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     HEADRequest,
  15 )
  16
  17
  18 class ViewsterIE(InfoExtractor):
  19     _VALID_URL = r'http://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)'
  20     _TESTS = [{
  21         # movie, Type=Movie
  22         'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/',
  23         'md5': '14d3cfffe66d57b41ae2d9c873416f01',
  24         'info_dict': {
  25             'id': '1140-11855-000',
  26             'ext': 'flv',
  27             'title': 'The listening Project',
  28             'description': 'md5:bac720244afd1a8ea279864e67baa071',
  29             'timestamp': 1214870400,
  30             'upload_date': '20080701',
  31             'duration': 4680,
  32         },
  33     }, {
  34         # series episode, Type=Episode
  35         'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/',
  36         'md5': 'd5434c80fcfdb61651cc2199a88d6ba3',
  37         'info_dict': {
  38             'id': '1284-19427-001',
  39             'ext': 'flv',
  40             'title': 'The World and a Wall',
  41             'description': 'md5:24814cf74d3453fdf5bfef9716d073e3',
  42             'timestamp': 1428192000,
  43             'upload_date': '20150405',
  44             'duration': 1500,
  45         },
  46     }, {
  47         # serie, Type=Serie
  48         'url': 'http://www.viewster.com/serie/1303-19426-000/',
  49         'info_dict': {
  50             'id': '1303-19426-000',
  51             'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?',
  52             'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11',
  53         },
  54         'playlist_count': 13,
  55     }, {
  56         # unfinished serie, no Type
  57         'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/',
  58         'info_dict': {
  59             'id': '1284-19427-000',
  60             'title': 'Baby Steps—Season 2',
  61             'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1',
  62         },
  63         'playlist_mincount': 16,
  64     }]
  65
  66     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
  67
  68     def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True):
  69         request = compat_urllib_request.Request(url)
  70         request.add_header('Accept', self._ACCEPT_HEADER)
  71         request.add_header('Auth-token', self._AUTH_TOKEN)
  72         return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal)
  73
  74     def _real_extract(self, url):
  75         video_id = self._match_id(url)
  76         # Get 'api_token' cookie
  77         self._request_webpage(HEADRequest(url), video_id)
  78         cookies = self._get_cookies(url)
  79         self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value)
  80
  81         info = self._download_json(
  82             'https://public-api.viewster.com/search/%s' % video_id,
  83             video_id, 'Downloading entry JSON')
  84
  85         entry_id = info.get('Id') or info['id']
  86
  87         # unfinished serie has no Type
  88         if info.get('Type') in ['Serie', None]:
  89             episodes = self._download_json(
  90                 'https://public-api.viewster.com/series/%s/episodes' % entry_id,
  91                 video_id, 'Downloading series JSON')
  92             entries = [
  93                 self.url_result(
  94                     'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster')
  95                 for episode in episodes]
  96             title = (info.get('Title') or info['Synopsis']['Title']).strip()
  97             description = info.get('Synopsis', {}).get('Detailed')
  98             return self.playlist_result(entries, video_id, title, description)
  99
 100         formats = []
 101         for media_type in ('application/f4m+xml', 'application/x-mpegURL'):
 102             media = self._download_json(
 103                 'https://public-api.viewster.com/movies/%s/video?mediaType=%s'
 104                 % (entry_id, compat_urllib_parse.quote(media_type)),
 105                 video_id, 'Downloading %s JSON' % media_type, fatal=False)
 106             if not media:
 107                 continue
 108             video_url = media.get('Uri')
 109             if not video_url:
 110                 continue
 111             ext = determine_ext(video_url)
 112             if ext == 'f4m':
 113                 video_url += '&' if '?' in video_url else '?'
 114                 video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
 115                 formats.extend(self._extract_f4m_formats(
 116                     video_url, video_id, f4m_id='hds'))
 117             elif ext == 'm3u8':
 118                 formats.extend(self._extract_m3u8_formats(
 119                     video_url, video_id, 'mp4', m3u8_id='hls',
 120                     fatal=False  # m3u8 sometimes fail
 121                 ))
 122             else:
 123                 formats.append({
 124                     'url': video_url,
 125                 })
 126         self._sort_formats(formats)
 127
 128         synopsis = info.get('Synopsis', {})
 129         # Prefer title outside synopsis since it's less messy
 130         title = (info.get('Title') or synopsis['Title']).strip()
 131         description = synopsis.get('Detailed') or info.get('Synopsis', {}).get('Short')
 132         duration = int_or_none(info.get('Duration'))
 133         timestamp = parse_iso8601(info.get('ReleaseDate'))
 134
 135         return {
 136             'id': video_id,
 137             'title': title,
 138             'description': description,
 139             'timestamp': timestamp,
 140             'duration': duration,
 141             'formats': formats,
 142         }