Raphaël G. Git Repositories - youtubedl/blob - 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 .common import InfoExtractor
  11 from ..utils import (
  12     ExtractorError,
  13     int_or_none,
  14     parse_age_limit,
  15     parse_iso8601,
  16     sanitized_Request,
  17 )
  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     _ERRORS = {
  34         'geo': 'Sorry, this content is not available in your region.',
  35         'upcoming': 'Sorry, this content is not yet available.',
  36         # 'paywall': 'paywall',
  37     }
  38
  39     def _prepare_call(self, path, timestamp=None, post_data=None):
  40         path += '?' if '?' not in path else '&'
  41         if not timestamp:
  42             timestamp = int(time.time())
  43         query = self._API_QUERY_TEMPLATE % (path, self._APP, timestamp)
  44         if self._token:
  45             query += '&token=%s' % self._token
  46         sig = hmac.new(
  47             self._APP_SECRET.encode('ascii'),
  48             query.encode('ascii'),
  49             hashlib.sha1
  50         ).hexdigest()
  51         url = self._API_URL_TEMPLATE % (query, sig)
  52         return sanitized_Request(
  53             url, json.dumps(post_data).encode('utf-8')) if post_data else url
  54
  55     def _call_api(self, path, video_id, note, timestamp=None, post_data=None):
  56         resp = self._download_json(
  57             self._prepare_call(path, timestamp, post_data), video_id, note)
  58
  59         error = resp.get('error')
  60         if error:
  61             if error == 'invalid timestamp':
  62                 resp = self._download_json(
  63                     self._prepare_call(path, int(resp['current_timestamp']), post_data),
  64                     video_id, '%s (retry)' % note)
  65                 error = resp.get('error')
  66             if error:
  67                 self._raise_error(resp['error'])
  68
  69         return resp
  70
  71     def _raise_error(self, error):
  72         raise ExtractorError(
  73             '%s returned error: %s' % (self.IE_NAME, error),
  74             expected=True)
  75
  76     def _check_errors(self, data):
  77         for reason, status in data.get('blocking', {}).items():
  78             if status and reason in self._ERRORS:
  79                 raise ExtractorError('%s said: %s' % (
  80                     self.IE_NAME, self._ERRORS[reason]), expected=True)
  81
  82     def _real_initialize(self):
  83         self._login()
  84
  85     def _login(self):
  86         (username, password) = self._get_login_info()
  87         if username is None:
  88             return
  89
  90         login_form = {
  91             'login_id': username,
  92             'password': password,
  93         }
  94
  95         login = self._call_api(
  96             'sessions.json', None,
  97             'Logging in as %s' % username, post_data=login_form)
  98
  99         self._token = login.get('token')
 100         if not self._token:
 101             self.report_warning('Unable to get session token, login has probably failed')
 102
 103     @staticmethod
 104     def dict_selection(dict_obj, preferred_key, allow_fallback=True):
 105         if preferred_key in dict_obj:
 106             return dict_obj.get(preferred_key)
 107
 108         if not allow_fallback:
 109             return
 110
 111         filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()]))
 112         return filtered_dict[0] if filtered_dict else None
 113
 114
 115 class VikiIE(VikiBaseIE):
 116     IE_NAME = 'viki'
 117     _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
 118     _TESTS = [{
 119         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
 120         'info_dict': {
 121             'id': '1023585v',
 122             'ext': 'mp4',
 123             'title': 'Heirs Episode 14',
 124             'uploader': 'SBS',
 125             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
 126             'upload_date': '20131121',
 127             'age_limit': 13,
 128         },
 129         'skip': 'Blocked in the US',
 130     }, {
 131         # clip
 132         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
 133         'md5': 'feea2b1d7b3957f70886e6dfd8b8be84',
 134         'info_dict': {
 135             'id': '1067139v',
 136             'ext': 'mp4',
 137             'title': "'The Avengers: Age of Ultron' Press Conference",
 138             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
 139             'duration': 352,
 140             'timestamp': 1430380829,
 141             'upload_date': '20150430',
 142             'uploader': 'Arirang TV',
 143             'like_count': int,
 144             'age_limit': 0,
 145         }
 146     }, {
 147         'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
 148         'info_dict': {
 149             'id': '1048879v',
 150             'ext': 'mp4',
 151             'title': 'Ankhon Dekhi',
 152             'duration': 6512,
 153             'timestamp': 1408532356,
 154             'upload_date': '20140820',
 155             'uploader': 'Spuul',
 156             'like_count': int,
 157             'age_limit': 13,
 158         },
 159         'params': {
 160             # m3u8 download
 161             'skip_download': True,
 162         },
 163         'skip': 'Blocked in the US',
 164     }, {
 165         # episode
 166         'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1',
 167         'md5': '1f54697dabc8f13f31bf06bb2e4de6db',
 168         'info_dict': {
 169             'id': '44699v',
 170             'ext': 'mp4',
 171             'title': 'Boys Over Flowers - Episode 1',
 172             'description': 'md5:b89cf50038b480b88b5b3c93589a9076',
 173             'duration': 4204,
 174             'timestamp': 1270496524,
 175             'upload_date': '20100405',
 176             'uploader': 'group8',
 177             'like_count': int,
 178             'age_limit': 13,
 179         }
 180     }, {
 181         # youtube external
 182         'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
 183         'md5': '63f8600c1da6f01b7640eee7eca4f1da',
 184         'info_dict': {
 185             'id': '50562v',
 186             'ext': 'webm',
 187             'title': 'Poor Nastya [COMPLETE] - Episode 1',
 188             'description': '',
 189             'duration': 606,
 190             'timestamp': 1274949505,
 191             'upload_date': '20101213',
 192             'uploader': 'ad14065n',
 193             'uploader_id': 'ad14065n',
 194             'like_count': int,
 195             'age_limit': 13,
 196         }
 197     }, {
 198         'url': 'http://www.viki.com/player/44699v',
 199         'only_matching': True,
 200     }, {
 201         # non-English description
 202         'url': 'http://www.viki.com/videos/158036v-love-in-magic',
 203         'md5': '013dc282714e22acf9447cad14ff1208',
 204         'info_dict': {
 205             'id': '158036v',
 206             'ext': 'mp4',
 207             'uploader': 'I Planet Entertainment',
 208             'upload_date': '20111122',
 209             'timestamp': 1321985454,
 210             'description': 'md5:44b1e46619df3a072294645c770cef36',
 211             'title': 'Love In Magic',
 212             'age_limit': 13,
 213         },
 214     }]
 215
 216     def _real_extract(self, url):
 217         video_id = self._match_id(url)
 218
 219         video = self._call_api(
 220             'videos/%s.json' % video_id, video_id, 'Downloading video JSON')
 221
 222         self._check_errors(video)
 223
 224         title = self.dict_selection(video.get('titles', {}), 'en', allow_fallback=False)
 225         if not title:
 226             title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id
 227             container_titles = video.get('container', {}).get('titles', {})
 228             container_title = self.dict_selection(container_titles, 'en')
 229             title = '%s - %s' % (container_title, title)
 230
 231         description = self.dict_selection(video.get('descriptions', {}), 'en')
 232
 233         duration = int_or_none(video.get('duration'))
 234         timestamp = parse_iso8601(video.get('created_at'))
 235         uploader = video.get('author')
 236         like_count = int_or_none(video.get('likes', {}).get('count'))
 237         age_limit = parse_age_limit(video.get('rating'))
 238
 239         thumbnails = []
 240         for thumbnail_id, thumbnail in video.get('images', {}).items():
 241             thumbnails.append({
 242                 'id': thumbnail_id,
 243                 'url': thumbnail.get('url'),
 244             })
 245
 246         subtitles = {}
 247         for subtitle_lang, _ in video.get('subtitle_completions', {}).items():
 248             subtitles[subtitle_lang] = [{
 249                 'ext': subtitles_format,
 250                 'url': self._prepare_call(
 251                     'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)),
 252             } for subtitles_format in ('srt', 'vtt')]
 253
 254         result = {
 255             'id': video_id,
 256             'title': title,
 257             'description': description,
 258             'duration': duration,
 259             'timestamp': timestamp,
 260             'uploader': uploader,
 261             'like_count': like_count,
 262             'age_limit': age_limit,
 263             'thumbnails': thumbnails,
 264             'subtitles': subtitles,
 265         }
 266
 267         streams = self._call_api(
 268             'videos/%s/streams.json' % video_id, video_id,
 269             'Downloading video streams JSON')
 270
 271         if 'external' in streams:
 272             result.update({
 273                 '_type': 'url_transparent',
 274                 'url': streams['external']['url'],
 275             })
 276             return result
 277
 278         formats = []
 279         for format_id, stream_dict in streams.items():
 280             height = int_or_none(self._search_regex(
 281                 r'^(\d+)[pP]$', format_id, 'height', default=None))
 282             for protocol, format_dict in stream_dict.items():
 283                 if format_id == 'm3u8':
 284                     formats.extend(self._extract_m3u8_formats(
 285                         format_dict['url'], video_id, 'mp4', 'm3u8_native',
 286                         m3u8_id='m3u8-%s' % protocol, fatal=False))
 287                 else:
 288                     formats.append({
 289                         'url': format_dict['url'],
 290                         'format_id': '%s-%s' % (format_id, protocol),
 291                         'height': height,
 292                     })
 293         self._sort_formats(formats)
 294
 295         result['formats'] = formats
 296         return result
 297
 298
 299 class VikiChannelIE(VikiBaseIE):
 300     IE_NAME = 'viki:channel'
 301     _VALID_URL = r'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE._VALID_URL_BASE
 302     _TESTS = [{
 303         'url': 'http://www.viki.com/tv/50c-boys-over-flowers',
 304         'info_dict': {
 305             'id': '50c',
 306             'title': 'Boys Over Flowers',
 307             'description': 'md5:ecd3cff47967fe193cff37c0bec52790',
 308         },
 309         'playlist_mincount': 71,
 310     }, {
 311         'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
 312         'info_dict': {
 313             'id': '1354c',
 314             'title': 'Poor Nastya [COMPLETE]',
 315             'description': 'md5:05bf5471385aa8b21c18ad450e350525',
 316         },
 317         'playlist_count': 127,
 318     }, {
 319         'url': 'http://www.viki.com/news/24569c-showbiz-korea',
 320         'only_matching': True,
 321     }, {
 322         'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005',
 323         'only_matching': True,
 324     }, {
 325         'url': 'http://www.viki.com/artists/2141c-shinee',
 326         'only_matching': True,
 327     }]
 328
 329     _PER_PAGE = 25
 330
 331     def _real_extract(self, url):
 332         channel_id = self._match_id(url)
 333
 334         channel = self._call_api(
 335             'containers/%s.json' % channel_id, channel_id,
 336             'Downloading channel JSON')
 337
 338         self._check_errors(channel)
 339
 340         title = self.dict_selection(channel['titles'], 'en')
 341
 342         description = self.dict_selection(channel['descriptions'], 'en')
 343
 344         entries = []
 345         for video_type in ('episodes', 'clips', 'movies'):
 346             for page_num in itertools.count(1):
 347                 page = self._call_api(
 348                     'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d'
 349                     % (channel_id, video_type, self._PER_PAGE, page_num), channel_id,
 350                     'Downloading %s JSON page #%d' % (video_type, page_num))
 351                 for video in page['response']:
 352                     video_id = video['id']
 353                     entries.append(self.url_result(
 354                         'http://www.viki.com/videos/%s' % video_id, 'Viki'))
 355                 if not page['pagination']['next']:
 356                     break
 357
 358         return self.playlist_result(entries, channel_id, title, description)