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             }
  70         },
  71         # Vevo video
  72         {
  73             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  74             'info_dict': {
  75                 'title': 'Roar (Official)',
  76                 'id': 'USUV71301934',
  77                 'ext': 'mp4',
  78                 'uploader': 'Katy Perry',
  79                 'upload_date': '20130905',
  80             },
  81             'params': {
  82                 'skip_download': True,
  83             },
  84             'skip': 'VEVO is only available in some countries',
  85         },
  86         # age-restricted video
  87         {
  88             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  89             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  90             'info_dict': {
  91                 'id': 'xyh2zz',
  92                 'ext': 'mp4',
  93                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  94                 'uploader': 'HotWaves1012',
  95                 'age_limit': 18,
  96             },
  97             'skip': 'video gone',
  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', default=None))
 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              r'var\s+config\s*=\s*({.+?});'],
 149             webpage, 'player v5', default=None)
 150         if player_v5:
 151             player = self._parse_json(player_v5, video_id)
 152             metadata = player['metadata']
 153
 154             self._check_error(metadata)
 155
 156             formats = []
 157             for quality, media_list in metadata['qualities'].items():
 158                 for media in media_list:
 159                     media_url = media.get('url')
 160                     if not media_url:
 161                         continue
 162                     type_ = media.get('type')
 163                     if type_ == 'application/vnd.lumberjack.manifest':
 164                         continue
 165                     ext = mimetype2ext(type_) or determine_ext(media_url)
 166                     if ext == 'm3u8':
 167                         formats.extend(self._extract_m3u8_formats(
 168                             media_url, video_id, 'mp4', preference=-1,
 169                             m3u8_id='hls', fatal=False))
 170                     elif ext == 'f4m':
 171                         formats.extend(self._extract_f4m_formats(
 172                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 173                     else:
 174                         f = {
 175                             'url': media_url,
 176                             'format_id': 'http-%s' % quality,
 177                             'ext': ext,
 178                         }
 179                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 180                         if m:
 181                             f.update({
 182                                 'width': int(m.group('width')),
 183                                 'height': int(m.group('height')),
 184                             })
 185                         formats.append(f)
 186             self._sort_formats(formats)
 187
 188             title = metadata['title']
 189             duration = int_or_none(metadata.get('duration'))
 190             timestamp = int_or_none(metadata.get('created_time'))
 191             thumbnail = metadata.get('poster_url')
 192             uploader = metadata.get('owner', {}).get('screenname')
 193             uploader_id = metadata.get('owner', {}).get('id')
 194
 195             subtitles = {}
 196             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 197             if subtitles_data and isinstance(subtitles_data, dict):
 198                 for subtitle_lang, subtitle in subtitles_data.items():
 199                     subtitles[subtitle_lang] = [{
 200                         'ext': determine_ext(subtitle_url),
 201                         'url': subtitle_url,
 202                     } for subtitle_url in subtitle.get('urls', [])]
 203
 204             return {
 205                 'id': video_id,
 206                 'title': title,
 207                 'description': description,
 208                 'thumbnail': thumbnail,
 209                 'duration': duration,
 210                 'timestamp': timestamp,
 211                 'uploader': uploader,
 212                 'uploader_id': uploader_id,
 213                 'age_limit': age_limit,
 214                 'view_count': view_count,
 215                 'comment_count': comment_count,
 216                 'formats': formats,
 217                 'subtitles': subtitles,
 218             }
 219
 220         # vevo embed
 221         vevo_id = self._search_regex(
 222             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 223             webpage, 'vevo embed', default=None)
 224         if vevo_id:
 225             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 226
 227         # fallback old player
 228         embed_page = self._download_webpage_no_ff(
 229             'https://www.dailymotion.com/embed/video/%s' % video_id,
 230             video_id, 'Downloading embed page')
 231
 232         timestamp = parse_iso8601(self._html_search_meta(
 233             'video:release_date', webpage, 'upload date'))
 234
 235         info = self._parse_json(
 236             self._search_regex(
 237                 r'var info = ({.*?}),$', embed_page,
 238                 'video info', flags=re.MULTILINE),
 239             video_id)
 240
 241         self._check_error(info)
 242
 243         formats = []
 244         for (key, format_id) in self._FORMATS:
 245             video_url = info.get(key)
 246             if video_url is not None:
 247                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 248                 if m_size is not None:
 249                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 250                 else:
 251                     width, height = None, None
 252                 formats.append({
 253                     'url': video_url,
 254                     'ext': 'mp4',
 255                     'format_id': format_id,
 256                     'width': width,
 257                     'height': height,
 258                 })
 259         self._sort_formats(formats)
 260
 261         # subtitles
 262         video_subtitles = self.extract_subtitles(video_id, webpage)
 263
 264         title = self._og_search_title(webpage, default=None)
 265         if title is None:
 266             title = self._html_search_regex(
 267                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 268                 'title')
 269
 270         return {
 271             'id': video_id,
 272             'formats': formats,
 273             'uploader': info['owner.screenname'],
 274             'timestamp': timestamp,
 275             'title': title,
 276             'description': description,
 277             'subtitles': video_subtitles,
 278             'thumbnail': info['thumbnail_url'],
 279             'age_limit': age_limit,
 280             'view_count': view_count,
 281             'duration': info['duration']
 282         }
 283
 284     def _check_error(self, info):
 285         if info.get('error') is not None:
 286             raise ExtractorError(
 287                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 288
 289     def _get_subtitles(self, video_id, webpage):
 290         try:
 291             sub_list = self._download_webpage(
 292                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 293                 video_id, note=False)
 294         except ExtractorError as err:
 295             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 296             return {}
 297         info = json.loads(sub_list)
 298         if (info['total'] > 0):
 299             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 300             return sub_lang_list
 301         self._downloader.report_warning('video doesn\'t have subtitles')
 302         return {}
 303
 304
 305 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 306     IE_NAME = 'dailymotion:playlist'
 307     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 308     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 309     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 310     _TESTS = [{
 311         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 312         'info_dict': {
 313             'title': 'SPORT',
 314             'id': 'xv4bw_nqtv_sport',
 315         },
 316         'playlist_mincount': 20,
 317     }]
 318
 319     def _extract_entries(self, id):
 320         video_ids = set()
 321         processed_urls = set()
 322         for pagenum in itertools.count(1):
 323             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 324             webpage, urlh = self._download_webpage_handle_no_ff(
 325                 page_url, id, 'Downloading page %s' % pagenum)
 326             if urlh.geturl() in processed_urls:
 327                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 328                     page_url, urlh.geturl()), id)
 329                 break
 330
 331             processed_urls.add(urlh.geturl())
 332
 333             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 334                 if video_id not in video_ids:
 335                     yield self.url_result(
 336                         'http://www.dailymotion.com/video/%s' % video_id,
 337                         DailymotionIE.ie_key(), video_id)
 338                     video_ids.add(video_id)
 339
 340             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 341                 break
 342
 343     def _real_extract(self, url):
 344         mobj = re.match(self._VALID_URL, url)
 345         playlist_id = mobj.group('id')
 346         webpage = self._download_webpage(url, playlist_id)
 347
 348         return {
 349             '_type': 'playlist',
 350             'id': playlist_id,
 351             'title': self._og_search_title(webpage),
 352             'entries': self._extract_entries(playlist_id),
 353         }
 354
 355
 356 class DailymotionUserIE(DailymotionPlaylistIE):
 357     IE_NAME = 'dailymotion:user'
 358     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 359     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 360     _TESTS = [{
 361         'url': 'https://www.dailymotion.com/user/nqtv',
 362         'info_dict': {
 363             'id': 'nqtv',
 364             'title': 'Rémi Gaillard',
 365         },
 366         'playlist_mincount': 100,
 367     }, {
 368         'url': 'http://www.dailymotion.com/user/UnderProject',
 369         'info_dict': {
 370             'id': 'UnderProject',
 371             'title': 'UnderProject',
 372         },
 373         'playlist_mincount': 1800,
 374         'expected_warnings': [
 375             'Stopped at duplicated page',
 376         ],
 377         'skip': 'Takes too long time',
 378     }]
 379
 380     def _real_extract(self, url):
 381         mobj = re.match(self._VALID_URL, url)
 382         user = mobj.group('user')
 383         webpage = self._download_webpage(
 384             'https://www.dailymotion.com/user/%s' % user, user)
 385         full_user = unescapeHTML(self._html_search_regex(
 386             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 387             webpage, 'user'))
 388
 389         return {
 390             '_type': 'playlist',
 391             'id': user,
 392             'title': full_user,
 393             'entries': self._extract_entries(user),
 394         }
 395
 396
 397 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 398     _VALID_URL_PREFIX = r'https?://api\.dmcloud\.net/(?:player/)?embed/'
 399     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 400     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 401
 402     _TESTS = [{
 403         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 404         # Tested at FranceTvInfo_2
 405         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 406         'only_matching': True,
 407     }, {
 408         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 409         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 410         'only_matching': True,
 411     }]
 412
 413     @classmethod
 414     def _extract_dmcloud_url(cls, webpage):
 415         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL, webpage)
 416         if mobj:
 417             return mobj.group(1)
 418
 419         mobj = re.search(
 420             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL,
 421             webpage)
 422         if mobj:
 423             return mobj.group(1)
 424
 425     def _real_extract(self, url):
 426         video_id = self._match_id(url)
 427
 428         webpage = self._download_webpage_no_ff(url, video_id)
 429
 430         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 431
 432         video_info = self._parse_json(self._search_regex(
 433             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 434
 435         # TODO: parse ios_url, which is in fact a manifest
 436         video_url = video_info['mp4_url']
 437
 438         return {
 439             'id': video_id,
 440             'url': video_url,
 441             'title': title,
 442             'thumbnail': video_info.get('thumbnail_url'),
 443         }