_ Git - youtube-dl/blob - youtube_dl/extractor/viewster.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..compat import (
   6     compat_urllib_request,
   7     compat_urllib_parse,
   8     compat_urllib_parse_unquote,
   9 )
  10 from ..utils import (
  11     determine_ext,
  12     int_or_none,
  13     parse_iso8601,
  14 )
  15
  16
  17 class ViewsterIE(InfoExtractor):
  18     _VALID_URL = r'http://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)'
  19     _TESTS = [{
  20         # movie, Type=Movie
  21         'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/',
  22         'md5': '14d3cfffe66d57b41ae2d9c873416f01',
  23         'info_dict': {
  24             'id': '1140-11855-000',
  25             'ext': 'flv',
  26             'title': 'The listening Project',
  27             'description': 'md5:bac720244afd1a8ea279864e67baa071',
  28             'timestamp': 1214870400,
  29             'upload_date': '20080701',
  30             'duration': 4680,
  31         },
  32     }, {
  33         # series episode, Type=Episode
  34         'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/',
  35         'md5': 'd5434c80fcfdb61651cc2199a88d6ba3',
  36         'info_dict': {
  37             'id': '1284-19427-001',
  38             'ext': 'flv',
  39             'title': 'The World and a Wall',
  40             'description': 'md5:24814cf74d3453fdf5bfef9716d073e3',
  41             'timestamp': 1428192000,
  42             'upload_date': '20150405',
  43             'duration': 1500,
  44         },
  45     }, {
  46         # serie, Type=Serie
  47         'url': 'http://www.viewster.com/serie/1303-19426-000/',
  48         'info_dict': {
  49             'id': '1303-19426-000',
  50             'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?',
  51             'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11',
  52         },
  53         'playlist_count': 13,
  54     }, {
  55         # unfinished serie, no Type
  56         'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/',
  57         'info_dict': {
  58             'id': '1284-19427-000',
  59             'title': 'Baby Steps—Season 2',
  60             'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1',
  61         },
  62         'playlist_mincount': 16,
  63     }]
  64
  65     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
  66
  67     def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True):
  68         request = compat_urllib_request.Request(url)
  69         request.add_header('Accept', self._ACCEPT_HEADER)
  70         request.add_header('Auth-token', self._AUTH_TOKEN)
  71         return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal)
  72
  73     def _real_extract(self, url):
  74         video_id = self._match_id(url)
  75         # Get 'api_token' cookie
  76         self._request_webpage(url, video_id)
  77         cookies = self._get_cookies(url)
  78         self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value)
  79
  80         info = self._download_json(
  81             'https://public-api.viewster.com/search/%s' % video_id,
  82             video_id, 'Downloading entry JSON')
  83
  84         entry_id = info.get('Id') or info['id']
  85
  86         # unfinished serie has no Type
  87         if info.get('Type') in ['Serie', None]:
  88             episodes = self._download_json(
  89                 'https://public-api.viewster.com/series/%s/episodes' % entry_id,
  90                 video_id, 'Downloading series JSON')
  91             entries = [
  92                 self.url_result(
  93                     'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster')
  94                 for episode in episodes]
  95             title = (info.get('Title') or info['Synopsis']['Title']).strip()
  96             description = info.get('Synopsis', {}).get('Detailed')
  97             return self.playlist_result(entries, video_id, title, description)
  98
  99         formats = []
 100         for media_type in ('application/f4m+xml', 'application/x-mpegURL'):
 101             media = self._download_json(
 102                 'https://public-api.viewster.com/movies/%s/video?mediaType=%s'
 103                 % (entry_id, compat_urllib_parse.quote(media_type)),
 104                 video_id, 'Downloading %s JSON' % media_type, fatal=False)
 105             if not media:
 106                 continue
 107             video_url = media.get('Uri')
 108             if not video_url:
 109                 continue
 110             ext = determine_ext(video_url)
 111             if ext == 'f4m':
 112                 video_url += '&' if '?' in video_url else '?'
 113                 video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
 114                 formats.extend(self._extract_f4m_formats(
 115                     video_url, video_id, f4m_id='hds'))
 116             elif ext == 'm3u8':
 117                 formats.extend(self._extract_m3u8_formats(
 118                     video_url, video_id, 'mp4', m3u8_id='hls',
 119                     fatal=False  # m3u8 sometimes fail
 120                 ))
 121             else:
 122                 formats.append({
 123                     'url': video_url,
 124                 })
 125         self._sort_formats(formats)
 126
 127         synopsis = info.get('Synopsis', {})
 128         # Prefer title outside synopsis since it's less messy
 129         title = (info.get('Title') or synopsis['Title']).strip()
 130         description = synopsis.get('Detailed') or info.get('Synopsis', {}).get('Short')
 131         duration = int_or_none(info.get('Duration'))
 132         timestamp = parse_iso8601(info.get('ReleaseDate'))
 133
 134         return {
 135             'id': video_id,
 136             'title': title,
 137             'description': description,
 138             'timestamp': timestamp,
 139             'duration': duration,
 140             'formats': formats,
 141         }