_ Git - youtube-dl/blob - youtube_dl/extractor/viu.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     clean_html,
  11 )
  12
  13
  14 class ViuBaseIE(InfoExtractor):
  15
  16     def _get_viu_auth(self, video_id):
  17         viu_auth_res = self._request_webpage(
  18             'https://www.viu.com/api/apps/v2/authenticate', video_id,
  19             note='Requesting Viu auth',
  20             query={
  21                 'acct': 'test', 'appid': 'viu_desktop', 'fmt': 'json',
  22                 'iid': 'guest', 'languageid': 'default', 'platform': 'desktop',
  23                 'userid': 'guest', 'useridtype': 'guest', 'ver': '1.0'
  24             })
  25         return viu_auth_res.info().get('X-VIU-AUTH')
  26
  27
  28 class ViuIE(ViuBaseIE):
  29     IE_NAME = 'viu:show'
  30     _VALID_URL = r'https?://www\.viu\.com/.+/(?:vod|media)/(?P<id>[0-9]+)'
  31     _TESTS = [{
  32         'url': 'http://www.viu.com/ott/sg/en-us/vod/3421/The%20Prime%20Minister%20and%20I',
  33         'info_dict': {
  34             'id': '3421',
  35             'ext': 'mp4',
  36             'title': 'The Prime Minister and I - Episode 17',
  37             'description': 'md5:1e7486a619b6399b25ba6a41c0fe5b2c',
  38         },
  39         'params': {
  40             'skip_download': 'm3u8 download',
  41         },
  42         'skip': 'Geo-restricted to Singapore',
  43     }, {
  44         'url': 'http://www.viu.com/ott/hk/zh-hk/vod/7123/%E5%A4%A7%E4%BA%BA%E5%A5%B3%E5%AD%90',
  45         'info_dict': {
  46             'id': '7123',
  47             'ext': 'mp4',
  48             'title': '大人女子 - Episode 10',
  49             'description': 'md5:4eb0d8b08cf04fcdc6bbbeb16043434f',
  50         },
  51         'params': {
  52             'skip_download': 'm3u8 download',
  53         },
  54         'skip': 'Geo-restricted to Hong Kong',
  55     }, {
  56         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
  57         'info_dict': {
  58             'id': '1116705532',
  59             'ext': 'mp4',
  60             'title': 'Citizen Khan - Episode 1',
  61             'description': 'md5:d7ea1604f49e5ba79c212c551ce2110e',
  62         },
  63         'params': {
  64             'skip_download': 'm3u8 download',
  65         },
  66         'skip': 'Geo-restricted to India',
  67     }, {
  68         'url': 'https://www.viu.com/en/media/1130599965',
  69         'info_dict': {
  70             'id': '1130599965',
  71             'ext': 'mp4',
  72             'title': 'Jealousy Incarnate - Episode 1',
  73             'description': 'md5:d3d82375cab969415d2720b6894361e9',
  74         },
  75         'params': {
  76             'skip_download': 'm3u8 download',
  77         },
  78         'skip': 'Geo-restricted to Indonesia',
  79     }]
  80
  81     def _real_extract(self, url):
  82         video_id = self._match_id(url)
  83
  84         webpage = self._download_webpage(
  85             url, video_id, note='Downloading video page')
  86
  87         mobj = re.search(
  88             r'<div class=["\']error-title[^<>]+?>(?P<err>.+?)</div>', webpage, flags=re.DOTALL)
  89
  90         if mobj:
  91             raise ExtractorError(clean_html(mobj.group('err')), expected=True)
  92
  93         config_js_url = self._search_regex(
  94             r'src=(["\'])(?P<api_url>.+?/js/config\.js)(?:\?.+?)?\1', webpage, 'config_js',
  95             group='api_url', default=None)
  96
  97         if not config_js_url:
  98             # content is from ID, IN, MY
  99             video_info = self._download_json(
 100                 'https://www.viu.com/api/clip/load', video_id,
 101                 headers={'X-VIU-AUTH': self._get_viu_auth(video_id)},
 102                 query={'appid': 'viu_desktop', 'fmt': 'json', 'id': video_id},
 103                 note='Downloading video info').get('response', {}).get('item', [{}])[0]
 104
 105             formats = self._extract_m3u8_formats(
 106                 video_info['href'], video_id, 'mp4',
 107                 m3u8_id='hls', fatal=False)
 108             self._sort_formats(formats)
 109
 110             subtitles = {}
 111             for key, value in list(video_info.items()):
 112                 mobj = re.match(r'^subtitle_(?P<lang>[^_]+?)_(?P<ext>(vtt|srt))', key)
 113                 if not mobj:
 114                     continue
 115                 if not subtitles.get(mobj.group('lang')):
 116                     subtitles[mobj.group('lang')] = []
 117                 subtitles[mobj.group('lang')].append(
 118                     {'url': value, 'ext': mobj.group('ext')})
 119
 120             title = '%s - Episode %s' % (video_info['moviealbumshowname'],
 121                                          video_info.get('episodeno'))
 122             description = video_info.get('description')
 123             duration = int_or_none(video_info.get('duration'))
 124             series = video_info.get('moviealbumshowname')
 125             episode_title = video_info.get('title')
 126             episode_num = int_or_none(video_info.get('episodeno'))
 127
 128             return {
 129                 'id': video_id,
 130                 'title': title,
 131                 'description': description,
 132                 'series': series,
 133                 'episode': episode_title,
 134                 'episode_number': episode_num,
 135                 'duration': duration,
 136                 'formats': formats,
 137                 'subtitles': subtitles,
 138             }
 139
 140         # content from HK, SG
 141         config_js = self._download_webpage(
 142             'http://www.viu.com' + config_js_url, video_id, note='Downloading config js')
 143
 144         # try to strip away commented code which contains test urls
 145         config_js = re.sub(r'^//.*?$', '', config_js, flags=re.MULTILINE)
 146         config_js = re.sub(r'/\*.*?\*/', '', config_js, flags=re.DOTALL)
 147
 148         # Slightly different api_url between HK and SG config.js
 149         # http://www.viu.com/ott/hk/v1/js/config.js =>  '//www.viu.com/ott/hk/index.php?r='
 150         # http://www.viu.com/ott/sg/v1/js/config.js => 'http://www.viu.com/ott/sg/index.php?r='
 151         api_url = self._proto_relative_url(
 152             self._search_regex(
 153                 r'var\s+api_url\s*=\s*(["\'])(?P<api_url>(?:https?:)?//.+?\?r=)\1',
 154                 config_js, 'api_url', group='api_url'), scheme='http:')
 155
 156         stream_info_url = self._proto_relative_url(
 157             self._search_regex(
 158                 r'var\s+video_url\s*=\s*(["\'])(?P<video_url>(?:https?:)?//.+?\?ccs_product_id=)\1',
 159                 config_js, 'video_url', group='video_url'), scheme='http:')
 160
 161         if url.startswith('https://'):
 162             api_url = re.sub('^http://', 'https://', api_url)
 163
 164         video_info = self._download_json(
 165             api_url + 'vod/ajax-detail&platform_flag_label=web&product_id=' + video_id,
 166             video_id, note='Downloading video info').get('data', {})
 167
 168         ccs_product_id = video_info.get('current_product', {}).get('ccs_product_id')
 169
 170         if not ccs_product_id:
 171             raise ExtractorError('This video is not available in your region.', expected=True)
 172
 173         stream_info = self._download_json(
 174             stream_info_url + ccs_product_id, video_id,
 175             note='Downloading stream info').get('data', {}).get('stream', {})
 176
 177         formats = []
 178         for vid_format, stream_url in stream_info.get('url', {}).items():
 179             br = int_or_none(self._search_regex(
 180                 r's(?P<br>[0-9]+)p', vid_format, 'bitrate', group='br'))
 181             formats.append({
 182                 'format_id': vid_format,
 183                 'url': stream_url,
 184                 'vbr': br,
 185                 'ext': 'mp4',
 186                 'filesize': stream_info.get('size', {}).get(vid_format)
 187             })
 188         self._sort_formats(formats)
 189
 190         subtitles = {}
 191         if video_info.get('current_product', {}).get('subtitle', []):
 192             for sub in video_info.get('current_product', {}).get('subtitle', []):
 193                 subtitles[sub.get('name')] = [{
 194                     'url': sub.get('url'),
 195                     'ext': 'srt',
 196                 }]
 197
 198         episode_info = next(
 199             p for p in video_info.get('series', {}).get('product', [])
 200             if p.get('product_id') == video_id)
 201
 202         title = '%s - Episode %s' % (video_info.get('series', {}).get('name'),
 203                                      episode_info.get('number'))
 204         description = episode_info.get('description')
 205         thumbnail = episode_info.get('cover_image_url')
 206         duration = int_or_none(stream_info.get('duration'))
 207         series = video_info.get('series', {}).get('name')
 208         episode_title = episode_info.get('synopsis')
 209         episode_num = int_or_none(episode_info.get('number'))
 210
 211         return {
 212             'id': video_id,
 213             'title': title,
 214             'description': description,
 215             'series': series,
 216             'episode': episode_title,
 217             'episode_number': episode_num,
 218             'duration': duration,
 219             'thumbnail': thumbnail,
 220             'formats': formats,
 221             'subtitles': subtitles,
 222         }
 223
 224
 225 class ViuPlaylistIE(ViuBaseIE):
 226     IE_NAME = 'viu:playlist'
 227     _VALID_URL = r'https?://www\.viu\.com/.+/listing/(?P<id>playlist\-[0-9]+)'
 228     _TEST = {
 229         'url': 'https://www.viu.com/en/listing/playlist-22461380',
 230         'info_dict': {
 231             'id': 'playlist-22461380',
 232             'title': 'The Good Wife',
 233         },
 234         'playlist_count': 16,
 235         'skip': 'Geo-restricted to Indonesia',
 236     }
 237
 238     def _real_extract(self, url):
 239         playlist_id = self._match_id(url)
 240         playlist_info = self._download_json(
 241             'https://www.viu.com/api/container/load', playlist_id,
 242             headers={'X-VIU-AUTH': self._get_viu_auth(playlist_id)},
 243             query={'appid': 'viu_desktop', 'fmt': 'json', 'id': playlist_id},
 244             note='Downloading playlist info').get('response', {}).get('container')
 245
 246         name = playlist_info['title']
 247         entries = [
 248             self.url_result(
 249                 'https://www.viu.com/en/media/%s' % item['id'],
 250                 'Viu', item['id'])
 251             for item in playlist_info['item'] if item['id']]
 252
 253         return self.playlist_result(entries, playlist_id, name)