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