youtube_dl/extractor/viki.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import time
   6 import hmac
   7 import hashlib
   8 import itertools
   9
  10 from ..utils import (
  11     ExtractorError,
  12     int_or_none,
  13     parse_age_limit,
  14     parse_iso8601,
  15 )
  16 from ..compat import compat_urllib_request
  17 from .common import InfoExtractor
  18
  19
  20 class VikiBaseIE(InfoExtractor):
  21     _VALID_URL_BASE = r'https?://(?:www\.)?viki\.(?:com|net|mx|jp|fr)/'
  22     _API_QUERY_TEMPLATE = '/v4/%sapp=%s&t=%s&site=www.viki.com'
  23     _API_URL_TEMPLATE = 'http://api.viki.io%s&sig=%s'
  24
  25     _APP = '65535a'
  26     _APP_VERSION = '2.2.5.1428709186'
  27     _APP_SECRET = '-$iJ}@p7!G@SyU/je1bEyWg}upLu-6V6-Lg9VD(]siH,r.,m-r|ulZ,U4LC/SeR)'
  28
  29     _NETRC_MACHINE = 'viki'
  30
  31     _token = None
  32
  33     def _prepare_call(self, path, timestamp=None, post_data=None):
  34         path += '?' if '?' not in path else '&'
  35         if not timestamp:
  36             timestamp = int(time.time())
  37         query = self._API_QUERY_TEMPLATE % (path, self._APP, timestamp)
  38         if self._token:
  39             query += '&token=%s' % self._token
  40         sig = hmac.new(
  41             self._APP_SECRET.encode('ascii'),
  42             query.encode('ascii'),
  43             hashlib.sha1
  44         ).hexdigest()
  45         url = self._API_URL_TEMPLATE % (query, sig)
  46         return compat_urllib_request.Request(
  47             url, json.dumps(post_data).encode('utf-8')) if post_data else url
  48
  49     def _call_api(self, path, video_id, note, timestamp=None, post_data=None):
  50         resp = self._download_json(
  51             self._prepare_call(path, timestamp, post_data), video_id, note)
  52
  53         error = resp.get('error')
  54         if error:
  55             if error == 'invalid timestamp':
  56                 resp = self._download_json(
  57                     self._prepare_call(path, int(resp['current_timestamp']), post_data),
  58                     video_id, '%s (retry)' % note)
  59                 error = resp.get('error')
  60             if error:
  61                 self._raise_error(resp['error'])
  62
  63         return resp
  64
  65     def _raise_error(self, error):
  66         raise ExtractorError(
  67             '%s returned error: %s' % (self.IE_NAME, error),
  68             expected=True)
  69
  70     def _real_initialize(self):
  71         self._login()
  72
  73     def _login(self):
  74         (username, password) = self._get_login_info()
  75         if username is None:
  76             return
  77
  78         login_form = {
  79             'login_id': username,
  80             'password': password,
  81         }
  82
  83         login = self._call_api(
  84             'sessions.json', None,
  85             'Logging in as %s' % username, post_data=login_form)
  86
  87         self._token = login.get('token')
  88         if not self._token:
  89             self.report_warning('Unable to get session token, login has probably failed')
  90
  91
  92 class VikiIE(VikiBaseIE):
  93     IE_NAME = 'viki'
  94     _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
  95     _TESTS = [{
  96         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
  97         'info_dict': {
  98             'id': '1023585v',
  99             'ext': 'mp4',
 100             'title': 'Heirs Episode 14',
 101             'uploader': 'SBS',
 102             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
 103             'upload_date': '20131121',
 104             'age_limit': 13,
 105         },
 106         'skip': 'Blocked in the US',
 107     }, {
 108         # clip
 109         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
 110         'md5': '86c0b5dbd4d83a6611a79987cc7a1989',
 111         'info_dict': {
 112             'id': '1067139v',
 113             'ext': 'mp4',
 114             'title': "'The Avengers: Age of Ultron' Press Conference",
 115             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
 116             'duration': 352,
 117             'timestamp': 1430380829,
 118             'upload_date': '20150430',
 119             'uploader': 'Arirang TV',
 120             'like_count': int,
 121             'age_limit': 0,
 122         }
 123     }, {
 124         'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
 125         'info_dict': {
 126             'id': '1048879v',
 127             'ext': 'mp4',
 128             'title': 'Ankhon Dekhi',
 129             'duration': 6512,
 130             'timestamp': 1408532356,
 131             'upload_date': '20140820',
 132             'uploader': 'Spuul',
 133             'like_count': int,
 134             'age_limit': 13,
 135         },
 136         'params': {
 137             # m3u8 download
 138             'skip_download': True,
 139         }
 140     }, {
 141         # episode
 142         'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1',
 143         'md5': '190f3ef426005ba3a080a63325955bc3',
 144         'info_dict': {
 145             'id': '44699v',
 146             'ext': 'mp4',
 147             'title': 'Boys Over Flowers - Episode 1',
 148             'description': 'md5:52617e4f729c7d03bfd4bcbbb6e946f2',
 149             'duration': 4155,
 150             'timestamp': 1270496524,
 151             'upload_date': '20100405',
 152             'uploader': 'group8',
 153             'like_count': int,
 154             'age_limit': 13,
 155         }
 156     }, {
 157         # youtube external
 158         'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
 159         'md5': '216d1afdc0c64d1febc1e9f2bd4b864b',
 160         'info_dict': {
 161             'id': '50562v',
 162             'ext': 'mp4',
 163             'title': 'Poor Nastya [COMPLETE] - Episode 1',
 164             'description': '',
 165             'duration': 607,
 166             'timestamp': 1274949505,
 167             'upload_date': '20101213',
 168             'uploader': 'ad14065n',
 169             'uploader_id': 'ad14065n',
 170             'like_count': int,
 171             'age_limit': 13,
 172         }
 173     }, {
 174         'url': 'http://www.viki.com/player/44699v',
 175         'only_matching': True,
 176     }, {
 177         # non-English description
 178         'url': 'http://www.viki.com/videos/158036v-love-in-magic',
 179         'md5': '1713ae35df5a521b31f6dc40730e7c9c',
 180         'info_dict': {
 181             'id': '158036v',
 182             'ext': 'mp4',
 183             'uploader': 'I Planet Entertainment',
 184             'upload_date': '20111122',
 185             'timestamp': 1321985454,
 186             'description': 'md5:44b1e46619df3a072294645c770cef36',
 187             'title': 'Love In Magic',
 188         },
 189     }]
 190
 191     def _real_extract(self, url):
 192         video_id = self._match_id(url)
 193
 194         video = self._call_api(
 195             'videos/%s.json' % video_id, video_id, 'Downloading video JSON')
 196
 197         title = None
 198         titles = video.get('titles')
 199         if titles:
 200             title = titles.get('en') or titles[titles.keys()[0]]
 201         if not title:
 202             title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id
 203             container_titles = video.get('container', {}).get('titles')
 204             if container_titles:
 205                 container_title = container_titles.get('en') or container_titles[container_titles.keys()[0]]
 206                 title = '%s - %s' % (container_title, title)
 207
 208         descriptions = video.get('descriptions', {})
 209         description = descriptions.get('en')
 210         if description is None:
 211             filtered_descriptions = list(filter(None, [descriptions.get(k) for k in titles.keys()]))
 212             if filtered_descriptions:
 213                 description = filtered_descriptions[0]
 214
 215         duration = int_or_none(video.get('duration'))
 216         timestamp = parse_iso8601(video.get('created_at'))
 217         uploader = video.get('author')
 218         like_count = int_or_none(video.get('likes', {}).get('count'))
 219         age_limit = parse_age_limit(video.get('rating'))
 220
 221         thumbnails = []
 222         for thumbnail_id, thumbnail in video.get('images', {}).items():
 223             thumbnails.append({
 224                 'id': thumbnail_id,
 225                 'url': thumbnail.get('url'),
 226             })
 227
 228         subtitles = {}
 229         for subtitle_lang, _ in video.get('subtitle_completions', {}).items():
 230             subtitles[subtitle_lang] = [{
 231                 'ext': subtitles_format,
 232                 'url': self._prepare_call(
 233                     'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)),
 234             } for subtitles_format in ('srt', 'vtt')]
 235
 236         result = {
 237             'id': video_id,
 238             'title': title,
 239             'description': description,
 240             'duration': duration,
 241             'timestamp': timestamp,
 242             'uploader': uploader,
 243             'like_count': like_count,
 244             'age_limit': age_limit,
 245             'thumbnails': thumbnails,
 246             'subtitles': subtitles,
 247         }
 248
 249         streams = self._call_api(
 250             'videos/%s/streams.json' % video_id, video_id,
 251             'Downloading video streams JSON')
 252
 253         if 'external' in streams:
 254             result.update({
 255                 '_type': 'url_transparent',
 256                 'url': streams['external']['url'],
 257             })
 258             return result
 259
 260         formats = []
 261         for format_id, stream_dict in streams.items():
 262             height = int_or_none(self._search_regex(
 263                 r'^(\d+)[pP]$', format_id, 'height', default=None))
 264             for protocol, format_dict in stream_dict.items():
 265                 if format_id == 'm3u8':
 266                     formats = self._extract_m3u8_formats(
 267                         format_dict['url'], video_id, 'mp4', m3u8_id='m3u8-%s' % protocol)
 268                 else:
 269                     formats.append({
 270                         'url': format_dict['url'],
 271                         'format_id': '%s-%s' % (format_id, protocol),
 272                         'height': height,
 273                     })
 274         self._sort_formats(formats)
 275
 276         result['formats'] = formats
 277         return result
 278
 279
 280 class VikiChannelIE(VikiBaseIE):
 281     IE_NAME = 'viki:channel'
 282     _VALID_URL = r'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE._VALID_URL_BASE
 283     _TESTS = [{
 284         'url': 'http://www.viki.com/tv/50c-boys-over-flowers',
 285         'info_dict': {
 286             'id': '50c',
 287             'title': 'Boys Over Flowers',
 288             'description': 'md5:ecd3cff47967fe193cff37c0bec52790',
 289         },
 290         'playlist_count': 70,
 291     }, {
 292         'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
 293         'info_dict': {
 294             'id': '1354c',
 295             'title': 'Poor Nastya [COMPLETE]',
 296             'description': 'md5:05bf5471385aa8b21c18ad450e350525',
 297         },
 298         'playlist_count': 127,
 299     }, {
 300         'url': 'http://www.viki.com/news/24569c-showbiz-korea',
 301         'only_matching': True,
 302     }, {
 303         'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005',
 304         'only_matching': True,
 305     }, {
 306         'url': 'http://www.viki.com/artists/2141c-shinee',
 307         'only_matching': True,
 308     }]
 309
 310     _PER_PAGE = 25
 311
 312     def _real_extract(self, url):
 313         channel_id = self._match_id(url)
 314
 315         channel = self._call_api(
 316             'containers/%s.json' % channel_id, channel_id,
 317             'Downloading channel JSON')
 318
 319         titles = channel['titles']
 320         title = titles.get('en') or titles[titles.keys()[0]]
 321
 322         descriptions = channel['descriptions']
 323         description = descriptions.get('en') or descriptions[descriptions.keys()[0]]
 324
 325         entries = []
 326         for video_type in ('episodes', 'clips', 'movies'):
 327             for page_num in itertools.count(1):
 328                 page = self._call_api(
 329                     'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d'
 330                     % (channel_id, video_type, self._PER_PAGE, page_num), channel_id,
 331                     'Downloading %s JSON page #%d' % (video_type, page_num))
 332                 for video in page['response']:
 333                     video_id = video['id']
 334                     entries.append(self.url_result(
 335                         'http://www.viki.com/videos/%s' % video_id, 'Viki'))
 336                 if not page['pagination']['next']:
 337                     break
 338
 339         return self.playlist_result(entries, channel_id, title, description)