Raphaël G. Git Repositories - youtubedl/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': r'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             'skip': 'video gone',
  99         },
 100         # geo-restricted, player v5
 101         {
 102             'url': 'http://www.dailymotion.com/video/xhza0o',
 103             'only_matching': True,
 104         },
 105         # with subtitles
 106         {
 107             'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 108             'only_matching': True,
 109         },
 110         {
 111             'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 112             'only_matching': True,
 113         }
 114     ]
 115
 116     @staticmethod
 117     def _extract_urls(webpage):
 118         # Look for embedded Dailymotion player
 119         matches = re.findall(
 120             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
 121         return list(map(lambda m: unescapeHTML(m[1]), matches))
 122
 123     def _real_extract(self, url):
 124         video_id = self._match_id(url)
 125
 126         webpage = self._download_webpage_no_ff(
 127             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 128
 129         age_limit = self._rta_search(webpage)
 130
 131         description = self._og_search_description(webpage) or self._html_search_meta(
 132             'description', webpage, 'description')
 133
 134         view_count_str = self._search_regex(
 135             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
 136              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
 137             webpage, 'view count', fatal=False)
 138         if view_count_str:
 139             view_count_str = re.sub(r'\s', '', view_count_str)
 140         view_count = str_to_int(view_count_str)
 141         comment_count = int_or_none(self._search_regex(
 142             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 143             webpage, 'comment count', fatal=False))
 144
 145         player_v5 = self._search_regex(
 146             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 147              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 148              r'buildPlayer\(({.+?})\);',
 149              r'var\s+config\s*=\s*({.+?});'],
 150             webpage, 'player v5', default=None)
 151         if player_v5:
 152             player = self._parse_json(player_v5, video_id)
 153             metadata = player['metadata']
 154
 155             self._check_error(metadata)
 156
 157             formats = []
 158             for quality, media_list in metadata['qualities'].items():
 159                 for media in media_list:
 160                     media_url = media.get('url')
 161                     if not media_url:
 162                         continue
 163                     type_ = media.get('type')
 164                     if type_ == 'application/vnd.lumberjack.manifest':
 165                         continue
 166                     ext = mimetype2ext(type_) or determine_ext(media_url)
 167                     if ext == 'm3u8':
 168                         formats.extend(self._extract_m3u8_formats(
 169                             media_url, video_id, 'mp4', preference=-1,
 170                             m3u8_id='hls', fatal=False))
 171                     elif ext == 'f4m':
 172                         formats.extend(self._extract_f4m_formats(
 173                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 174                     else:
 175                         f = {
 176                             'url': media_url,
 177                             'format_id': 'http-%s' % quality,
 178                             'ext': ext,
 179                         }
 180                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 181                         if m:
 182                             f.update({
 183                                 'width': int(m.group('width')),
 184                                 'height': int(m.group('height')),
 185                             })
 186                         formats.append(f)
 187             self._sort_formats(formats)
 188
 189             title = metadata['title']
 190             duration = int_or_none(metadata.get('duration'))
 191             timestamp = int_or_none(metadata.get('created_time'))
 192             thumbnail = metadata.get('poster_url')
 193             uploader = metadata.get('owner', {}).get('screenname')
 194             uploader_id = metadata.get('owner', {}).get('id')
 195
 196             subtitles = {}
 197             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 198             if subtitles_data and isinstance(subtitles_data, dict):
 199                 for subtitle_lang, subtitle in subtitles_data.items():
 200                     subtitles[subtitle_lang] = [{
 201                         'ext': determine_ext(subtitle_url),
 202                         'url': subtitle_url,
 203                     } for subtitle_url in subtitle.get('urls', [])]
 204
 205             return {
 206                 'id': video_id,
 207                 'title': title,
 208                 'description': description,
 209                 'thumbnail': thumbnail,
 210                 'duration': duration,
 211                 'timestamp': timestamp,
 212                 'uploader': uploader,
 213                 'uploader_id': uploader_id,
 214                 'age_limit': age_limit,
 215                 'view_count': view_count,
 216                 'comment_count': comment_count,
 217                 'formats': formats,
 218                 'subtitles': subtitles,
 219             }
 220
 221         # vevo embed
 222         vevo_id = self._search_regex(
 223             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 224             webpage, 'vevo embed', default=None)
 225         if vevo_id:
 226             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 227
 228         # fallback old player
 229         embed_page = self._download_webpage_no_ff(
 230             'https://www.dailymotion.com/embed/video/%s' % video_id,
 231             video_id, 'Downloading embed page')
 232
 233         timestamp = parse_iso8601(self._html_search_meta(
 234             'video:release_date', webpage, 'upload date'))
 235
 236         info = self._parse_json(
 237             self._search_regex(
 238                 r'var info = ({.*?}),$', embed_page,
 239                 'video info', flags=re.MULTILINE),
 240             video_id)
 241
 242         self._check_error(info)
 243
 244         formats = []
 245         for (key, format_id) in self._FORMATS:
 246             video_url = info.get(key)
 247             if video_url is not None:
 248                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 249                 if m_size is not None:
 250                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 251                 else:
 252                     width, height = None, None
 253                 formats.append({
 254                     'url': video_url,
 255                     'ext': 'mp4',
 256                     'format_id': format_id,
 257                     'width': width,
 258                     'height': height,
 259                 })
 260         self._sort_formats(formats)
 261
 262         # subtitles
 263         video_subtitles = self.extract_subtitles(video_id, webpage)
 264
 265         title = self._og_search_title(webpage, default=None)
 266         if title is None:
 267             title = self._html_search_regex(
 268                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 269                 'title')
 270
 271         return {
 272             'id': video_id,
 273             'formats': formats,
 274             'uploader': info['owner.screenname'],
 275             'timestamp': timestamp,
 276             'title': title,
 277             'description': description,
 278             'subtitles': video_subtitles,
 279             'thumbnail': info['thumbnail_url'],
 280             'age_limit': age_limit,
 281             'view_count': view_count,
 282             'duration': info['duration']
 283         }
 284
 285     def _check_error(self, info):
 286         if info.get('error') is not None:
 287             raise ExtractorError(
 288                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 289
 290     def _get_subtitles(self, video_id, webpage):
 291         try:
 292             sub_list = self._download_webpage(
 293                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 294                 video_id, note=False)
 295         except ExtractorError as err:
 296             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 297             return {}
 298         info = json.loads(sub_list)
 299         if (info['total'] > 0):
 300             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 301             return sub_lang_list
 302         self._downloader.report_warning('video doesn\'t have subtitles')
 303         return {}
 304
 305
 306 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 307     IE_NAME = 'dailymotion:playlist'
 308     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 309     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 310     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 311     _TESTS = [{
 312         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 313         'info_dict': {
 314             'title': 'SPORT',
 315             'id': 'xv4bw_nqtv_sport',
 316         },
 317         'playlist_mincount': 20,
 318     }]
 319
 320     def _extract_entries(self, id):
 321         video_ids = set()
 322         processed_urls = set()
 323         for pagenum in itertools.count(1):
 324             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 325             webpage, urlh = self._download_webpage_handle_no_ff(
 326                 page_url, id, 'Downloading page %s' % pagenum)
 327             if urlh.geturl() in processed_urls:
 328                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 329                     page_url, urlh.geturl()), id)
 330                 break
 331
 332             processed_urls.add(urlh.geturl())
 333
 334             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 335                 if video_id not in video_ids:
 336                     yield self.url_result(
 337                         'http://www.dailymotion.com/video/%s' % video_id,
 338                         DailymotionIE.ie_key(), video_id)
 339                     video_ids.add(video_id)
 340
 341             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 342                 break
 343
 344     def _real_extract(self, url):
 345         mobj = re.match(self._VALID_URL, url)
 346         playlist_id = mobj.group('id')
 347         webpage = self._download_webpage(url, playlist_id)
 348
 349         return {
 350             '_type': 'playlist',
 351             'id': playlist_id,
 352             'title': self._og_search_title(webpage),
 353             'entries': self._extract_entries(playlist_id),
 354         }
 355
 356
 357 class DailymotionUserIE(DailymotionPlaylistIE):
 358     IE_NAME = 'dailymotion:user'
 359     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 360     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 361     _TESTS = [{
 362         'url': 'https://www.dailymotion.com/user/nqtv',
 363         'info_dict': {
 364             'id': 'nqtv',
 365             'title': 'Rémi Gaillard',
 366         },
 367         'playlist_mincount': 100,
 368     }, {
 369         'url': 'http://www.dailymotion.com/user/UnderProject',
 370         'info_dict': {
 371             'id': 'UnderProject',
 372             'title': 'UnderProject',
 373         },
 374         'playlist_mincount': 1800,
 375         'expected_warnings': [
 376             'Stopped at duplicated page',
 377         ],
 378         'skip': 'Takes too long time',
 379     }]
 380
 381     def _real_extract(self, url):
 382         mobj = re.match(self._VALID_URL, url)
 383         user = mobj.group('user')
 384         webpage = self._download_webpage(
 385             'https://www.dailymotion.com/user/%s' % user, user)
 386         full_user = unescapeHTML(self._html_search_regex(
 387             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 388             webpage, 'user'))
 389
 390         return {
 391             '_type': 'playlist',
 392             'id': user,
 393             'title': full_user,
 394             'entries': self._extract_entries(user),
 395         }
 396
 397
 398 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 399     _VALID_URL_PREFIX = r'https?://api\.dmcloud\.net/(?:player/)?embed/'
 400     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 401     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 402
 403     _TESTS = [{
 404         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 405         # Tested at FranceTvInfo_2
 406         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 407         'only_matching': True,
 408     }, {
 409         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 410         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 411         'only_matching': True,
 412     }]
 413
 414     @classmethod
 415     def _extract_dmcloud_url(cls, webpage):
 416         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL, webpage)
 417         if mobj:
 418             return mobj.group(1)
 419
 420         mobj = re.search(
 421             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL,
 422             webpage)
 423         if mobj:
 424             return mobj.group(1)
 425
 426     def _real_extract(self, url):
 427         video_id = self._match_id(url)
 428
 429         webpage = self._download_webpage_no_ff(url, video_id)
 430
 431         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 432
 433         video_info = self._parse_json(self._search_regex(
 434             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 435
 436         # TODO: parse ios_url, which is in fact a manifest
 437         video_url = video_info['mp4_url']
 438
 439         return {
 440             'id': video_id,
 441             'url': video_url,
 442             'title': title,
 443             'thumbnail': video_info.get('thumbnail_url'),
 444         }