yt_dlp/extractor/viu.py

   1 import json
   2 import random
   3 import re
   4 import urllib.parse
   5 import uuid
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     ExtractorError,
  10     int_or_none,
  11     remove_end,
  12     smuggle_url,
  13     strip_or_none,
  14     traverse_obj,
  15     try_get,
  16     unified_timestamp,
  17     unsmuggle_url,
  18     url_or_none,
  19 )
  20
  21
  22 class ViuBaseIE(InfoExtractor):
  23     def _call_api(self, path, *args, headers={}, **kwargs):
  24         response = self._download_json(
  25             f'https://www.viu.com/api/{path}', *args, **kwargs,
  26             headers={**self.geo_verification_headers(), **headers})['response']
  27         if response.get('status') != 'success':
  28             raise ExtractorError(f'{self.IE_NAME} said: {response["message"]}', expected=True)
  29         return response
  30
  31
  32 class ViuIE(ViuBaseIE):
  33     _VALID_URL = r'(?:viu:|https?://[^/]+\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
  34     _TESTS = [{
  35         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
  36         'info_dict': {
  37             'id': '1116705532',
  38             'ext': 'mp4',
  39             'title': 'Citizen Khan - Ep 1',
  40             'description': 'md5:d7ea1604f49e5ba79c212c551ce2110e',
  41         },
  42         'params': {
  43             'skip_download': 'm3u8 download',
  44         },
  45         'skip': 'Geo-restricted to India',
  46     }, {
  47         'url': 'https://www.viu.com/en/media/1130599965',
  48         'info_dict': {
  49             'id': '1130599965',
  50             'ext': 'mp4',
  51             'title': 'Jealousy Incarnate - Episode 1',
  52             'description': 'md5:d3d82375cab969415d2720b6894361e9',
  53         },
  54         'params': {
  55             'skip_download': 'm3u8 download',
  56         },
  57         'skip': 'Geo-restricted to Indonesia',
  58     }, {
  59         'url': 'https://india.viu.com/en/media/1126286865',
  60         'only_matching': True,
  61     }]
  62
  63     def _real_extract(self, url):
  64         video_id = self._match_id(url)
  65
  66         video_data = self._call_api(
  67             'clip/load', video_id, 'Downloading video data', query={
  68                 'appid': 'viu_desktop',
  69                 'fmt': 'json',
  70                 'id': video_id,
  71             })['item'][0]
  72
  73         title = video_data['title']
  74
  75         m3u8_url = None
  76         url_path = video_data.get('urlpathd') or video_data.get('urlpath')
  77         tdirforwhole = video_data.get('tdirforwhole')
  78         # #EXT-X-BYTERANGE is not supported by native hls downloader
  79         # and ffmpeg (#10955)
  80         # FIXME: It is supported in yt-dlp
  81         # hls_file = video_data.get('hlsfile')
  82         hls_file = video_data.get('jwhlsfile')
  83         if url_path and tdirforwhole and hls_file:
  84             m3u8_url = f'{url_path}/{tdirforwhole}/{hls_file}'
  85         else:
  86             # m3u8_url = re.sub(
  87             #     r'(/hlsc_)[a-z]+(\d+\.m3u8)',
  88             #     r'\1whe\2', video_data['href'])
  89             m3u8_url = video_data['href']
  90         formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4')
  91
  92         for key, value in video_data.items():
  93             mobj = re.match(r'subtitle_(?P<lang>[^_]+)_(?P<ext>(vtt|srt))', key)
  94             if not mobj:
  95                 continue
  96             subtitles.setdefault(mobj.group('lang'), []).append({
  97                 'url': value,
  98                 'ext': mobj.group('ext'),
  99             })
 100
 101         return {
 102             'id': video_id,
 103             'title': title,
 104             'description': video_data.get('description'),
 105             'series': video_data.get('moviealbumshowname'),
 106             'episode': title,
 107             'episode_number': int_or_none(video_data.get('episodeno')),
 108             'duration': int_or_none(video_data.get('duration')),
 109             'formats': formats,
 110             'subtitles': subtitles,
 111         }
 112
 113
 114 class ViuPlaylistIE(ViuBaseIE):
 115     IE_NAME = 'viu:playlist'
 116     _VALID_URL = r'https?://www\.viu\.com/[^/]+/listing/playlist-(?P<id>\d+)'
 117     _TEST = {
 118         'url': 'https://www.viu.com/en/listing/playlist-22461380',
 119         'info_dict': {
 120             'id': '22461380',
 121             'title': 'The Good Wife',
 122         },
 123         'playlist_count': 16,
 124         'skip': 'Geo-restricted to Indonesia',
 125     }
 126
 127     def _real_extract(self, url):
 128         playlist_id = self._match_id(url)
 129         playlist_data = self._call_api(
 130             'container/load', playlist_id,
 131             'Downloading playlist info', query={
 132                 'appid': 'viu_desktop',
 133                 'fmt': 'json',
 134                 'id': 'playlist-' + playlist_id,
 135             })['container']
 136
 137         entries = []
 138         for item in playlist_data.get('item', []):
 139             item_id = item.get('id')
 140             if not item_id:
 141                 continue
 142             item_id = str(item_id)
 143             entries.append(self.url_result(
 144                 'viu:' + item_id, 'Viu', item_id))
 145
 146         return self.playlist_result(
 147             entries, playlist_id, playlist_data.get('title'))
 148
 149
 150 class ViuOTTIE(InfoExtractor):
 151     IE_NAME = 'viu:ott'
 152     _NETRC_MACHINE = 'viu'
 153     _VALID_URL = r'https?://(?:www\.)?viu\.com/ott/(?P<country_code>[a-z]{2})/(?P<lang_code>[a-z]{2}-[a-z]{2})/vod/(?P<id>\d+)'
 154     _TESTS = [{
 155         'url': 'http://www.viu.com/ott/sg/en-us/vod/3421/The%20Prime%20Minister%20and%20I',
 156         'info_dict': {
 157             'id': '3421',
 158             'ext': 'mp4',
 159             'title': 'A New Beginning',
 160             'description': 'md5:1e7486a619b6399b25ba6a41c0fe5b2c',
 161         },
 162         'params': {
 163             'skip_download': 'm3u8 download',
 164             'noplaylist': True,
 165         },
 166         'skip': 'Geo-restricted to Singapore',
 167     }, {
 168         'url': 'https://www.viu.com/ott/hk/zh-hk/vod/430078/%E7%AC%AC%E5%85%AD%E6%84%9F-3',
 169         'info_dict': {
 170             'id': '430078',
 171             'ext': 'mp4',
 172             'title': '大韓民國的1%',
 173             'description': 'md5:74d6db47ddd9ddb9c89a05739103ccdb',
 174             'episode_number': 1,
 175             'duration': 6614,
 176             'episode': '大韓民國的1%',
 177             'series': '第六感 3',
 178             'thumbnail': 'https://d2anahhhmp1ffz.cloudfront.net/1313295781/d2b14f48d008ef2f3a9200c98d8e9b63967b9cc2',
 179         },
 180         'params': {
 181             'skip_download': 'm3u8 download',
 182             'noplaylist': True,
 183         },
 184         'skip': 'Geo-restricted to Hong Kong',
 185     }, {
 186         'url': 'https://www.viu.com/ott/hk/zh-hk/vod/444666/%E6%88%91%E7%9A%84%E5%AE%A4%E5%8F%8B%E6%98%AF%E4%B9%9D%E5%B0%BE%E7%8B%90',
 187         'playlist_count': 16,
 188         'info_dict': {
 189             'id': '23807',
 190             'title': '我的室友是九尾狐',
 191             'description': 'md5:b42c95f2b4a316cdd6ae14ca695f33b9',
 192         },
 193         'params': {
 194             'skip_download': 'm3u8 download',
 195             'noplaylist': False,
 196         },
 197         'skip': 'Geo-restricted to Hong Kong',
 198     }]
 199
 200     _AREA_ID = {
 201         'HK': 1,
 202         'SG': 2,
 203         'TH': 4,
 204         'PH': 5,
 205     }
 206     _LANGUAGE_FLAG = {
 207         'zh-hk': 1,
 208         'zh-cn': 2,
 209         'en-us': 3,
 210     }
 211
 212     _user_token = None
 213     _auth_codes = {}
 214
 215     def _detect_error(self, response):
 216         code = try_get(response, lambda x: x['status']['code'])
 217         if code and code > 0:
 218             message = try_get(response, lambda x: x['status']['message'])
 219             raise ExtractorError(f'{self.IE_NAME} said: {message} ({code})', expected=True)
 220         return response.get('data') or {}
 221
 222     def _login(self, country_code, video_id):
 223         if self._user_token is None:
 224             username, password = self._get_login_info()
 225             if username is None:
 226                 return
 227             headers = {
 228                 'Authorization': f'Bearer {self._auth_codes[country_code]}',
 229                 'Content-Type': 'application/json',
 230             }
 231             data = self._download_json(
 232                 'https://api-gateway-global.viu.com/api/account/validate',
 233                 video_id, 'Validating email address', headers=headers,
 234                 data=json.dumps({
 235                     'principal': username,
 236                     'provider': 'email',
 237                 }).encode())
 238             if not data.get('exists'):
 239                 raise ExtractorError('Invalid email address')
 240
 241             data = self._download_json(
 242                 'https://api-gateway-global.viu.com/api/auth/login',
 243                 video_id, 'Logging in', headers=headers,
 244                 data=json.dumps({
 245                     'email': username,
 246                     'password': password,
 247                     'provider': 'email',
 248                 }).encode())
 249             self._detect_error(data)
 250             self._user_token = data.get('identity')
 251             # need to update with valid user's token else will throw an error again
 252             self._auth_codes[country_code] = data.get('token')
 253         return self._user_token
 254
 255     def _get_token(self, country_code, video_id):
 256         rand = ''.join(random.choices('0123456789', k=10))
 257         return self._download_json(
 258             f'https://api-gateway-global.viu.com/api/auth/token?v={rand}000', video_id,
 259             headers={'Content-Type': 'application/json'}, note='Getting bearer token',
 260             data=json.dumps({
 261                 'countryCode': country_code.upper(),
 262                 'platform': 'browser',
 263                 'platformFlagLabel': 'web',
 264                 'language': 'en',
 265                 'uuid': str(uuid.uuid4()),
 266                 'carrierId': '0',
 267             }).encode())['token']
 268
 269     def _real_extract(self, url):
 270         url, idata = unsmuggle_url(url, {})
 271         country_code, lang_code, video_id = self._match_valid_url(url).groups()
 272
 273         query = {
 274             'r': 'vod/ajax-detail',
 275             'platform_flag_label': 'web',
 276             'product_id': video_id,
 277         }
 278
 279         area_id = self._AREA_ID.get(country_code.upper())
 280         if area_id:
 281             query['area_id'] = area_id
 282
 283         product_data = self._download_json(
 284             f'http://www.viu.com/ott/{country_code}/index.php', video_id,
 285             'Downloading video info', query=query)['data']
 286
 287         video_data = product_data.get('current_product')
 288         if not video_data:
 289             self.raise_geo_restricted()
 290
 291         series_id = video_data.get('series_id')
 292         if self._yes_playlist(series_id, video_id, idata):
 293             series = product_data.get('series') or {}
 294             product = series.get('product')
 295             if product:
 296                 entries = []
 297                 for entry in sorted(product, key=lambda x: int_or_none(x.get('number', 0))):
 298                     item_id = entry.get('product_id')
 299                     if not item_id:
 300                         continue
 301                     entries.append(self.url_result(
 302                         smuggle_url(f'http://www.viu.com/ott/{country_code}/{lang_code}/vod/{item_id}/',
 303                                     {'force_noplaylist': True}),
 304                         ViuOTTIE, str(item_id), entry.get('synopsis', '').strip()))
 305
 306                 return self.playlist_result(entries, series_id, series.get('name'), series.get('description'))
 307
 308         duration_limit = False
 309         query = {
 310             'ccs_product_id': video_data['ccs_product_id'],
 311             'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or '3',
 312         }
 313
 314         def download_playback():
 315             stream_data = self._download_json(
 316                 'https://api-gateway-global.viu.com/api/playback/distribute',
 317                 video_id=video_id, query=query, fatal=False, note='Downloading stream info',
 318                 headers={
 319                     'Authorization': f'Bearer {self._auth_codes[country_code]}',
 320                     'Referer': url,
 321                     'Origin': url,
 322                 })
 323             return self._detect_error(stream_data).get('stream')
 324
 325         if not self._auth_codes.get(country_code):
 326             self._auth_codes[country_code] = self._get_token(country_code, video_id)
 327
 328         stream_data = None
 329         try:
 330             stream_data = download_playback()
 331         except (ExtractorError, KeyError):
 332             token = self._login(country_code, video_id)
 333             if token is not None:
 334                 query['identity'] = token
 335             else:
 336                 # The content is Preview or for VIP only.
 337                 # We can try to bypass the duration which is limited to 3mins only
 338                 duration_limit, query['duration'] = True, '180'
 339             try:
 340                 stream_data = download_playback()
 341             except (ExtractorError, KeyError):
 342                 if token is not None:
 343                     raise
 344                 self.raise_login_required(method='password')
 345         if not stream_data:
 346             raise ExtractorError('Cannot get stream info', expected=True)
 347
 348         formats = []
 349         for vid_format, stream_url in (stream_data.get('url') or {}).items():
 350             height = int(self._search_regex(r's(\d+)p', vid_format, 'height', default=None))
 351
 352             # bypass preview duration limit
 353             if duration_limit:
 354                 old_stream_url = urllib.parse.urlparse(stream_url)
 355                 query = dict(urllib.parse.parse_qsl(old_stream_url.query, keep_blank_values=True))
 356                 query.update({
 357                     'duration': video_data.get('time_duration') or '9999999',
 358                     'duration_start': '0',
 359                 })
 360                 stream_url = old_stream_url._replace(query=urllib.parse.urlencode(query)).geturl()
 361
 362             formats.append({
 363                 'format_id': vid_format,
 364                 'url': stream_url,
 365                 'height': height,
 366                 'ext': 'mp4',
 367                 'filesize': try_get(stream_data, lambda x: x['size'][vid_format], int),
 368             })
 369
 370         subtitles = {}
 371         for sub in video_data.get('subtitle') or []:
 372             lang = sub.get('name') or 'und'
 373             if sub.get('url'):
 374                 subtitles.setdefault(lang, []).append({
 375                     'url': sub['url'],
 376                     'ext': 'srt',
 377                     'name': f'Spoken text for {lang}',
 378                 })
 379             if sub.get('second_subtitle_url'):
 380                 subtitles.setdefault(f'{lang}_ost', []).append({
 381                     'url': sub['second_subtitle_url'],
 382                     'ext': 'srt',
 383                     'name': f'On-screen text for {lang}',
 384                 })
 385
 386         title = strip_or_none(video_data.get('synopsis'))
 387         return {
 388             'id': video_id,
 389             'title': title,
 390             'description': video_data.get('description'),
 391             'series': try_get(product_data, lambda x: x['series']['name']),
 392             'episode': title,
 393             'episode_number': int_or_none(video_data.get('number')),
 394             'duration': int_or_none(stream_data.get('duration')),
 395             'thumbnail': url_or_none(video_data.get('cover_image_url')),
 396             'formats': formats,
 397             'subtitles': subtitles,
 398         }
 399
 400
 401 class ViuOTTIndonesiaBaseIE(InfoExtractor):
 402     _BASE_QUERY = {
 403         'ver': 1.0,
 404         'fmt': 'json',
 405         'aver': 5.0,
 406         'appver': 2.0,
 407         'appid': 'viu_desktop',
 408         'platform': 'desktop',
 409     }
 410
 411     _DEVICE_ID = str(uuid.uuid4())
 412     _SESSION_ID = str(uuid.uuid4())
 413     _TOKEN = None
 414
 415     _HEADERS = {
 416         'x-session-id': _SESSION_ID,
 417         'x-client': 'browser',
 418     }
 419
 420     _AGE_RATINGS_MAPPER = {
 421         'ADULTS': 18,
 422         'teens': 13,
 423     }
 424
 425     def _real_initialize(self):
 426         ViuOTTIndonesiaBaseIE._TOKEN = self._download_json(
 427             'https://um.viuapi.io/user/identity', None,
 428             headers={'Content-type': 'application/json', **self._HEADERS},
 429             query={**self._BASE_QUERY, 'iid': self._DEVICE_ID},
 430             data=json.dumps({'deviceId': self._DEVICE_ID}).encode(),
 431             note='Downloading token information')['token']
 432
 433
 434 class ViuOTTIndonesiaIE(ViuOTTIndonesiaBaseIE):
 435     _VALID_URL = r'https?://www\.viu\.com/ott/\w+/\w+/all/video-[\w-]+-(?P<id>\d+)'
 436     _TESTS = [{
 437         'url': 'https://www.viu.com/ott/id/id/all/video-japanese-drama-tv_shows-detective_conan_episode_793-1165863142?containerId=playlist-26271226',
 438         'info_dict': {
 439             'id': '1165863142',
 440             'ext': 'mp4',
 441             'episode_number': 793,
 442             'episode': 'Episode 793',
 443             'title': 'Detective Conan - Episode 793',
 444             'duration': 1476,
 445             'description': 'md5:b79d55345bc1e0217ece22616267c9a5',
 446             'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1165863189/d-1',
 447             'upload_date': '20210101',
 448             'timestamp': 1609459200,
 449         },
 450     }, {
 451         'url': 'https://www.viu.com/ott/id/id/all/video-korean-reality-tv_shows-entertainment_weekly_episode_1622-1118617054',
 452         'info_dict': {
 453             'id': '1118617054',
 454             'ext': 'mp4',
 455             'episode_number': 1622,
 456             'episode': 'Episode 1622',
 457             'description': 'md5:6d68ca450004020113e9bf27ad99f0f8',
 458             'title': 'Entertainment Weekly - Episode 1622',
 459             'duration': 4729,
 460             'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1120187848/d-1',
 461             'timestamp': 1420070400,
 462             'upload_date': '20150101',
 463             'cast': ['Shin Hyun-joon', 'Lee Da-Hee'],
 464         },
 465     }, {
 466         # age-limit test
 467         'url': 'https://www.viu.com/ott/id/id/all/video-japanese-trailer-tv_shows-trailer_jujutsu_kaisen_ver_01-1166044219?containerId=playlist-26273140',
 468         'info_dict': {
 469             'id': '1166044219',
 470             'ext': 'mp4',
 471             'upload_date': '20200101',
 472             'timestamp': 1577836800,
 473             'title': 'Trailer \'Jujutsu Kaisen\' Ver.01',
 474             'duration': 92,
 475             'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1166044240/d-1',
 476             'description': 'Trailer \'Jujutsu Kaisen\' Ver.01',
 477             'cast': ['Junya Enoki', ' Yûichi Nakamura', ' Yuma Uchida', 'Asami Seto'],
 478             'age_limit': 13,
 479         },
 480     }, {
 481         # json ld metadata type equal to Movie instead of TVEpisodes
 482         'url': 'https://www.viu.com/ott/id/id/all/video-japanese-animation-movies-demon_slayer_kimetsu_no_yaiba_the_movie_mugen_train-1165892707?containerId=1675060691786',
 483         'info_dict': {
 484             'id': '1165892707',
 485             'ext': 'mp4',
 486             'timestamp': 1577836800,
 487             'upload_date': '20200101',
 488             'title': 'Demon Slayer - Kimetsu no Yaiba - The Movie: Mugen Train',
 489             'age_limit': 13,
 490             'cast': 'count:9',
 491             'thumbnail': 'https://vuclipi-a.akamaihd.net/p/cloudinary/h_171,w_304,dpr_1.5,f_auto,c_thumb,q_auto:low/1165895279/d-1',
 492             'description': 'md5:1ce9c35a3aeab384085533f746c87469',
 493             'duration': 7021,
 494         },
 495     }]
 496
 497     def _real_extract(self, url):
 498         display_id = self._match_id(url)
 499         webpage = self._download_webpage(url, display_id)
 500
 501         video_data = self._download_json(
 502             f'https://um.viuapi.io/drm/v1/content/{display_id}', display_id, data=b'',
 503             headers={'Authorization': ViuOTTIndonesiaBaseIE._TOKEN, **self._HEADERS, 'ccode': 'ID'})
 504         formats, subtitles = self._extract_m3u8_formats_and_subtitles(video_data['playUrl'], display_id)
 505
 506         initial_state = self._search_json(
 507             r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state',
 508             display_id)['content']['clipDetails']
 509         for key, url in initial_state.items():
 510             lang, ext = self._search_regex(
 511                 r'^subtitle_(?P<lang>[\w-]+)_(?P<ext>\w+)$', key, 'subtitle metadata',
 512                 default=(None, None), group=('lang', 'ext'))
 513             if lang and ext:
 514                 subtitles.setdefault(lang, []).append({
 515                     'ext': ext,
 516                     'url': url,
 517                 })
 518
 519                 if ext == 'vtt':
 520                     subtitles[lang].append({
 521                         'ext': 'srt',
 522                         'url': f'{remove_end(initial_state[key], "vtt")}srt',
 523                     })
 524
 525         episode = traverse_obj(list(filter(
 526             lambda x: x.get('@type') in ('TVEpisode', 'Movie'), self._yield_json_ld(webpage, display_id))), 0) or {}
 527         return {
 528             'id': display_id,
 529             'title': (traverse_obj(initial_state, 'title', 'display_title')
 530                       or episode.get('name')),
 531             'description': initial_state.get('description') or episode.get('description'),
 532             'duration': initial_state.get('duration'),
 533             'thumbnail': traverse_obj(episode, ('image', 'url')),
 534             'timestamp': unified_timestamp(episode.get('dateCreated')),
 535             'formats': formats,
 536             'subtitles': subtitles,
 537             'episode_number': (traverse_obj(initial_state, 'episode_no', 'episodeno', expected_type=int_or_none)
 538                                or int_or_none(episode.get('episodeNumber'))),
 539             'cast': traverse_obj(episode, ('actor', ..., 'name'), default=None),
 540             'age_limit': self._AGE_RATINGS_MAPPER.get(initial_state.get('internal_age_rating')),
 541         }