youtube_dl/extractor/viewster.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_HTTPError,
   9     compat_urllib_parse,
  10     compat_urllib_parse_unquote,
  11 )
  12 from ..utils import (
  13     determine_ext,
  14     ExtractorError,
  15     int_or_none,
  16     parse_iso8601,
  17     sanitized_Request,
  18     HEADRequest,
  19     url_basename,
  20 )
  21
  22
  23 class ViewsterIE(InfoExtractor):
  24     _VALID_URL = r'https?://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)'
  25     _TESTS = [{
  26         # movie, Type=Movie
  27         'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/',
  28         'md5': 'e642d1b27fcf3a4ffa79f194f5adde36',
  29         'info_dict': {
  30             'id': '1140-11855-000',
  31             'ext': 'mp4',
  32             'title': 'The listening Project',
  33             'description': 'md5:bac720244afd1a8ea279864e67baa071',
  34             'timestamp': 1214870400,
  35             'upload_date': '20080701',
  36             'duration': 4680,
  37         },
  38     }, {
  39         # series episode, Type=Episode
  40         'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/',
  41         'md5': '9243079a8531809efe1b089db102c069',
  42         'info_dict': {
  43             'id': '1284-19427-001',
  44             'ext': 'mp4',
  45             'title': 'The World and a Wall',
  46             'description': 'md5:24814cf74d3453fdf5bfef9716d073e3',
  47             'timestamp': 1428192000,
  48             'upload_date': '20150405',
  49             'duration': 1500,
  50         },
  51     }, {
  52         # serie, Type=Serie
  53         'url': 'http://www.viewster.com/serie/1303-19426-000/',
  54         'info_dict': {
  55             'id': '1303-19426-000',
  56             'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?',
  57             'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11',
  58         },
  59         'playlist_count': 13,
  60     }, {
  61         # unfinished serie, no Type
  62         'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/',
  63         'info_dict': {
  64             'id': '1284-19427-000',
  65             'title': 'Baby Steps—Season 2',
  66             'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1',
  67         },
  68         'playlist_mincount': 16,
  69     }, {
  70         # geo restricted series
  71         'url': 'https://www.viewster.com/serie/1280-18794-002/',
  72         'only_matching': True,
  73     }, {
  74         # geo restricted video
  75         'url': 'https://www.viewster.com/serie/1280-18794-002/what-is-extraterritoriality-lawo/',
  76         'only_matching': True,
  77     }]
  78
  79     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
  80
  81     def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True):
  82         request = sanitized_Request(url)
  83         request.add_header('Accept', self._ACCEPT_HEADER)
  84         request.add_header('Auth-token', self._AUTH_TOKEN)
  85         return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal)
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89         # Get 'api_token' cookie
  90         self._request_webpage(HEADRequest('http://www.viewster.com/'), video_id)
  91         cookies = self._get_cookies('http://www.viewster.com/')
  92         self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value)
  93
  94         info = self._download_json(
  95             'https://public-api.viewster.com/search/%s' % video_id,
  96             video_id, 'Downloading entry JSON')
  97
  98         entry_id = info.get('Id') or info['id']
  99
 100         # unfinished serie has no Type
 101         if info.get('Type') in ('Serie', None):
 102             try:
 103                 episodes = self._download_json(
 104                     'https://public-api.viewster.com/series/%s/episodes' % entry_id,
 105                     video_id, 'Downloading series JSON')
 106             except ExtractorError as e:
 107                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 108                     self.raise_geo_restricted()
 109                 else:
 110                     raise
 111             entries = [
 112                 self.url_result(
 113                     'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster')
 114                 for episode in episodes]
 115             title = (info.get('Title') or info['Synopsis']['Title']).strip()
 116             description = info.get('Synopsis', {}).get('Detailed')
 117             return self.playlist_result(entries, video_id, title, description)
 118
 119         formats = []
 120         manifest_url = None
 121         m3u8_formats = []
 122         for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
 123             media = self._download_json(
 124                 'https://public-api.viewster.com/movies/%s/video?mediaType=%s'
 125                 % (entry_id, compat_urllib_parse.quote(media_type)),
 126                 video_id, 'Downloading %s JSON' % media_type, fatal=False)
 127             if not media:
 128                 continue
 129             video_url = media.get('Uri')
 130             if not video_url:
 131                 continue
 132             ext = determine_ext(video_url)
 133             if ext == 'f4m':
 134                 manifest_url = video_url
 135                 video_url += '&' if '?' in video_url else '?'
 136                 video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
 137                 formats.extend(self._extract_f4m_formats(
 138                     video_url, video_id, f4m_id='hds'))
 139             elif ext == 'm3u8':
 140                 manifest_url = video_url
 141                 m3u8_formats = self._extract_m3u8_formats(
 142                     video_url, video_id, 'mp4', m3u8_id='hls',
 143                     fatal=False)  # m3u8 sometimes fail
 144                 if m3u8_formats:
 145                     formats.extend(m3u8_formats)
 146             else:
 147                 qualities_basename = self._search_regex(
 148                     '/([^/]+)\.csmil/',
 149                     manifest_url, 'qualities basename', default=None)
 150                 if not qualities_basename:
 151                     continue
 152                 QUALITIES_RE = r'((,\d+k)+,?)'
 153                 qualities = self._search_regex(
 154                     QUALITIES_RE, qualities_basename,
 155                     'qualities', default=None)
 156                 if not qualities:
 157                     continue
 158                 qualities = list(map(lambda q: int(q[:-1]), qualities.strip(',').split(',')))
 159                 qualities.sort()
 160                 http_template = re.sub(QUALITIES_RE, r'%dk', qualities_basename)
 161                 http_url_basename = url_basename(video_url)
 162                 if m3u8_formats:
 163                     self._sort_formats(m3u8_formats)
 164                     m3u8_formats = list(filter(
 165                         lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
 166                         m3u8_formats))
 167                 if len(qualities) == len(m3u8_formats):
 168                     for q, m3u8_format in zip(qualities, m3u8_formats):
 169                         f = m3u8_format.copy()
 170                         f.update({
 171                             'url': video_url.replace(http_url_basename, http_template % q),
 172                             'format_id': f['format_id'].replace('hls', 'http'),
 173                             'protocol': 'http',
 174                         })
 175                         formats.append(f)
 176                 else:
 177                     for q in qualities:
 178                         formats.append({
 179                             'url': video_url.replace(http_url_basename, http_template % q),
 180                             'ext': 'mp4',
 181                             'format_id': 'http-%d' % q,
 182                             'tbr': q,
 183                         })
 184
 185         if not formats and not info.get('LanguageSets') and not info.get('VODSettings'):
 186             self.raise_geo_restricted()
 187
 188         self._sort_formats(formats)
 189
 190         synopsis = info.get('Synopsis') or {}
 191         # Prefer title outside synopsis since it's less messy
 192         title = (info.get('Title') or synopsis['Title']).strip()
 193         description = synopsis.get('Detailed') or (info.get('Synopsis') or {}).get('Short')
 194         duration = int_or_none(info.get('Duration'))
 195         timestamp = parse_iso8601(info.get('ReleaseDate'))
 196
 197         return {
 198             'id': video_id,
 199             'title': title,
 200             'description': description,
 201             'timestamp': timestamp,
 202             'duration': duration,
 203             'formats': formats,
 204         }