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 )
  20
  21
  22 class DailymotionBaseInfoExtractor(InfoExtractor):
  23     @staticmethod
  24     def _build_request(url):
  25         """Build a request with the family filter disabled"""
  26         request = sanitized_Request(url)
  27         request.add_header('Cookie', 'family_filter=off; ff=off')
  28         return request
  29
  30     def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
  31         request = self._build_request(url)
  32         return self._download_webpage_handle(request, *args, **kwargs)
  33
  34     def _download_webpage_no_ff(self, url, *args, **kwargs):
  35         request = self._build_request(url)
  36         return self._download_webpage(request, *args, **kwargs)
  37
  38
  39 class DailymotionIE(DailymotionBaseInfoExtractor):
  40     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  41     IE_NAME = 'dailymotion'
  42
  43     _FORMATS = [
  44         ('stream_h264_ld_url', 'ld'),
  45         ('stream_h264_url', 'standard'),
  46         ('stream_h264_hq_url', 'hq'),
  47         ('stream_h264_hd_url', 'hd'),
  48         ('stream_h264_hd1080_url', 'hd180'),
  49     ]
  50
  51     _TESTS = [
  52         {
  53             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  54             'md5': '2137c41a8e78554bb09225b8eb322406',
  55             'info_dict': {
  56                 'id': 'x2iuewm',
  57                 'ext': 'mp4',
  58                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  59                 'description': 'Several come bundled with the Steam Controller.',
  60                 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
  61                 'duration': 74,
  62                 'timestamp': 1425657362,
  63                 'upload_date': '20150306',
  64                 'uploader': 'IGN',
  65                 'uploader_id': 'xijv66',
  66                 'age_limit': 0,
  67                 'view_count': int,
  68                 'comment_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         },
  98         # geo-restricted, player v5
  99         {
 100             'url': 'http://www.dailymotion.com/video/xhza0o',
 101             'only_matching': True,
 102         },
 103         # with subtitles
 104         {
 105             'url': 'http://www.dailymotion.com/video/x20su5f_the-power-of-nightmares-1-the-rise-of-the-politics-of-fear-bbc-2004_news',
 106             'only_matching': True,
 107         }
 108     ]
 109
 110     def _real_extract(self, url):
 111         video_id = self._match_id(url)
 112
 113         webpage = self._download_webpage_no_ff(
 114             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 115
 116         age_limit = self._rta_search(webpage)
 117
 118         description = self._og_search_description(webpage) or self._html_search_meta(
 119             'description', webpage, 'description')
 120
 121         view_count = str_to_int(self._search_regex(
 122             [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
 123              r'video_views_count[^>]+>\s+([\d\.,]+)'],
 124             webpage, 'view count', fatal=False))
 125         comment_count = int_or_none(self._search_regex(
 126             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 127             webpage, 'comment count', fatal=False))
 128
 129         player_v5 = self._search_regex(
 130             [r'buildPlayer\(({.+?})\);\n',  # See https://github.com/rg3/youtube-dl/issues/7826
 131              r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);',
 132              r'buildPlayer\(({.+?})\);'],
 133             webpage, 'player v5', default=None)
 134         if player_v5:
 135             player = self._parse_json(player_v5, video_id)
 136             metadata = player['metadata']
 137
 138             self._check_error(metadata)
 139
 140             formats = []
 141             for quality, media_list in metadata['qualities'].items():
 142                 for media in media_list:
 143                     media_url = media.get('url')
 144                     if not media_url:
 145                         continue
 146                     type_ = media.get('type')
 147                     if type_ == 'application/vnd.lumberjack.manifest':
 148                         continue
 149                     ext = determine_ext(media_url)
 150                     if type_ == 'application/x-mpegURL' or ext == 'm3u8':
 151                         formats.extend(self._extract_m3u8_formats(
 152                             media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
 153                     elif type_ == 'application/f4m' or ext == 'f4m':
 154                         formats.extend(self._extract_f4m_formats(
 155                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
 156                     else:
 157                         f = {
 158                             'url': media_url,
 159                             'format_id': quality,
 160                         }
 161                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 162                         if m:
 163                             f.update({
 164                                 'width': int(m.group('width')),
 165                                 'height': int(m.group('height')),
 166                             })
 167                         formats.append(f)
 168             self._sort_formats(formats)
 169
 170             title = metadata['title']
 171             duration = int_or_none(metadata.get('duration'))
 172             timestamp = int_or_none(metadata.get('created_time'))
 173             thumbnail = metadata.get('poster_url')
 174             uploader = metadata.get('owner', {}).get('screenname')
 175             uploader_id = metadata.get('owner', {}).get('id')
 176
 177             subtitles = {}
 178             subtitles_data = metadata.get('subtitles', {}).get('data', {})
 179             if subtitles_data and isinstance(subtitles_data, dict):
 180                 for subtitle_lang, subtitle in subtitles_data.items():
 181                     subtitles[subtitle_lang] = [{
 182                         'ext': determine_ext(subtitle_url),
 183                         'url': subtitle_url,
 184                     } for subtitle_url in subtitle.get('urls', [])]
 185
 186             return {
 187                 'id': video_id,
 188                 'title': title,
 189                 'description': description,
 190                 'thumbnail': thumbnail,
 191                 'duration': duration,
 192                 'timestamp': timestamp,
 193                 'uploader': uploader,
 194                 'uploader_id': uploader_id,
 195                 'age_limit': age_limit,
 196                 'view_count': view_count,
 197                 'comment_count': comment_count,
 198                 'formats': formats,
 199                 'subtitles': subtitles,
 200             }
 201
 202         # vevo embed
 203         vevo_id = self._search_regex(
 204             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 205             webpage, 'vevo embed', default=None)
 206         if vevo_id:
 207             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 208
 209         # fallback old player
 210         embed_page = self._download_webpage_no_ff(
 211             'https://www.dailymotion.com/embed/video/%s' % video_id,
 212             video_id, 'Downloading embed page')
 213
 214         timestamp = parse_iso8601(self._html_search_meta(
 215             'video:release_date', webpage, 'upload date'))
 216
 217         info = self._parse_json(
 218             self._search_regex(
 219                 r'var info = ({.*?}),$', embed_page,
 220                 'video info', flags=re.MULTILINE),
 221             video_id)
 222
 223         self._check_error(info)
 224
 225         formats = []
 226         for (key, format_id) in self._FORMATS:
 227             video_url = info.get(key)
 228             if video_url is not None:
 229                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 230                 if m_size is not None:
 231                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 232                 else:
 233                     width, height = None, None
 234                 formats.append({
 235                     'url': video_url,
 236                     'ext': 'mp4',
 237                     'format_id': format_id,
 238                     'width': width,
 239                     'height': height,
 240                 })
 241         self._sort_formats(formats)
 242
 243         # subtitles
 244         video_subtitles = self.extract_subtitles(video_id, webpage)
 245
 246         title = self._og_search_title(webpage, default=None)
 247         if title is None:
 248             title = self._html_search_regex(
 249                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 250                 'title')
 251
 252         return {
 253             'id': video_id,
 254             'formats': formats,
 255             'uploader': info['owner.screenname'],
 256             'timestamp': timestamp,
 257             'title': title,
 258             'description': description,
 259             'subtitles': video_subtitles,
 260             'thumbnail': info['thumbnail_url'],
 261             'age_limit': age_limit,
 262             'view_count': view_count,
 263             'duration': info['duration']
 264         }
 265
 266     def _check_error(self, info):
 267         if info.get('error') is not None:
 268             raise ExtractorError(
 269                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 270
 271     def _get_subtitles(self, video_id, webpage):
 272         try:
 273             sub_list = self._download_webpage(
 274                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 275                 video_id, note=False)
 276         except ExtractorError as err:
 277             self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err))
 278             return {}
 279         info = json.loads(sub_list)
 280         if (info['total'] > 0):
 281             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 282             return sub_lang_list
 283         self._downloader.report_warning('video doesn\'t have subtitles')
 284         return {}
 285
 286
 287 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 288     IE_NAME = 'dailymotion:playlist'
 289     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 290     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 291     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 292     _TESTS = [{
 293         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 294         'info_dict': {
 295             'title': 'SPORT',
 296             'id': 'xv4bw_nqtv_sport',
 297         },
 298         'playlist_mincount': 20,
 299     }]
 300
 301     def _extract_entries(self, id):
 302         video_ids = set()
 303         processed_urls = set()
 304         for pagenum in itertools.count(1):
 305             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 306             webpage, urlh = self._download_webpage_handle_no_ff(
 307                 page_url, id, 'Downloading page %s' % pagenum)
 308             if urlh.geturl() in processed_urls:
 309                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 310                     page_url, urlh.geturl()), id)
 311                 break
 312
 313             processed_urls.add(urlh.geturl())
 314
 315             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 316                 if video_id not in video_ids:
 317                     yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 318                     video_ids.add(video_id)
 319
 320             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 321                 break
 322
 323     def _real_extract(self, url):
 324         mobj = re.match(self._VALID_URL, url)
 325         playlist_id = mobj.group('id')
 326         webpage = self._download_webpage(url, playlist_id)
 327
 328         return {
 329             '_type': 'playlist',
 330             'id': playlist_id,
 331             'title': self._og_search_title(webpage),
 332             'entries': self._extract_entries(playlist_id),
 333         }
 334
 335
 336 class DailymotionUserIE(DailymotionPlaylistIE):
 337     IE_NAME = 'dailymotion:user'
 338     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 339     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 340     _TESTS = [{
 341         'url': 'https://www.dailymotion.com/user/nqtv',
 342         'info_dict': {
 343             'id': 'nqtv',
 344             'title': 'Rémi Gaillard',
 345         },
 346         'playlist_mincount': 100,
 347     }, {
 348         'url': 'http://www.dailymotion.com/user/UnderProject',
 349         'info_dict': {
 350             'id': 'UnderProject',
 351             'title': 'UnderProject',
 352         },
 353         'playlist_mincount': 1800,
 354         'expected_warnings': [
 355             'Stopped at duplicated page',
 356         ],
 357         'skip': 'Takes too long time',
 358     }]
 359
 360     def _real_extract(self, url):
 361         mobj = re.match(self._VALID_URL, url)
 362         user = mobj.group('user')
 363         webpage = self._download_webpage(
 364             'https://www.dailymotion.com/user/%s' % user, user)
 365         full_user = unescapeHTML(self._html_search_regex(
 366             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 367             webpage, 'user'))
 368
 369         return {
 370             '_type': 'playlist',
 371             'id': user,
 372             'title': full_user,
 373             'entries': self._extract_entries(user),
 374         }
 375
 376
 377 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 378     _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
 379     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 380     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 381
 382     _TESTS = [{
 383         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 384         # Tested at FranceTvInfo_2
 385         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 386         'only_matching': True,
 387     }, {
 388         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 389         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 390         'only_matching': True,
 391     }]
 392
 393     @classmethod
 394     def _extract_dmcloud_url(self, webpage):
 395         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
 396         if mobj:
 397             return mobj.group(1)
 398
 399         mobj = re.search(
 400             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
 401             webpage)
 402         if mobj:
 403             return mobj.group(1)
 404
 405     def _real_extract(self, url):
 406         video_id = self._match_id(url)
 407
 408         webpage = self._download_webpage_no_ff(url, video_id)
 409
 410         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 411
 412         video_info = self._parse_json(self._search_regex(
 413             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 414
 415         # TODO: parse ios_url, which is in fact a manifest
 416         video_url = video_info['mp4_url']
 417
 418         return {
 419             'id': video_id,
 420             'url': video_url,
 421             'title': title,
 422             'thumbnail': video_info.get('thumbnail_url'),
 423         }