_ Git - youtube-dl/blob - 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, query={}):
  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, query=query)
  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         for language_set in info.get('LanguageSets', []):
 121             manifest_url = None
 122             m3u8_formats = []
 123             audio = language_set.get('Audio') or ''
 124             subtitle = language_set.get('Subtitle') or ''
 125             base_format_id = audio
 126             if subtitle:
 127                 base_format_id += '-%s' % subtitle
 128
 129             def concat(suffix, sep='-'):
 130                 return (base_format_id + '%s%s' % (sep, suffix)) if base_format_id else suffix
 131
 132             for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
 133                 media = self._download_json(
 134                     'https://public-api.viewster.com/movies/%s/video' % entry_id,
 135                     video_id, 'Downloading %s JSON' % concat(media_type, ' '), fatal=False, query={
 136                         'mediaType': media_type,
 137                         'language': audio,
 138                         'subtitle': subtitle,
 139                     })
 140                 if not media:
 141                     continue
 142                 video_url = media.get('Uri')
 143                 if not video_url:
 144                     continue
 145                 ext = determine_ext(video_url)
 146                 if ext == 'f4m':
 147                     manifest_url = video_url
 148                     video_url += '&' if '?' in video_url else '?'
 149                     video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
 150                     formats.extend(self._extract_f4m_formats(
 151                         video_url, video_id, f4m_id=concat('hds')))
 152                 elif ext == 'm3u8':
 153                     manifest_url = video_url
 154                     m3u8_formats = self._extract_m3u8_formats(
 155                         video_url, video_id, 'mp4', m3u8_id=concat('hls'),
 156                         fatal=False)  # m3u8 sometimes fail
 157                     if m3u8_formats:
 158                         formats.extend(m3u8_formats)
 159                 else:
 160                     qualities_basename = self._search_regex(
 161                         '/([^/]+)\.csmil/',
 162                         manifest_url, 'qualities basename', default=None)
 163                     if not qualities_basename:
 164                         continue
 165                     QUALITIES_RE = r'((,\d+k)+,?)'
 166                     qualities = self._search_regex(
 167                         QUALITIES_RE, qualities_basename,
 168                         'qualities', default=None)
 169                     if not qualities:
 170                         continue
 171                     qualities = list(map(lambda q: int(q[:-1]), qualities.strip(',').split(',')))
 172                     qualities.sort()
 173                     http_template = re.sub(QUALITIES_RE, r'%dk', qualities_basename)
 174                     http_url_basename = url_basename(video_url)
 175                     if m3u8_formats:
 176                         self._sort_formats(m3u8_formats)
 177                         m3u8_formats = list(filter(
 178                             lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
 179                             m3u8_formats))
 180                     if len(qualities) == len(m3u8_formats):
 181                         for q, m3u8_format in zip(qualities, m3u8_formats):
 182                             f = m3u8_format.copy()
 183                             f.update({
 184                                 'url': video_url.replace(http_url_basename, http_template % q),
 185                                 'format_id': f['format_id'].replace('hls', 'http'),
 186                                 'protocol': 'http',
 187                             })
 188                             formats.append(f)
 189                     else:
 190                         for q in qualities:
 191                             formats.append({
 192                                 'url': video_url.replace(http_url_basename, http_template % q),
 193                                 'ext': 'mp4',
 194                                 'format_id': 'http-%d' % q,
 195                                 'tbr': q,
 196                             })
 197
 198         if not formats and not info.get('VODSettings'):
 199             self.raise_geo_restricted()
 200
 201         self._sort_formats(formats)
 202
 203         synopsis = info.get('Synopsis') or {}
 204         # Prefer title outside synopsis since it's less messy
 205         title = (info.get('Title') or synopsis['Title']).strip()
 206         description = synopsis.get('Detailed') or (info.get('Synopsis') or {}).get('Short')
 207         duration = int_or_none(info.get('Duration'))
 208         timestamp = parse_iso8601(info.get('ReleaseDate'))
 209
 210         return {
 211             'id': video_id,
 212             'title': title,
 213             'description': description,
 214             'timestamp': timestamp,
 215             'duration': duration,
 216             'formats': formats,
 217         }