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|swf)/(?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         'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
  54         'md5': '074b95bdee76b9e3654137aee9c79dfe',
  55         'info_dict': {
  56             'id': 'x5kesuj',
  57             'ext': 'mp4',
  58             'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
  59             'description': 'Office Christmas Party Review -  Jason Bateman, Olivia Munn, T.J. Miller',
  60             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
  61             'duration': 187,
  62             'timestamp': 1493651285,
  63             'upload_date': '20170501',
  64             'uploader': 'Deadline',
  65             'uploader_id': 'x1xm8ri',
  66             'age_limit': 0,
  67             'view_count': int,
  68         },
  69     }, {
  70         'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  71         'md5': '2137c41a8e78554bb09225b8eb322406',
  72         'info_dict': {
  73             'id': 'x2iuewm',
  74             'ext': 'mp4',
  75             'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  76             'description': 'Several come bundled with the Steam Controller.',
  77             'thumbnail': r're:^https?:.*\.(?:jpg|png)$',
  78             'duration': 74,
  79             'timestamp': 1425657362,
  80             'upload_date': '20150306',
  81             'uploader': 'IGN',
  82             'uploader_id': 'xijv66',
  83             'age_limit': 0,
  84             'view_count': int,
  85         },
  86         'skip': 'video gone',
  87     }, {
  88         # Vevo video
  89         'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  90         'info_dict': {
  91             'title': 'Roar (Official)',
  92             'id': 'USUV71301934',
  93             'ext': 'mp4',
  94             'uploader': 'Katy Perry',
  95             'upload_date': '20130905',
  96         },
  97         'params': {
  98             'skip_download': True,
  99         },
 100         'skip': 'VEVO is only available in some countries',
 101     }, {
 102         # age-restricted video
 103         'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
 104         'md5': '0d667a7b9cebecc3c89ee93099c4159d',
 105         'info_dict': {
 106             'id': 'xyh2zz',
 107             'ext': 'mp4',
 108             'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
 109             'uploader': 'HotWaves1012',
 110             'age_limit': 18,
 111         },
 112         'skip': 'video gone',
 113     }, {
 114         # geo-restricted, player v5
 115         'url': 'http://www.dailymotion.com/video/xhza0o',
 116         'only_matching': True,
 117     }, {
 118         # with subtitles
 119         'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 120         'only_matching': True,
 121     }, {
 122         'url': 'http://www.dailymotion.com/swf/video/x3n92nf',
 123         'only_matching': True,
 124     }, {
 125         'url': 'http://www.dailymotion.com/swf/x3ss1m_funny-magic-trick-barry-and-stuart_fun',
 126         'only_matching': True,
 127     }]
 128
 129     @staticmethod
 130     def _extract_urls(webpage):
 131         # Look for embedded Dailymotion player
 132         matches = re.findall(
 133             r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
 134         return list(map(lambda m: unescapeHTML(m[1]), matches))
 135
 136     def _real_extract(self, url):
 137         video_id = self._match_id(url)
 138
 139         webpage = self._download_webpage_no_ff(
 140             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 141
 142         age_limit = self._rta_search(webpage)
 143
 144         description = self._og_search_description(webpage) or self._html_search_meta(
 145             'description', webpage, 'description')
 146
 147         view_count_str = self._search_regex(
 148             (r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:([\s\d,.]+)"',
 149              r'video_views_count[^>]+>\s+([\s\d\,.]+)'),
 150             webpage, 'view count', default=None)
 151         if view_count_str:
 152             view_count_str = re.sub(r'\s', '', view_count_str)
 153         view_count = str_to_int(view_count_str)
 154         comment_count = int_or_none(self._search_regex(
 155             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 156             webpage, 'comment count', default=None))
 157
 158         player_v5 = self._search_regex(
 159             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 160              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 161              r'buildPlayer\(({.+?})\);',
 162              r'var\s+config\s*=\s*({.+?});',
 163              # New layout regex (see https://github.com/rg3/youtube-dl/issues/13580)
 164              r'__PLAYER_CONFIG__\s*=\s*({.+?});'],
 165             webpage, 'player v5', default=None)
 166         if player_v5:
 167             player = self._parse_json(player_v5, video_id)
 168             metadata = player['metadata']
 169
 170             self._check_error(metadata)
 171
 172             formats = []
 173             for quality, media_list in metadata['qualities'].items():
 174                 for media in media_list:
 175                     media_url = media.get('url')
 176                     if not media_url:
 177                         continue
 178                     type_ = media.get('type')
 179                     if type_ == 'application/vnd.lumberjack.manifest':
 180                         continue
 181                     ext = mimetype2ext(type_) or determine_ext(media_url)
 182                     if ext == 'm3u8':
 183                         formats.extend(self._extract_m3u8_formats(
 184                             media_url, video_id, 'mp4', preference=-1,
 185                             m3u8_id='hls', fatal=False))
 186                     elif ext == 'f4m':
 187                         formats.extend(self._extract_f4m_formats(
 188                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 189                     else:
 190                         f = {
 191                             'url': media_url,
 192                             'format_id': 'http-%s' % quality,
 193                             'ext': ext,
 194                         }
 195                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 196                         if m:
 197                             f.update({
 198                                 'width': int(m.group('width')),
 199                                 'height': int(m.group('height')),
 200                             })
 201                         formats.append(f)
 202             self._sort_formats(formats)
 203
 204             title = metadata['title']
 205             duration = int_or_none(metadata.get('duration'))
 206             timestamp = int_or_none(metadata.get('created_time'))
 207             thumbnail = metadata.get('poster_url')
 208             uploader = metadata.get('owner', {}).get('screenname')
 209             uploader_id = metadata.get('owner', {}).get('id')
 210
 211             subtitles = {}
 212             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 213             if subtitles_data and isinstance(subtitles_data, dict):
 214                 for subtitle_lang, subtitle in subtitles_data.items():
 215                     subtitles[subtitle_lang] = [{
 216                         'ext': determine_ext(subtitle_url),
 217                         'url': subtitle_url,
 218                     } for subtitle_url in subtitle.get('urls', [])]
 219
 220             return {
 221                 'id': video_id,
 222                 'title': title,
 223                 'description': description,
 224                 'thumbnail': thumbnail,
 225                 'duration': duration,
 226                 'timestamp': timestamp,
 227                 'uploader': uploader,
 228                 'uploader_id': uploader_id,
 229                 'age_limit': age_limit,
 230                 'view_count': view_count,
 231                 'comment_count': comment_count,
 232                 'formats': formats,
 233                 'subtitles': subtitles,
 234             }
 235
 236         # vevo embed
 237         vevo_id = self._search_regex(
 238             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 239             webpage, 'vevo embed', default=None)
 240         if vevo_id:
 241             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 242
 243         # fallback old player
 244         embed_page = self._download_webpage_no_ff(
 245             'https://www.dailymotion.com/embed/video/%s' % video_id,
 246             video_id, 'Downloading embed page')
 247
 248         timestamp = parse_iso8601(self._html_search_meta(
 249             'video:release_date', webpage, 'upload date'))
 250
 251         info = self._parse_json(
 252             self._search_regex(
 253                 r'var info = ({.*?}),$', embed_page,
 254                 'video info', flags=re.MULTILINE),
 255             video_id)
 256
 257         self._check_error(info)
 258
 259         formats = []
 260         for (key, format_id) in self._FORMATS:
 261             video_url = info.get(key)
 262             if video_url is not None:
 263                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 264                 if m_size is not None:
 265                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 266                 else:
 267                     width, height = None, None
 268                 formats.append({
 269                     'url': video_url,
 270                     'ext': 'mp4',
 271                     'format_id': format_id,
 272                     'width': width,
 273                     'height': height,
 274                 })
 275         self._sort_formats(formats)
 276
 277         # subtitles
 278         video_subtitles = self.extract_subtitles(video_id, webpage)
 279
 280         title = self._og_search_title(webpage, default=None)
 281         if title is None:
 282             title = self._html_search_regex(
 283                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 284                 'title')
 285
 286         return {
 287             'id': video_id,
 288             'formats': formats,
 289             'uploader': info['owner.screenname'],
 290             'timestamp': timestamp,
 291             'title': title,
 292             'description': description,
 293             'subtitles': video_subtitles,
 294             'thumbnail': info['thumbnail_url'],
 295             'age_limit': age_limit,
 296             'view_count': view_count,
 297             'duration': info['duration']
 298         }
 299
 300     def _check_error(self, info):
 301         error = info.get('error')
 302         if info.get('error') is not None:
 303             title = error['title']
 304             # See https://developer.dailymotion.com/api#access-error
 305             if error.get('code') == 'DM007':
 306                 self.raise_geo_restricted(msg=title)
 307             raise ExtractorError(
 308                 '%s said: %s' % (self.IE_NAME, title), expected=True)
 309
 310     def _get_subtitles(self, video_id, webpage):
 311         try:
 312             sub_list = self._download_webpage(
 313                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 314                 video_id, note=False)
 315         except ExtractorError as err:
 316             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 317             return {}
 318         info = json.loads(sub_list)
 319         if (info['total'] > 0):
 320             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 321             return sub_lang_list
 322         self._downloader.report_warning('video doesn\'t have subtitles')
 323         return {}
 324
 325
 326 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 327     IE_NAME = 'dailymotion:playlist'
 328     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>[^/?#&]+)'
 329     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 330     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 331     _TESTS = [{
 332         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 333         'info_dict': {
 334             'title': 'SPORT',
 335             'id': 'xv4bw_nqtv_sport',
 336         },
 337         'playlist_mincount': 20,
 338     }]
 339
 340     def _extract_entries(self, id):
 341         video_ids = set()
 342         processed_urls = set()
 343         for pagenum in itertools.count(1):
 344             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 345             webpage, urlh = self._download_webpage_handle_no_ff(
 346                 page_url, id, 'Downloading page %s' % pagenum)
 347             if urlh.geturl() in processed_urls:
 348                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 349                     page_url, urlh.geturl()), id)
 350                 break
 351
 352             processed_urls.add(urlh.geturl())
 353
 354             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 355                 if video_id not in video_ids:
 356                     yield self.url_result(
 357                         'http://www.dailymotion.com/video/%s' % video_id,
 358                         DailymotionIE.ie_key(), video_id)
 359                     video_ids.add(video_id)
 360
 361             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 362                 break
 363
 364     def _real_extract(self, url):
 365         mobj = re.match(self._VALID_URL, url)
 366         playlist_id = mobj.group('id')
 367         webpage = self._download_webpage(url, playlist_id)
 368
 369         return {
 370             '_type': 'playlist',
 371             'id': playlist_id,
 372             'title': self._og_search_title(webpage),
 373             'entries': self._extract_entries(playlist_id),
 374         }
 375
 376
 377 class DailymotionUserIE(DailymotionPlaylistIE):
 378     IE_NAME = 'dailymotion:user'
 379     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 380     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 381     _TESTS = [{
 382         'url': 'https://www.dailymotion.com/user/nqtv',
 383         'info_dict': {
 384             'id': 'nqtv',
 385             'title': 'Rémi Gaillard',
 386         },
 387         'playlist_mincount': 100,
 388     }, {
 389         'url': 'http://www.dailymotion.com/user/UnderProject',
 390         'info_dict': {
 391             'id': 'UnderProject',
 392             'title': 'UnderProject',
 393         },
 394         'playlist_mincount': 1800,
 395         'expected_warnings': [
 396             'Stopped at duplicated page',
 397         ],
 398         'skip': 'Takes too long time',
 399     }]
 400
 401     def _real_extract(self, url):
 402         mobj = re.match(self._VALID_URL, url)
 403         user = mobj.group('user')
 404         webpage = self._download_webpage(
 405             'https://www.dailymotion.com/user/%s' % user, user)
 406         full_user = unescapeHTML(self._html_search_regex(
 407             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 408             webpage, 'user'))
 409
 410         return {
 411             '_type': 'playlist',
 412             'id': user,
 413             'title': full_user,
 414             'entries': self._extract_entries(user),
 415         }
 416
 417
 418 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 419     _VALID_URL_PREFIX = r'https?://api\.dmcloud\.net/(?:player/)?embed/'
 420     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 421     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 422
 423     _TESTS = [{
 424         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 425         # Tested at FranceTvInfo_2
 426         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 427         'only_matching': True,
 428     }, {
 429         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 430         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 431         'only_matching': True,
 432     }]
 433
 434     @classmethod
 435     def _extract_dmcloud_url(cls, webpage):
 436         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL, webpage)
 437         if mobj:
 438             return mobj.group(1)
 439
 440         mobj = re.search(
 441             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % cls._VALID_EMBED_URL,
 442             webpage)
 443         if mobj:
 444             return mobj.group(1)
 445
 446     def _real_extract(self, url):
 447         video_id = self._match_id(url)
 448
 449         webpage = self._download_webpage_no_ff(url, video_id)
 450
 451         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 452
 453         video_info = self._parse_json(self._search_regex(
 454             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 455
 456         # TODO: parse ios_url, which is in fact a manifest
 457         video_url = video_info['mp4_url']
 458
 459         return {
 460             'id': video_id,
 461             'url': video_url,
 462             'title': title,
 463             'thumbnail': video_info.get('thumbnail_url'),
 464         }