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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_etree_fromstring,
   9     compat_parse_qs,
  10     compat_urllib_parse_unquote,
  11     compat_urllib_parse_urlparse,
  12 )
  13 from ..utils import (
  14     ExtractorError,
  15     unified_strdate,
  16     int_or_none,
  17     qualities,
  18     unescapeHTML,
  19     urlencode_postdata,
  20 )
  21
  22
  23 class OdnoklassnikiIE(InfoExtractor):
  24     _VALID_URL = r'''(?x)
  25                 https?://
  26                     (?:(?:www|m|mobile)\.)?
  27                     (?:odnoklassniki|ok)\.ru/
  28                     (?:
  29                         video(?:embed)?/|
  30                         web-api/video/moviePlayer/|
  31                         live/|
  32                         dk\?.*?st\.mvId=
  33                     )
  34                     (?P<id>[\d-]+)
  35                 '''
  36     _TESTS = [{
  37         # metadata in JSON
  38         'url': 'http://ok.ru/video/20079905452',
  39         'md5': '0b62089b479e06681abaaca9d204f152',
  40         'info_dict': {
  41             'id': '20079905452',
  42             'ext': 'mp4',
  43             'title': 'Культура меняет нас (прекрасный ролик!))',
  44             'duration': 100,
  45             'upload_date': '20141207',
  46             'uploader_id': '330537914540',
  47             'uploader': 'Виталий Добровольский',
  48             'like_count': int,
  49             'age_limit': 0,
  50         },
  51     }, {
  52         # metadataUrl
  53         'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
  54         'md5': '6ff470ea2dd51d5d18c295a355b0b6bc',
  55         'info_dict': {
  56             'id': '63567059965189-0',
  57             'ext': 'mp4',
  58             'title': 'Девушка без комплексов ...',
  59             'duration': 191,
  60             'upload_date': '20150518',
  61             'uploader_id': '534380003155',
  62             'uploader': '☭ Андрей Мещанинов ☭',
  63             'like_count': int,
  64             'age_limit': 0,
  65             'start_time': 5,
  66         },
  67     }, {
  68         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
  69         'url': 'http://ok.ru/video/64211978996595-1',
  70         'md5': '2f206894ffb5dbfcce2c5a14b909eea5',
  71         'info_dict': {
  72             'id': 'V_VztHT5BzY',
  73             'ext': 'mp4',
  74             'title': 'Космическая среда от 26 августа 2015',
  75             'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0',
  76             'duration': 440,
  77             'upload_date': '20150826',
  78             'uploader_id': 'tvroscosmos',
  79             'uploader': 'Телестудия Роскосмоса',
  80             'age_limit': 0,
  81         },
  82     }, {
  83         # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
  84         'url': 'http://ok.ru/video/62036049272859-0',
  85         'info_dict': {
  86             'id': '62036049272859-0',
  87             'ext': 'mp4',
  88             'title': 'МУЗЫКА     ДОЖДЯ .',
  89             'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
  90             'upload_date': '20120106',
  91             'uploader_id': '473534735899',
  92             'uploader': 'МARINA D',
  93             'age_limit': 0,
  94         },
  95         'params': {
  96             'skip_download': True,
  97         },
  98         'skip': 'Video has not been found',
  99     }, {
 100         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
 101         'only_matching': True,
 102     }, {
 103         'url': 'http://www.ok.ru/video/20648036891',
 104         'only_matching': True,
 105     }, {
 106         'url': 'http://www.ok.ru/videoembed/20648036891',
 107         'only_matching': True,
 108     }, {
 109         'url': 'http://m.ok.ru/video/20079905452',
 110         'only_matching': True,
 111     }, {
 112         'url': 'http://mobile.ok.ru/video/20079905452',
 113         'only_matching': True,
 114     }, {
 115         'url': 'https://www.ok.ru/live/484531969818',
 116         'only_matching': True,
 117     }, {
 118         'url': 'https://m.ok.ru/dk?st.cmd=movieLayer&st.discId=863789452017&st.retLoc=friend&st.rtu=%2Fdk%3Fst.cmd%3DfriendMovies%26st.mode%3Down%26st.mrkId%3D%257B%2522uploadedMovieMarker%2522%253A%257B%2522marker%2522%253A%25221519410114503%2522%252C%2522hasMore%2522%253Atrue%257D%252C%2522sharedMovieMarker%2522%253A%257B%2522marker%2522%253Anull%252C%2522hasMore%2522%253Afalse%257D%257D%26st.friendId%3D561722190321%26st.frwd%3Don%26_prevCmd%3DfriendMovies%26tkn%3D7257&st.discType=MOVIE&st.mvId=863789452017&_prevCmd=friendMovies&tkn=3648#lst#',
 119         'only_matching': True,
 120     }, {
 121         # Paid video
 122         'url': 'https://ok.ru/video/954886983203',
 123         'only_matching': True,
 124     }]
 125
 126     @staticmethod
 127     def _extract_url(webpage):
 128         mobj = re.search(
 129             r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:odnoklassniki|ok)\.ru/videoembed/.+?)\1', webpage)
 130         if mobj:
 131             return mobj.group('url')
 132
 133     def _real_extract(self, url):
 134         start_time = int_or_none(compat_parse_qs(
 135             compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
 136
 137         video_id = self._match_id(url)
 138
 139         webpage = self._download_webpage(
 140             'http://ok.ru/video/%s' % video_id, video_id)
 141
 142         error = self._search_regex(
 143             r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<',
 144             webpage, 'error', default=None)
 145         if error:
 146             raise ExtractorError(error, expected=True)
 147
 148         player = self._parse_json(
 149             unescapeHTML(self._search_regex(
 150                 r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
 151                 webpage, 'player', group='player')),
 152             video_id)
 153
 154         flashvars = player['flashvars']
 155
 156         metadata = flashvars.get('metadata')
 157         if metadata:
 158             metadata = self._parse_json(metadata, video_id)
 159         else:
 160             data = {}
 161             st_location = flashvars.get('location')
 162             if st_location:
 163                 data['st.location'] = st_location
 164             metadata = self._download_json(
 165                 compat_urllib_parse_unquote(flashvars['metadataUrl']),
 166                 video_id, 'Downloading metadata JSON',
 167                 data=urlencode_postdata(data))
 168
 169         movie = metadata['movie']
 170
 171         # Some embedded videos may not contain title in movie dict (e.g.
 172         # http://ok.ru/video/62036049272859-0) thus we allow missing title
 173         # here and it's going to be extracted later by an extractor that
 174         # will process the actual embed.
 175         provider = metadata.get('provider')
 176         title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
 177
 178         thumbnail = movie.get('poster')
 179         duration = int_or_none(movie.get('duration'))
 180
 181         author = metadata.get('author', {})
 182         uploader_id = author.get('id')
 183         uploader = author.get('name')
 184
 185         upload_date = unified_strdate(self._html_search_meta(
 186             'ya:ovs:upload_date', webpage, 'upload date', default=None))
 187
 188         age_limit = None
 189         adult = self._html_search_meta(
 190             'ya:ovs:adult', webpage, 'age limit', default=None)
 191         if adult:
 192             age_limit = 18 if adult == 'true' else 0
 193
 194         like_count = int_or_none(metadata.get('likeCount'))
 195
 196         info = {
 197             'id': video_id,
 198             'title': title,
 199             'thumbnail': thumbnail,
 200             'duration': duration,
 201             'upload_date': upload_date,
 202             'uploader': uploader,
 203             'uploader_id': uploader_id,
 204             'like_count': like_count,
 205             'age_limit': age_limit,
 206             'start_time': start_time,
 207         }
 208
 209         if provider == 'USER_YOUTUBE':
 210             info.update({
 211                 '_type': 'url_transparent',
 212                 'url': movie['contentId'],
 213             })
 214             return info
 215
 216         assert title
 217         if provider == 'LIVE_TV_APP':
 218             info['title'] = self._live_title(title)
 219
 220         quality = qualities(('4', '0', '1', '2', '3', '5'))
 221
 222         formats = [{
 223             'url': f['url'],
 224             'ext': 'mp4',
 225             'format_id': f['name'],
 226         } for f in metadata['videos']]
 227
 228         m3u8_url = metadata.get('hlsManifestUrl')
 229         if m3u8_url:
 230             formats.extend(self._extract_m3u8_formats(
 231                 m3u8_url, video_id, 'mp4', 'm3u8_native',
 232                 m3u8_id='hls', fatal=False))
 233
 234         dash_manifest = metadata.get('metadataEmbedded')
 235         if dash_manifest:
 236             formats.extend(self._parse_mpd_formats(
 237                 compat_etree_fromstring(dash_manifest), 'mpd'))
 238
 239         for fmt in formats:
 240             fmt_type = self._search_regex(
 241                 r'\btype[/=](\d)', fmt['url'],
 242                 'format type', default=None)
 243             if fmt_type:
 244                 fmt['quality'] = quality(fmt_type)
 245
 246         # Live formats
 247         m3u8_url = metadata.get('hlsMasterPlaylistUrl')
 248         if m3u8_url:
 249             formats.extend(self._extract_m3u8_formats(
 250                 m3u8_url, video_id, 'mp4', entry_protocol='m3u8',
 251                 m3u8_id='hls', fatal=False))
 252         rtmp_url = metadata.get('rtmpUrl')
 253         if rtmp_url:
 254             formats.append({
 255                 'url': rtmp_url,
 256                 'format_id': 'rtmp',
 257                 'ext': 'flv',
 258             })
 259
 260         if not formats:
 261             payment_info = metadata.get('paymentInfo')
 262             if payment_info:
 263                 raise ExtractorError('This video is paid, subscribe to download it', expected=True)
 264
 265         self._sort_formats(formats)
 266
 267         info['formats'] = formats
 268         return info