_ Git - youtube-dl/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 ..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                     ext = determine_ext(media_url)
 145                     if type_ == 'application/x-mpegURL' or ext == 'm3u8':
 146                         m3u8_formats = self._extract_m3u8_formats(
 147                             media_url, video_id, 'mp4', m3u8_id='hls', fatal=False)
 148                         if m3u8_formats:
 149                             formats.extend(m3u8_formats)
 150                     elif type_ == 'application/f4m' or ext == 'f4m':
 151                         f4m_formats = self._extract_f4m_formats(
 152                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False)
 153                         if f4m_formats:
 154                             formats.extend(f4m_formats)
 155                     else:
 156                         f = {
 157                             'url': media_url,
 158                             'format_id': quality,
 159                         }
 160                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
 161                         if m:
 162                             f.update({
 163                                 'width': int(m.group('width')),
 164                                 'height': int(m.group('height')),
 165                             })
 166                         formats.append(f)
 167             self._sort_formats(formats)
 168
 169             title = metadata['title']
 170             duration = int_or_none(metadata.get('duration'))
 171             timestamp = int_or_none(metadata.get('created_time'))
 172             thumbnail = metadata.get('poster_url')
 173             uploader = metadata.get('owner', {}).get('screenname')
 174             uploader_id = metadata.get('owner', {}).get('id')
 175
 176             subtitles = {}
 177             for subtitle_lang, subtitle in metadata.get('subtitles', {}).get('data', {}).items():
 178                 subtitles[subtitle_lang] = [{
 179                     'ext': determine_ext(subtitle_url),
 180                     'url': subtitle_url,
 181                 } for subtitle_url in subtitle.get('urls', [])]
 182
 183             return {
 184                 'id': video_id,
 185                 'title': title,
 186                 'description': description,
 187                 'thumbnail': thumbnail,
 188                 'duration': duration,
 189                 'timestamp': timestamp,
 190                 'uploader': uploader,
 191                 'uploader_id': uploader_id,
 192                 'age_limit': age_limit,
 193                 'view_count': view_count,
 194                 'comment_count': comment_count,
 195                 'formats': formats,
 196                 'subtitles': subtitles,
 197             }
 198
 199         # vevo embed
 200         vevo_id = self._search_regex(
 201             r'<link rel="video_src" href="[^"]*?vevo.com[^"]*?video=(?P<id>[\w]*)',
 202             webpage, 'vevo embed', default=None)
 203         if vevo_id:
 204             return self.url_result('vevo:%s' % vevo_id, 'Vevo')
 205
 206         # fallback old player
 207         embed_page = self._download_webpage_no_ff(
 208             'https://www.dailymotion.com/embed/video/%s' % video_id,
 209             video_id, 'Downloading embed page')
 210
 211         timestamp = parse_iso8601(self._html_search_meta(
 212             'video:release_date', webpage, 'upload date'))
 213
 214         info = self._parse_json(
 215             self._search_regex(
 216                 r'var info = ({.*?}),$', embed_page,
 217                 'video info', flags=re.MULTILINE),
 218             video_id)
 219
 220         self._check_error(info)
 221
 222         formats = []
 223         for (key, format_id) in self._FORMATS:
 224             video_url = info.get(key)
 225             if video_url is not None:
 226                 m_size = re.search(r'H264-(\d+)x(\d+)', video_url)
 227                 if m_size is not None:
 228                     width, height = map(int_or_none, (m_size.group(1), m_size.group(2)))
 229                 else:
 230                     width, height = None, None
 231                 formats.append({
 232                     'url': video_url,
 233                     'ext': 'mp4',
 234                     'format_id': format_id,
 235                     'width': width,
 236                     'height': height,
 237                 })
 238         self._sort_formats(formats)
 239
 240         # subtitles
 241         video_subtitles = self.extract_subtitles(video_id, webpage)
 242
 243         title = self._og_search_title(webpage, default=None)
 244         if title is None:
 245             title = self._html_search_regex(
 246                 r'(?s)<span\s+id="video_title"[^>]*>(.*?)</span>', webpage,
 247                 'title')
 248
 249         return {
 250             'id': video_id,
 251             'formats': formats,
 252             'uploader': info['owner.screenname'],
 253             'timestamp': timestamp,
 254             'title': title,
 255             'description': description,
 256             'subtitles': video_subtitles,
 257             'thumbnail': info['thumbnail_url'],
 258             'age_limit': age_limit,
 259             'view_count': view_count,
 260             'duration': info['duration']
 261         }
 262
 263     def _check_error(self, info):
 264         if info.get('error') is not None:
 265             raise ExtractorError(
 266                 '%s said: %s' % (self.IE_NAME, info['error']['title']), expected=True)
 267
 268     def _get_subtitles(self, video_id, webpage):
 269         try:
 270             sub_list = self._download_webpage(
 271                 'https://api.dailymotion.com/video/%s/subtitles?fields=id,language,url' % video_id,
 272                 video_id, note=False)
 273         except ExtractorError as err:
 274             self._downloader.report_warning('unable to download video subtitles: %s' % compat_str(err))
 275             return {}
 276         info = json.loads(sub_list)
 277         if (info['total'] > 0):
 278             sub_lang_list = dict((l['language'], [{'url': l['url'], 'ext': 'srt'}]) for l in info['list'])
 279             return sub_lang_list
 280         self._downloader.report_warning('video doesn\'t have subtitles')
 281         return {}
 282
 283
 284 class DailymotionPlaylistIE(DailymotionBaseInfoExtractor):
 285     IE_NAME = 'dailymotion:playlist'
 286     _VALID_URL = r'(?:https?://)?(?:www\.)?dailymotion\.[a-z]{2,3}/playlist/(?P<id>.+?)/'
 287     _MORE_PAGES_INDICATOR = r'(?s)<div class="pages[^"]*">.*?<a\s+class="[^"]*?icon-arrow_right[^"]*?"'
 288     _PAGE_TEMPLATE = 'https://www.dailymotion.com/playlist/%s/%s'
 289     _TESTS = [{
 290         'url': 'http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q',
 291         'info_dict': {
 292             'title': 'SPORT',
 293             'id': 'xv4bw_nqtv_sport',
 294         },
 295         'playlist_mincount': 20,
 296     }]
 297
 298     def _extract_entries(self, id):
 299         video_ids = set()
 300         processed_urls = set()
 301         for pagenum in itertools.count(1):
 302             page_url = self._PAGE_TEMPLATE % (id, pagenum)
 303             webpage, urlh = self._download_webpage_handle_no_ff(
 304                 page_url, id, 'Downloading page %s' % pagenum)
 305             if urlh.geturl() in processed_urls:
 306                 self.report_warning('Stopped at duplicated page %s, which is the same as %s' % (
 307                     page_url, urlh.geturl()), id)
 308                 break
 309
 310             processed_urls.add(urlh.geturl())
 311
 312             for video_id in re.findall(r'data-xid="(.+?)"', webpage):
 313                 if video_id not in video_ids:
 314                     yield self.url_result('http://www.dailymotion.com/video/%s' % video_id, 'Dailymotion')
 315                     video_ids.add(video_id)
 316
 317             if re.search(self._MORE_PAGES_INDICATOR, webpage) is None:
 318                 break
 319
 320     def _real_extract(self, url):
 321         mobj = re.match(self._VALID_URL, url)
 322         playlist_id = mobj.group('id')
 323         webpage = self._download_webpage(url, playlist_id)
 324
 325         return {
 326             '_type': 'playlist',
 327             'id': playlist_id,
 328             'title': self._og_search_title(webpage),
 329             'entries': self._extract_entries(playlist_id),
 330         }
 331
 332
 333 class DailymotionUserIE(DailymotionPlaylistIE):
 334     IE_NAME = 'dailymotion:user'
 335     _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|#|video|playlist)/)(?:(?:old/)?user/)?(?P<user>[^/]+)'
 336     _PAGE_TEMPLATE = 'http://www.dailymotion.com/user/%s/%s'
 337     _TESTS = [{
 338         'url': 'https://www.dailymotion.com/user/nqtv',
 339         'info_dict': {
 340             'id': 'nqtv',
 341             'title': 'Rémi Gaillard',
 342         },
 343         'playlist_mincount': 100,
 344     }, {
 345         'url': 'http://www.dailymotion.com/user/UnderProject',
 346         'info_dict': {
 347             'id': 'UnderProject',
 348             'title': 'UnderProject',
 349         },
 350         'playlist_mincount': 1800,
 351         'expected_warnings': [
 352             'Stopped at duplicated page',
 353         ],
 354         'skip': 'Takes too long time',
 355     }]
 356
 357     def _real_extract(self, url):
 358         mobj = re.match(self._VALID_URL, url)
 359         user = mobj.group('user')
 360         webpage = self._download_webpage(
 361             'https://www.dailymotion.com/user/%s' % user, user)
 362         full_user = unescapeHTML(self._html_search_regex(
 363             r'<a class="nav-image" title="([^"]+)" href="/%s">' % re.escape(user),
 364             webpage, 'user'))
 365
 366         return {
 367             '_type': 'playlist',
 368             'id': user,
 369             'title': full_user,
 370             'entries': self._extract_entries(user),
 371         }
 372
 373
 374 class DailymotionCloudIE(DailymotionBaseInfoExtractor):
 375     _VALID_URL_PREFIX = r'http://api\.dmcloud\.net/(?:player/)?embed/'
 376     _VALID_URL = r'%s[^/]+/(?P<id>[^/?]+)' % _VALID_URL_PREFIX
 377     _VALID_EMBED_URL = r'%s[^/]+/[^\'"]+' % _VALID_URL_PREFIX
 378
 379     _TESTS = [{
 380         # From http://www.francetvinfo.fr/economie/entreprises/les-entreprises-familiales-le-secret-de-la-reussite_933271.html
 381         # Tested at FranceTvInfo_2
 382         'url': 'http://api.dmcloud.net/embed/4e7343f894a6f677b10006b4/556e03339473995ee145930c?auth=1464865870-0-jyhsm84b-ead4c701fb750cf9367bf4447167a3db&autoplay=1',
 383         'only_matching': True,
 384     }, {
 385         # http://www.francetvinfo.fr/societe/larguez-les-amarres-le-cobaturage-se-developpe_980101.html
 386         'url': 'http://api.dmcloud.net/player/embed/4e7343f894a6f677b10006b4/559545469473996d31429f06?auth=1467430263-0-90tglw2l-a3a4b64ed41efe48d7fccad85b8b8fda&autoplay=1',
 387         'only_matching': True,
 388     }]
 389
 390     @classmethod
 391     def _extract_dmcloud_url(self, webpage):
 392         mobj = re.search(r'<iframe[^>]+src=[\'"](%s)[\'"]' % self._VALID_EMBED_URL, webpage)
 393         if mobj:
 394             return mobj.group(1)
 395
 396         mobj = re.search(
 397             r'<input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=[\'"](%s)[\'"]' % self._VALID_EMBED_URL,
 398             webpage)
 399         if mobj:
 400             return mobj.group(1)
 401
 402     def _real_extract(self, url):
 403         video_id = self._match_id(url)
 404
 405         webpage = self._download_webpage_no_ff(url, video_id)
 406
 407         title = self._html_search_regex(r'<title>([^>]+)</title>', webpage, 'title')
 408
 409         video_info = self._parse_json(self._search_regex(
 410             r'var\s+info\s*=\s*([^;]+);', webpage, 'video info'), video_id)
 411
 412         # TODO: parse ios_url, which is in fact a manifest
 413         video_url = video_info['mp4_url']
 414
 415         return {
 416             'id': video_id,
 417             'url': video_url,
 418             'title': title,
 419             'thumbnail': video_info.get('thumbnail_url'),
 420         }