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     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 sanitized_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     @staticmethod
  92     def dict_selection(dict_obj, preferred_key):
  93         if preferred_key in dict_obj:
  94             return dict_obj.get(preferred_key)
  95
  96         filtered_dict = list(filter(None, [dict_obj.get(k) for k in dict_obj.keys()]))
  97         return filtered_dict[0] if filtered_dict else None
  98
  99
 100 class VikiIE(VikiBaseIE):
 101     IE_NAME = 'viki'
 102     _VALID_URL = r'%s(?:videos|player)/(?P<id>[0-9]+v)' % VikiBaseIE._VALID_URL_BASE
 103     _TESTS = [{
 104         'url': 'http://www.viki.com/videos/1023585v-heirs-episode-14',
 105         'info_dict': {
 106             'id': '1023585v',
 107             'ext': 'mp4',
 108             'title': 'Heirs Episode 14',
 109             'uploader': 'SBS',
 110             'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e',
 111             'upload_date': '20131121',
 112             'age_limit': 13,
 113         },
 114         'skip': 'Blocked in the US',
 115     }, {
 116         # clip
 117         'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference',
 118         'md5': '86c0b5dbd4d83a6611a79987cc7a1989',
 119         'info_dict': {
 120             'id': '1067139v',
 121             'ext': 'mp4',
 122             'title': "'The Avengers: Age of Ultron' Press Conference",
 123             'description': 'md5:d70b2f9428f5488321bfe1db10d612ea',
 124             'duration': 352,
 125             'timestamp': 1430380829,
 126             'upload_date': '20150430',
 127             'uploader': 'Arirang TV',
 128             'like_count': int,
 129             'age_limit': 0,
 130         }
 131     }, {
 132         'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi',
 133         'info_dict': {
 134             'id': '1048879v',
 135             'ext': 'mp4',
 136             'title': 'Ankhon Dekhi',
 137             'duration': 6512,
 138             'timestamp': 1408532356,
 139             'upload_date': '20140820',
 140             'uploader': 'Spuul',
 141             'like_count': int,
 142             'age_limit': 13,
 143         },
 144         'params': {
 145             # m3u8 download
 146             'skip_download': True,
 147         }
 148     }, {
 149         # episode
 150         'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1',
 151         'md5': '190f3ef426005ba3a080a63325955bc3',
 152         'info_dict': {
 153             'id': '44699v',
 154             'ext': 'mp4',
 155             'title': 'Boys Over Flowers - Episode 1',
 156             'description': 'md5:52617e4f729c7d03bfd4bcbbb6e946f2',
 157             'duration': 4155,
 158             'timestamp': 1270496524,
 159             'upload_date': '20100405',
 160             'uploader': 'group8',
 161             'like_count': int,
 162             'age_limit': 13,
 163         }
 164     }, {
 165         # youtube external
 166         'url': 'http://www.viki.com/videos/50562v-poor-nastya-complete-episode-1',
 167         'md5': '216d1afdc0c64d1febc1e9f2bd4b864b',
 168         'info_dict': {
 169             'id': '50562v',
 170             'ext': 'mp4',
 171             'title': 'Poor Nastya [COMPLETE] - Episode 1',
 172             'description': '',
 173             'duration': 607,
 174             'timestamp': 1274949505,
 175             'upload_date': '20101213',
 176             'uploader': 'ad14065n',
 177             'uploader_id': 'ad14065n',
 178             'like_count': int,
 179             'age_limit': 13,
 180         }
 181     }, {
 182         'url': 'http://www.viki.com/player/44699v',
 183         'only_matching': True,
 184     }, {
 185         # non-English description
 186         'url': 'http://www.viki.com/videos/158036v-love-in-magic',
 187         'md5': '1713ae35df5a521b31f6dc40730e7c9c',
 188         'info_dict': {
 189             'id': '158036v',
 190             'ext': 'mp4',
 191             'uploader': 'I Planet Entertainment',
 192             'upload_date': '20111122',
 193             'timestamp': 1321985454,
 194             'description': 'md5:44b1e46619df3a072294645c770cef36',
 195             'title': 'Love In Magic',
 196         },
 197     }]
 198
 199     def _real_extract(self, url):
 200         video_id = self._match_id(url)
 201
 202         video = self._call_api(
 203             'videos/%s.json' % video_id, video_id, 'Downloading video JSON')
 204
 205         title = self.dict_selection(video.get('titles', {}), 'en')
 206         if not title:
 207             title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id
 208             container_titles = video.get('container', {}).get('titles', {})
 209             container_title = self.dict_selection(container_titles, 'en')
 210             title = '%s - %s' % (container_title, title)
 211
 212         description = self.dict_selection(video.get('descriptions', {}), 'en')
 213
 214         duration = int_or_none(video.get('duration'))
 215         timestamp = parse_iso8601(video.get('created_at'))
 216         uploader = video.get('author')
 217         like_count = int_or_none(video.get('likes', {}).get('count'))
 218         age_limit = parse_age_limit(video.get('rating'))
 219
 220         thumbnails = []
 221         for thumbnail_id, thumbnail in video.get('images', {}).items():
 222             thumbnails.append({
 223                 'id': thumbnail_id,
 224                 'url': thumbnail.get('url'),
 225             })
 226
 227         subtitles = {}
 228         for subtitle_lang, _ in video.get('subtitle_completions', {}).items():
 229             subtitles[subtitle_lang] = [{
 230                 'ext': subtitles_format,
 231                 'url': self._prepare_call(
 232                     'videos/%s/subtitles/%s.%s' % (video_id, subtitle_lang, subtitles_format)),
 233             } for subtitles_format in ('srt', 'vtt')]
 234
 235         result = {
 236             'id': video_id,
 237             'title': title,
 238             'description': description,
 239             'duration': duration,
 240             'timestamp': timestamp,
 241             'uploader': uploader,
 242             'like_count': like_count,
 243             'age_limit': age_limit,
 244             'thumbnails': thumbnails,
 245             'subtitles': subtitles,
 246         }
 247
 248         streams = self._call_api(
 249             'videos/%s/streams.json' % video_id, video_id,
 250             'Downloading video streams JSON')
 251
 252         if 'external' in streams:
 253             result.update({
 254                 '_type': 'url_transparent',
 255                 'url': streams['external']['url'],
 256             })
 257             return result
 258
 259         formats = []
 260         for format_id, stream_dict in streams.items():
 261             height = int_or_none(self._search_regex(
 262                 r'^(\d+)[pP]$', format_id, 'height', default=None))
 263             for protocol, format_dict in stream_dict.items():
 264                 if format_id == 'm3u8':
 265                     formats = self._extract_m3u8_formats(
 266                         format_dict['url'], video_id, 'mp4', m3u8_id='m3u8-%s' % protocol)
 267                 else:
 268                     formats.append({
 269                         'url': format_dict['url'],
 270                         'format_id': '%s-%s' % (format_id, protocol),
 271                         'height': height,
 272                     })
 273         self._sort_formats(formats)
 274
 275         result['formats'] = formats
 276         return result
 277
 278
 279 class VikiChannelIE(VikiBaseIE):
 280     IE_NAME = 'viki:channel'
 281     _VALID_URL = r'%s(?:tv|news|movies|artists)/(?P<id>[0-9]+c)' % VikiBaseIE._VALID_URL_BASE
 282     _TESTS = [{
 283         'url': 'http://www.viki.com/tv/50c-boys-over-flowers',
 284         'info_dict': {
 285             'id': '50c',
 286             'title': 'Boys Over Flowers',
 287             'description': 'md5:ecd3cff47967fe193cff37c0bec52790',
 288         },
 289         'playlist_count': 70,
 290     }, {
 291         'url': 'http://www.viki.com/tv/1354c-poor-nastya-complete',
 292         'info_dict': {
 293             'id': '1354c',
 294             'title': 'Poor Nastya [COMPLETE]',
 295             'description': 'md5:05bf5471385aa8b21c18ad450e350525',
 296         },
 297         'playlist_count': 127,
 298     }, {
 299         'url': 'http://www.viki.com/news/24569c-showbiz-korea',
 300         'only_matching': True,
 301     }, {
 302         'url': 'http://www.viki.com/movies/22047c-pride-and-prejudice-2005',
 303         'only_matching': True,
 304     }, {
 305         'url': 'http://www.viki.com/artists/2141c-shinee',
 306         'only_matching': True,
 307     }]
 308
 309     _PER_PAGE = 25
 310
 311     def _real_extract(self, url):
 312         channel_id = self._match_id(url)
 313
 314         channel = self._call_api(
 315             'containers/%s.json' % channel_id, channel_id,
 316             'Downloading channel JSON')
 317
 318         title = self.dict_selection(channel['titles'], 'en')
 319
 320         description = self.dict_selection(channel['descriptions'], 'en')
 321
 322         entries = []
 323         for video_type in ('episodes', 'clips', 'movies'):
 324             for page_num in itertools.count(1):
 325                 page = self._call_api(
 326                     'containers/%s/%s.json?per_page=%d&sort=number&direction=asc&with_paging=true&page=%d'
 327                     % (channel_id, video_type, self._PER_PAGE, page_num), channel_id,
 328                     'Downloading %s JSON page #%d' % (video_type, page_num))
 329                 for video in page['response']:
 330                     video_id = video['id']
 331                     entries.append(self.url_result(
 332                         'http://www.viki.com/videos/%s' % video_id, 'Viki'))
 333                 if not page['pagination']['next']:
 334                     break
 335
 336         return self.playlist_result(entries, channel_id, title, description)