': formats,
 181                 'subtitles': subtitles,
 182             }
 183
 184         # vevo embed
 185         vevo_id = self._search_regex(
 186             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 187             webpage, 'vevo embed', default=None)
 188         if vevo_id:
 189             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 190
 191         # fallback old player
 192         embed_page = self._download_webpage_no_ff(
 193             'https://www.dailymotion.com/embed/video/%s' % video_id,
 194             video_id, 'Downloading embed page')
 195
 196         timestamp = parse_iso8601(self._html_search_meta(
 197             'video:release_date', webpage, 'upload date'))
 198
 199         info = self._parse_json(
 200             self._search_regex(
 201                 r'var info = ({.*?}),$', embed_page,
 202                 'video info', flags=re.MULTILINE),
 203             video_id)
 204
 205         if info.get('error') is not None:
 206             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 207             raise ExtractorError(msg, expected=True)
 208
 209         formats = []
 210         for (key, format_id) in self._FORMATS:
 211             video_url = info.get(key)
 212             if video_url is not None:
 213                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 214                 if m_size is not None:
 215                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 216                 else:
 217                     width, height = None, None
 218                 formats.append({
 219                     'url': video_url,
 220                     'ext': 'mp4',
 221                     'format_id': format_id,
 222                     'width': width,
 223                     'height': height,
 224                 })
 225         self._sort_formats(formats)
 226
 227         # subtitles
 228         video_subtitles = self.extract_subtitles(video_id, webpage)
 229
 230         title = self._og_search_title(webpage, default=None)
 231         if title is None:
 232             title = self._html_search_regex(
 233                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 234                 'title')
 235
 236         return {
 237             'id': video_id,
 238             'formats': formats,
 239             'uploader': info['owner.screenname'],
 240             'timestamp': timestamp,
 241             'title': title,
 242             'description': description,
 243             'subtitles': video_subtitles,
 244             'thumbnail': info['thumbnail_url'],
 245             'age_limit': age_limit,
 246             'view_count': view_count,
 247             'duration': info['duration']
 248         }
 249
 250     def _get_subtitles(self, video_id, webpage):
 251         try:
 252             sub_list = self._download_webpage(
 253                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 254                 video_id, note=False)
 255         except ExtractorError as err:
 256             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 257             return {}
 258         info = json.loads(sub_list)
 259         if (info['total'] > 0):
 260             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 261             return sub_lang_list
 262         self._downloader.report_warning('video doesn\'t have subtitles')
 263         return {}
 264
 265
 266 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 267     IE_NAME = 'dailymotion:playlist'
 268     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 269     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 270     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 271     _TESTS = [{
 272         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 273         'info_dict': {
 274             'title': 'SPORT',
 275             'id': 'xv4bw_nqtv_sport',
 276         },
 277         'playlist_mincount': 20,
 278     }]
 279
 280     def _extract_entries(self, id):
 281         video_ids = []
 282         processed_urls = set()
 283         for pagenum in itertools.count(1):
 284             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 285             webpage, urlh = self._download_webpage_handle_no_ff(
 286                 page_url, id, 'Downloading page %s' % pagenum)
 287             if urlh.geturl() in processed_urls:
 288                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 289                     page_url, urlh.geturl()), id)
 290                 break
 291
 292             processed_urls.add(urlh.geturl())
 293
 294             video_ids.extend(re.findall(r'data-xid="(.+?)"', webpage))
 295
 296             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 297                 break
 298         return [self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 299                 for video_id in orderedSet(video_ids)]
 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         }