Raphaël G. Git Repositories - youtubedl/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_unquote,
  10 )
  11 from ..utils import (
  12     determine_ext,
  13     ExtractorError,
  14     int_or_none,
  15     parse_iso8601,
  16     sanitized_Request,
  17     HEADRequest,
  18     url_basename,
  19 )
  20
  21
  22 class ViewsterIE(InfoExtractor):
  23     _VALID_URL = r'https?://(?:www\.)?viewster\.com/(?:serie|movie)/(?P<id>\d+-\d+-\d+)'
  24     _TESTS = [{
  25         # movie, Type=Movie
  26         'url': 'http://www.viewster.com/movie/1140-11855-000/the-listening-project/',
  27         'md5': 'e642d1b27fcf3a4ffa79f194f5adde36',
  28         'info_dict': {
  29             'id': '1140-11855-000',
  30             'ext': 'mp4',
  31             'title': 'The listening Project',
  32             'description': 'md5:bac720244afd1a8ea279864e67baa071',
  33             'timestamp': 1214870400,
  34             'upload_date': '20080701',
  35             'duration': 4680,
  36         },
  37     }, {
  38         # series episode, Type=Episode
  39         'url': 'http://www.viewster.com/serie/1284-19427-001/the-world-and-a-wall/',
  40         'md5': '9243079a8531809efe1b089db102c069',
  41         'info_dict': {
  42             'id': '1284-19427-001',
  43             'ext': 'mp4',
  44             'title': 'The World and a Wall',
  45             'description': 'md5:24814cf74d3453fdf5bfef9716d073e3',
  46             'timestamp': 1428192000,
  47             'upload_date': '20150405',
  48             'duration': 1500,
  49         },
  50     }, {
  51         # serie, Type=Serie
  52         'url': 'http://www.viewster.com/serie/1303-19426-000/',
  53         'info_dict': {
  54             'id': '1303-19426-000',
  55             'title': 'Is It Wrong to Try to Pick up Girls in a Dungeon?',
  56             'description': 'md5:eeda9bef25b0d524b3a29a97804c2f11',
  57         },
  58         'playlist_count': 13,
  59     }, {
  60         # unfinished serie, no Type
  61         'url': 'http://www.viewster.com/serie/1284-19427-000/baby-steps-season-2/',
  62         'info_dict': {
  63             'id': '1284-19427-000',
  64             'title': 'Baby Steps—Season 2',
  65             'description': 'md5:e7097a8fc97151e25f085c9eb7a1cdb1',
  66         },
  67         'playlist_mincount': 16,
  68     }, {
  69         # geo restricted series
  70         'url': 'https://www.viewster.com/serie/1280-18794-002/',
  71         'only_matching': True,
  72     }, {
  73         # geo restricted video
  74         'url': 'https://www.viewster.com/serie/1280-18794-002/what-is-extraterritoriality-lawo/',
  75         'only_matching': True,
  76     }]
  77
  78     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
  79
  80     def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True, query={}):
  81         request = sanitized_Request(url)
  82         request.add_header('Accept', self._ACCEPT_HEADER)
  83         request.add_header('Auth-token', self._AUTH_TOKEN)
  84         return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal, query=query)
  85
  86     def _real_extract(self, url):
  87         video_id = self._match_id(url)
  88         # Get 'api_token' cookie
  89         self._request_webpage(
  90             HEADRequest('http://www.viewster.com/'),
  91             video_id, headers=self.geo_verification_headers())
  92         cookies = self._get_cookies('http://www.viewster.com/')
  93         self._AUTH_TOKEN = compat_urllib_parse_unquote(cookies['api_token'].value)
  94
  95         info = self._download_json(
  96             'https://public-api.viewster.com/search/%s' % video_id,
  97             video_id, 'Downloading entry JSON')
  98
  99         entry_id = info.get('Id') or info['id']
 100
 101         # unfinished serie has no Type
 102         if info.get('Type') in ('Serie', None):
 103             try:
 104                 episodes = self._download_json(
 105                     'https://public-api.viewster.com/series/%s/episodes' % entry_id,
 106                     video_id, 'Downloading series JSON')
 107             except ExtractorError as e:
 108                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404:
 109                     self.raise_geo_restricted()
 110                 else:
 111                     raise
 112             entries = [
 113                 self.url_result(
 114                     'http://www.viewster.com/movie/%s' % episode['OriginId'], 'Viewster')
 115                 for episode in episodes]
 116             title = (info.get('Title') or info['Synopsis']['Title']).strip()
 117             description = info.get('Synopsis', {}).get('Detailed')
 118             return self.playlist_result(entries, video_id, title, description)
 119
 120         formats = []
 121         for language_set in info.get('LanguageSets', []):
 122             manifest_url = None
 123             m3u8_formats = []
 124             audio = language_set.get('Audio') or ''
 125             subtitle = language_set.get('Subtitle') or ''
 126             base_format_id = audio
 127             if subtitle:
 128                 base_format_id += '-%s' % subtitle
 129
 130             def concat(suffix, sep='-'):
 131                 return (base_format_id + '%s%s' % (sep, suffix)) if base_format_id else suffix
 132
 133             for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
 134                 media = self._download_json(
 135                     'https://public-api.viewster.com/movies/%s/video' % entry_id,
 136                     video_id, 'Downloading %s JSON' % concat(media_type, ' '), fatal=False, query={
 137                         'mediaType': media_type,
 138                         'language': audio,
 139                         'subtitle': subtitle,
 140                     })
 141                 if not media:
 142                     continue
 143                 video_url = media.get('Uri')
 144                 if not video_url:
 145                     continue
 146                 ext = determine_ext(video_url)
 147                 if ext == 'f4m':
 148                     manifest_url = video_url
 149                     video_url += '&' if '?' in video_url else '?'
 150                     video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
 151                     formats.extend(self._extract_f4m_formats(
 152                         video_url, video_id, f4m_id=concat('hds')))
 153                 elif ext == 'm3u8':
 154                     manifest_url = video_url
 155                     m3u8_formats = self._extract_m3u8_formats(
 156                         video_url, video_id, 'mp4', m3u8_id=concat('hls'),
 157                         fatal=False)  # m3u8 sometimes fail
 158                     if m3u8_formats:
 159                         formats.extend(m3u8_formats)
 160                 else:
 161                     qualities_basename = self._search_regex(
 162                         r'/([^/]+)\.csmil/',
 163                         manifest_url, 'qualities basename', default=None)
 164                     if not qualities_basename:
 165                         continue
 166                     QUALITIES_RE = r'((,\d+k)+,?)'
 167                     qualities = self._search_regex(
 168                         QUALITIES_RE, qualities_basename,
 169                         'qualities', default=None)
 170                     if not qualities:
 171                         continue
 172                     qualities = list(map(lambda q: int(q[:-1]), qualities.strip(',').split(',')))
 173                     qualities.sort()
 174                     http_template = re.sub(QUALITIES_RE, r'%dk', qualities_basename)
 175                     http_url_basename = url_basename(video_url)
 176                     if m3u8_formats:
 177                         self._sort_formats(m3u8_formats)
 178                         m3u8_formats = list(filter(
 179                             lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
 180                             m3u8_formats))
 181                     if len(qualities) == len(m3u8_formats):
 182                         for q, m3u8_format in zip(qualities, m3u8_formats):
 183                             f = m3u8_format.copy()
 184                             f.update({
 185                                 'url': video_url.replace(http_url_basename, http_template % q),
 186                                 'format_id': f['format_id'].replace('hls', 'http'),
 187                                 'protocol': 'http',
 188                             })
 189                             formats.append(f)
 190                     else:
 191                         for q in qualities:
 192                             formats.append({
 193                                 'url': video_url.replace(http_url_basename, http_template % q),
 194                                 'ext': 'mp4',
 195                                 'format_id': 'http-%d' % q,
 196                                 'tbr': q,
 197                             })
 198
 199         if not formats and not info.get('VODSettings'):
 200             self.raise_geo_restricted()
 201
 202         self._sort_formats(formats)
 203
 204         synopsis = info.get('Synopsis') or {}
 205         # Prefer title outside synopsis since it's less messy
 206         title = (info.get('Title') or synopsis['Title']).strip()
 207         description = synopsis.get('Detailed') or (info.get('Synopsis') or {}).get('Short')
 208         duration = int_or_none(info.get('Duration'))
 209         timestamp = parse_iso8601(info.get('ReleaseDate'))
 210
 211         return {
 212             'id': video_id,
 213             'title': title,
 214             'description': description,
 215             'timestamp': timestamp,
 216             'duration': duration,
 217             'formats': formats,
 218         }