Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/odnoklassniki.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_etree_fromstring,
   7     compat_parse_qs,
   8     compat_urllib_parse_unquote,
   9     compat_urllib_parse_urlparse,
  10 )
  11 from ..utils import (
  12     ExtractorError,
  13     unified_strdate,
  14     int_or_none,
  15     qualities,
  16     unescapeHTML,
  17 )
  18
  19
  20 class OdnoklassnikiIE(InfoExtractor):
  21     _VALID_URL = r'https?://(?:(?:www|m|mobile)\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
  22     _TESTS = [{
  23         # metadata in JSON
  24         'url': 'http://ok.ru/video/20079905452',
  25         'md5': '6ba728d85d60aa2e6dd37c9e70fdc6bc',
  26         'info_dict': {
  27             'id': '20079905452',
  28             'ext': 'mp4',
  29             'title': 'Культура меняет нас (прекрасный ролик!))',
  30             'duration': 100,
  31             'upload_date': '20141207',
  32             'uploader_id': '330537914540',
  33             'uploader': 'Виталий Добровольский',
  34             'like_count': int,
  35             'age_limit': 0,
  36         },
  37         'skip': 'Video has been blocked',
  38     }, {
  39         # metadataUrl
  40         'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
  41         'md5': '6ff470ea2dd51d5d18c295a355b0b6bc',
  42         'info_dict': {
  43             'id': '63567059965189-0',
  44             'ext': 'mp4',
  45             'title': 'Девушка без комплексов ...',
  46             'duration': 191,
  47             'upload_date': '20150518',
  48             'uploader_id': '534380003155',
  49             'uploader': '☭ Андрей Мещанинов ☭',
  50             'like_count': int,
  51             'age_limit': 0,
  52             'start_time': 5,
  53         },
  54     }, {
  55         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
  56         'url': 'http://ok.ru/video/64211978996595-1',
  57         'md5': '2f206894ffb5dbfcce2c5a14b909eea5',
  58         'info_dict': {
  59             'id': '64211978996595-1',
  60             'ext': 'mp4',
  61             'title': 'Космическая среда от 26 августа 2015',
  62             'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0',
  63             'duration': 440,
  64             'upload_date': '20150826',
  65             'uploader_id': 'tvroscosmos',
  66             'uploader': 'Телестудия Роскосмоса',
  67             'age_limit': 0,
  68         },
  69     }, {
  70         # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
  71         'url': 'http://ok.ru/video/62036049272859-0',
  72         'info_dict': {
  73             'id': '62036049272859-0',
  74             'ext': 'mp4',
  75             'title': 'МУЗЫКА     ДОЖДЯ .',
  76             'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
  77             'upload_date': '20120106',
  78             'uploader_id': '473534735899',
  79             'uploader': 'МARINA D',
  80             'age_limit': 0,
  81         },
  82         'params': {
  83             'skip_download': True,
  84         },
  85         'skip': 'Video has not been found',
  86     }, {
  87         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
  88         'only_matching': True,
  89     }, {
  90         'url': 'http://www.ok.ru/video/20648036891',
  91         'only_matching': True,
  92     }, {
  93         'url': 'http://www.ok.ru/videoembed/20648036891',
  94         'only_matching': True,
  95     }, {
  96         'url': 'http://m.ok.ru/video/20079905452',
  97         'only_matching': True,
  98     }, {
  99         'url': 'http://mobile.ok.ru/video/20079905452',
 100         'only_matching': True,
 101     }]
 102
 103     def _real_extract(self, url):
 104         start_time = int_or_none(compat_parse_qs(
 105             compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
 106
 107         video_id = self._match_id(url)
 108
 109         webpage = self._download_webpage(
 110             'http://ok.ru/video/%s' % video_id, video_id)
 111
 112         error = self._search_regex(
 113             r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<',
 114             webpage, 'error', default=None)
 115         if error:
 116             raise ExtractorError(error, expected=True)
 117
 118         player = self._parse_json(
 119             unescapeHTML(self._search_regex(
 120                 r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
 121                 webpage, 'player', group='player')),
 122             video_id)
 123
 124         flashvars = player['flashvars']
 125
 126         metadata = flashvars.get('metadata')
 127         if metadata:
 128             metadata = self._parse_json(metadata, video_id)
 129         else:
 130             metadata = self._download_json(
 131                 compat_urllib_parse_unquote(flashvars['metadataUrl']),
 132                 video_id, 'Downloading metadata JSON')
 133
 134         movie = metadata['movie']
 135
 136         # Some embedded videos may not contain title in movie dict (e.g.
 137         # http://ok.ru/video/62036049272859-0) thus we allow missing title
 138         # here and it's going to be extracted later by an extractor that
 139         # will process the actual embed.
 140         provider = metadata.get('provider')
 141         title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
 142
 143         thumbnail = movie.get('poster')
 144         duration = int_or_none(movie.get('duration'))
 145
 146         author = metadata.get('author', {})
 147         uploader_id = author.get('id')
 148         uploader = author.get('name')
 149
 150         upload_date = unified_strdate(self._html_search_meta(
 151             'ya:ovs:upload_date', webpage, 'upload date', default=None))
 152
 153         age_limit = None
 154         adult = self._html_search_meta(
 155             'ya:ovs:adult', webpage, 'age limit', default=None)
 156         if adult:
 157             age_limit = 18 if adult == 'true' else 0
 158
 159         like_count = int_or_none(metadata.get('likeCount'))
 160
 161         info = {
 162             'id': video_id,
 163             'title': title,
 164             'thumbnail': thumbnail,
 165             'duration': duration,
 166             'upload_date': upload_date,
 167             'uploader': uploader,
 168             'uploader_id': uploader_id,
 169             'like_count': like_count,
 170             'age_limit': age_limit,
 171             'start_time': start_time,
 172         }
 173
 174         if provider == 'USER_YOUTUBE':
 175             info.update({
 176                 '_type': 'url_transparent',
 177                 'url': movie['contentId'],
 178             })
 179             return info
 180
 181         quality = qualities(('4', '0', '1', '2', '3', '5'))
 182
 183         formats = [{
 184             'url': f['url'],
 185             'ext': 'mp4',
 186             'format_id': f['name'],
 187         } for f in metadata['videos']]
 188
 189         m3u8_url = metadata.get('hlsManifestUrl')
 190         if m3u8_url:
 191             formats.extend(self._extract_m3u8_formats(
 192                 m3u8_url, video_id, 'mp4', 'm3u8_native',
 193                 m3u8_id='hls', fatal=False))
 194
 195         dash_manifest = metadata.get('metadataEmbedded')
 196         if dash_manifest:
 197             formats.extend(self._parse_mpd_formats(
 198                 compat_etree_fromstring(dash_manifest), 'mpd'))
 199
 200         for fmt in formats:
 201             fmt_type = self._search_regex(
 202                 r'\btype[/=](\d)', fmt['url'],
 203                 'format type', default=None)
 204             if fmt_type:
 205                 fmt['quality'] = quality(fmt_type)
 206
 207         self._sort_formats(formats)
 208
 209         info['formats'] = formats
 210         return info