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     parse_iso8601,
  19     str_to_int,
  20     unescapeHTML,
  21 )
  22
  23
  24 class DailymotionBaseInfoExtractor(InfoExtractor):
  25     @staticmethod
  26     def _build_request(url):
  27         """Build a request with the family filter disabled"""
  28         request = compat_urllib_request.Request(url)
  29         request.add_header('Cookie', 'family_filter=off; ff=off')
  30         return request
  31
  32     def _download_webpage_handle_no_ff(self, url, *args, **kwargs):
  33         request = self._build_request(url)
  34         return self._download_webpage_handle(request, *args, **kwargs)
  35
  36     def _download_webpage_no_ff(self, url, *args, **kwargs):
  37         request = self._build_request(url)
  38         return self._download_webpage(request, *args, **kwargs)
  39
  40
  41 class DailymotionIE(DailymotionBaseInfoExtractor):
  42     _VALID_URL = r'(?i)(?:https?://)?(?:(www|touch)\.)?dailymotion\.[a-z]{2,3}/(?:(embed|#)/)?video/(?P<id>[^/?_]+)'
  43     IE_NAME = 'dailymotion'
  44
  45     _FORMATS = [
  46         ('stream_h264_ld_url', 'ld'),
  47         ('stream_h264_url', 'standard'),
  48         ('stream_h264_hq_url', 'hq'),
  49         ('stream_h264_hd_url', 'hd'),
  50         ('stream_h264_hd1080_url', 'hd180'),
  51     ]
  52
  53     _TESTS = [
  54         {
  55             'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
  56             'md5': '2137c41a8e78554bb09225b8eb322406',
  57             'info_dict': {
  58                 'id': 'x2iuewm',
  59                 'ext': 'mp4',
  60                 'title': 'Steam Machine Models, Pricing Listed on Steam Store - IGN News',
  61                 'description': 'Several come bundled with the Steam Controller.',
  62                 'thumbnail': 're:^https?:.*\.(?:jpg|png)$',
  63                 'duration': 74,
  64                 'timestamp': 1425657362,
  65                 'upload_date': '20150306',
  66                 'uploader': 'IGN',
  67                 'uploader_id': 'xijv66',
  68                 'age_limit': 0,
  69                 'view_count': int,
  70                 'comment_count': int,
  71             }
  72         },
  73         # Vevo video
  74         {
  75             'url': 'http://www.dailymotion.com/video/x149uew_katy-perry-roar-official_musi',
  76             'info_dict': {
  77                 'title': 'Roar (Official)',
  78                 'id': 'USUV71301934',
  79                 'ext': 'mp4',
  80                 'uploader': 'Katy Perry',
  81                 'upload_date': '20130905',
  82             },
  83             'params': {
  84                 'skip_download': True,
  85             },
  86             'skip': 'VEVO is only available in some countries',
  87         },
  88         # age-restricted video
  89         {
  90             'url': 'http://www.dailymotion.com/video/xyh2zz_leanna-decker-cyber-girl-of-the-year-desires-nude-playboy-plus_redband',
  91             'md5': '0d667a7b9cebecc3c89ee93099c4159d',
  92             'info_dict': {
  93                 'id': 'xyh2zz',
  94                 'ext': 'mp4',
  95                 'title': 'Leanna Decker - Cyber Girl Of The Year Desires Nude [Playboy Plus]',
  96                 'uploader': 'HotWaves1012',
  97                 'age_limit': 18,
  98             }
  99         },
 100         # geo-restricted, player v5
 101         {
 102             'url': 'http://www.dailymotion.com/video/xhza0o',
 103             'only_matching': True,
 104         }
 105     ]
 106
 107     def _real_extract(self, url):
 108         video_id = self._match_id(url)
 109
 110         webpage = self._download_webpage_no_ff(
 111             'https://www.dailymotion.com/video/%s' % video_id, video_id)
 112
 113         age_limit = self._rta_search(webpage)
 114
 115         description = self._og_search_description(webpage) or self._html_search_meta(
 116             'description', webpage, 'description')
 117
 118         view_count = str_to_int(self._search_regex(
 119             [r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserPlays:(\d+)"',
 120              r'video_views_count[^>]+>\s+([\d\.,]+)'],
 121             webpage, 'view count', fatal=False))
 122         comment_count = int_or_none(self._search_regex(
 123             r'<meta[^>]+itemprop="interactionCount"[^>]+content="UserComments:(\d+)"',
 124             webpage, 'comment count', fatal=False))
 125
 126         player_v5 = self._search_regex(
 127             [r'buildPlayer\(({.+?})\);', r'playerV5\s*=\s*dmp\.create\([^,]+?,\s*({.+?})\);'],
 128             webpage, 'player v5', default=None)
 129         if player_v5:
 130             player = self._parse_json(player_v5, video_id)
 131             metadata = player['metadata']
 132
 133             self._check_error(metadata)
 134
 135             formats = []
 136             for quality, media_list in metadata['qualities'].items():
 137                 for media in media_list:
 138                     media_url = media.get('url')
 139                     if not media_url:
 140                         continue
 141                     type_ = media.get('type')
 142                     if type_ == 'application/vnd.lumberjack.manifest':
 143                         continue
 144                     if type_ == 'application/x-mpegURL' or determine_ext(media_url) == 'm3u8':
 145                         formats.extend(self._extract_m3u8_formats(
 146                             media_url, video_id, 'mp4', m3u8_id='hls'))
 147                     else:
 148                         f = {
 149                             'url': media_url,
 150                             'format_id': quality,
 151                         }
 152                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 153                         if m:
 154                             f.update({
 155                                 'width': int(m.group('width')),
 156                                 'height': int(m.group('height')),
 157                             })
 158                         formats.append(f)
 159             self._sort_formats(formats)
 160
 161             title = metadata['title']
 162             duration = int_or_none(metadata.get('duration'))
 163             timestamp = int_or_none(metadata.get('created_time'))
 164             thumbnail = metadata.get('poster_url')
 165             uploader = metadata.get('owner', {}).get('screenname')
 166             uploader_id = metadata.get('owner', {}).get('id')
 167
 168             subtitles = {}
 169             for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items():
 170                 subtitles[subtitle_lang] = [{
 171                     'ext': determine_ext(subtitle_url),
 172                     'url': subtitle_url,
 173                 } for subtitle_url in subtitle.get('urls', [])]
 174
 175             return {
 176                 'id': video_id,
 177                 'title': title,
 178                 'description': description,
 179                 'thumbnail': thumbnail,
 180                 'duration': duration,
 181                 'timestamp': timestamp,
 182                 'uploader': uploader,
 183                 'uploader_id': uploader_id,
 184                 'age_limit': age_limit,
 185                 'view_count': view_count,
 186                 'comment_count': comment_count,
 187                 'formats': formats,
 188                 'subtitles': subtitles,
 189             }
 190
 191         # vevo embed
 192         vevo_id = self._search_regex(
 193             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 194             webpage, 'vevo embed', default=None)
 195         if vevo_id:
 196             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 197
 198         # fallback old player
 199         embed_page = self._download_webpage_no_ff(
 200             'https://www.dailymotion.com/embed/video/%s' % video_id,
 201             video_id, 'Downloading embed page')
 202
 203         timestamp = parse_iso8601(self._html_search_meta(
 204             'video:release_date', webpage, 'upload date'))
 205
 206         info = self._parse_json(
 207             self._search_regex(
 208                 r'var info = ({.*?}),$', embed_page,
 209                 'video info', flags=re.MULTILINE),
 210             video_id)
 211
 212         self._check_error(info)
 213
 214         formats = []
 215         for (key, format_id) in self._FORMATS:
 216             video_url = info.get(key)
 217             if video_url is not None:
 218                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 219                 if m_size is not None:
 220                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 221                 else:
 222                     width, height = None, None
 223                 formats.append({
 224                     'url': video_url,
 225                     'ext': 'mp4',
 226                     'format_id': format_id,
 227                     'width': width,
 228                     'height': height,
 229                 })
 230         self._sort_formats(formats)
 231
 232         # subtitles
 233         video_subtitles = self.extract_subtitles(video_id, webpage)
 234
 235         title = self._og_search_title(webpage, default=None)
 236         if title is None:
 237             title = self._html_search_regex(
 238                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 239                 'title')
 240
 241         return {
 242             'id': video_id,
 243             'formats': formats,
 244             'uploader': info['owner.screenname'],
 245             'timestamp': timestamp,
 246             'title': title,
 247             'description': description,
 248             'subtitles': video_subtitles,
 249             'thumbnail': info['thumbnail_url'],
 250             'age_limit': age_limit,
 251             'view_count': view_count,
 252             'duration': info['duration']
 253         }
 254
 255     def _check_error(self, info):
 256         if info.get('error') is not None:
 257             msg = 'Couldn\'t get video, Dailymotion says: %s' % info['error']['title']
 258             raise ExtractorError(msg, expected=True)
 259
 260     def _get_subtitles(self, video_id, webpage):
 261         try:
 262             sub_list = self._download_webpage(
 263                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 264                 video_id, note=False)
 265         except ExtractorError as err:
 266             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 267             return {}
 268         info = json.loads(sub_list)
 269         if (info['total'] > 0):
 270             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 271             return sub_lang_list
 272         self._downloader.report_warning('video doesn\'t have subtitles')
 273         return {}
 274
 275
 276 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 277     IE_NAME = 'dailymotion:playlist'
 278     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 279     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 280     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 281     _TESTS = [{
 282         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 283         'info_dict': {
 284             'title': 'SPORT',
 285             'id': 'xv4bw_nqtv_sport',
 286         },
 287         'playlist_mincount': 20,
 288     }]
 289
 290     def _extract_entries(self, id):
 291         video_ids = set()
 292         processed_urls = set()
 293         for pagenum in itertools.count(1):
 294             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 295             webpage, urlh = self._download_webpage_handle_no_ff(
 296                 page_url, id, 'Downloading page %s' % pagenum)
 297             if urlh.geturl() in processed_urls:
 298                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 299                     page_url, urlh.geturl()), id)
 300                 break
 301
 302             processed_urls.add(urlh.geturl())
 303
 304             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 305                 if video_id not in video_ids:
 306                     yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 307                     video_ids.add(video_id)
 308
 309             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 310                 break
 311
 312     def _real_extract(self, url):
 313         mobj = re.match(self._VALID_URL, url)
 314         playlist_id = mobj.group('id')
 315         webpage = self._download_webpage(url, playlist_id)
 316
 317         return {
 318             '_type': 'playlist',
 319             'id': playlist_id,
 320             'title': self._og_search_title(webpage),
 321             'entries': self._extract_entries(playlist_id),
 322         }
 323
 324
 325 class DailymotionUserIE(DailymotionPlaylistIE):
 326     IE_NAME = 'dailymotion:user'
 327     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 328     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 329     _TESTS = [{
 330         'url': 'https://www.dailymotion.com/user/nqtv',
 331         'info_dict': {
 332             'id': 'nqtv',
 333             'title': 'Rémi Gaillard',
 334         },
 335         'playlist_mincount': 100,
 336     }, {
 337         'url': 'http://www.dailymotion.com/user/UnderProject',
 338         'info_dict': {
 339             'id': 'UnderProject',
 340             'title': 'UnderProject',
 341         },
 342         'playlist_mincount': 1800,
 343         'expected_warnings': [
 344             'Stopped at duplicated page',
 345         ],
 346         'skip': 'Takes too long time',
 347     }]
 348
 349     def _real_extract(self, url):
 350         mobj = re.match(self._VALID_URL, url)
 351         user = mobj.group('user')
 352         webpage = self._download_webpage(
 353             'https://www.dailymotion.com/user/%s' % user, user)
 354         full_user = unescapeHTML(self._html_search_regex(
 355             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 356             webpage, 'user'))
 357
 358         return {
 359             '_type': 'playlist',
 360             'id': user,
 361             'title': full_user,
 362             'entries': self._extract_entries(user),
 363         }
 364
 365
 366 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 367     _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
 368     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 369     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 370
 371     _TESTS = [{
 372         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 373         # Tested at FranceTvInfo_2
 374         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 375         'only_matching': True,
 376     }, {
 377         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 378         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 379         'only_matching': True,
 380     }]
 381
 382     @classmethod
 383     def _extract_dmcloud_url(self, webpage):
 384         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
 385         if mobj:
 386             return mobj.group(1)
 387
 388         mobj = re.search(
 389             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
 390             webpage)
 391         if mobj:
 392             return mobj.group(1)
 393
 394     def _real_extract(self, url):
 395         video_id = self._match_id(url)
 396
 397         webpage = self._download_webpage_no_ff(url, video_id)
 398
 399         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 400
 401         video_info = self._parse_json(self._search_regex(
 402             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 403
 404         # TODO: parse ios_url, which is in fact a manifest
 405         video_url = video_info['mp4_url']
 406
 407         return {
 408             'id': video_id,
 409             'url': video_url,
 410             'title': title,
 411             'thumbnail': video_info.get('thumbnail_url'),
 412         }