yt_dlp/extractor/viu.py

   1 import re
   2 import json
   3 import uuid
   4 import random
   5 import urllib.parse
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_str
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     strip_or_none,
  13     try_get,
  14     smuggle_url,
  15     unsmuggle_url,
  16     url_or_none,
  17 )
  18
  19
  20 class ViuBaseIE(InfoExtractor):
  21     def _call_api(self, path, *args, headers={}, **kwargs):
  22         response = self._download_json(
  23             f'https://www.viu.com/api/{path}', *args, **kwargs,
  24             headers={**self.geo_verification_headers(), **headers})['response']
  25         if response.get('status') != 'success':
  26             raise ExtractorError(f'{self.IE_NAME} said: {response["message"]}', expected=True)
  27         return response
  28
  29
  30 class ViuIE(ViuBaseIE):
  31     _VALID_URL = r'(?:viu:|https?://[^/]+\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
  32     _TESTS = [{
  33         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
  34         'info_dict': {
  35             'id': '1116705532',
  36             'ext': 'mp4',
  37             'title': 'Citizen Khan - Ep 1',
  38             'description': 'md5:d7ea1604f49e5ba79c212c551ce2110e',
  39         },
  40         'params': {
  41             'skip_download': 'm3u8 download',
  42         },
  43         'skip': 'Geo-restricted to India',
  44     }, {
  45         'url': 'https://www.viu.com/en/media/1130599965',
  46         'info_dict': {
  47             'id': '1130599965',
  48             'ext': 'mp4',
  49             'title': 'Jealousy Incarnate - Episode 1',
  50             'description': 'md5:d3d82375cab969415d2720b6894361e9',
  51         },
  52         'params': {
  53             'skip_download': 'm3u8 download',
  54         },
  55         'skip': 'Geo-restricted to Indonesia',
  56     }, {
  57         'url': 'https://india.viu.com/en/media/1126286865',
  58         'only_matching': True,
  59     }]
  60
  61     def _real_extract(self, url):
  62         video_id = self._match_id(url)
  63
  64         video_data = self._call_api(
  65             'clip/load', video_id, 'Downloading video data', query={
  66                 'appid': 'viu_desktop',
  67                 'fmt': 'json',
  68                 'id': video_id
  69             })['item'][0]
  70
  71         title = video_data['title']
  72
  73         m3u8_url = None
  74         url_path = video_data.get('urlpathd') or video_data.get('urlpath')
  75         tdirforwhole = video_data.get('tdirforwhole')
  76         # #EXT-X-BYTERANGE is not supported by native hls downloader
  77         # and ffmpeg (#10955)
  78         # FIXME: It is supported in yt-dlp
  79         # hls_file = video_data.get('hlsfile')
  80         hls_file = video_data.get('jwhlsfile')
  81         if url_path and tdirforwhole and hls_file:
  82             m3u8_url = '%s/%s/%s' % (url_path, tdirforwhole, hls_file)
  83         else:
  84             # m3u8_url = re.sub(
  85             #     r'(/hlsc_)[a-z]+(\d+\.m3u8)',
  86             #     r'\1whe\2', video_data['href'])
  87             m3u8_url = video_data['href']
  88         formats, subtitles = self._extract_m3u8_formats_and_subtitles(m3u8_url, video_id, 'mp4')
  89         self._sort_formats(formats)
  90
  91         for key, value in video_data.items():
  92             mobj = re.match(r'^subtitle_(?P<lang>[^_]+)_(?P<ext>(vtt|srt))', key)
  93             if not mobj:
  94                 continue
  95             subtitles.setdefault(mobj.group('lang'), []).append({
  96                 'url': value,
  97                 'ext': mobj.group('ext')
  98             })
  99
 100         return {
 101             'id': video_id,
 102             'title': title,
 103             'description': video_data.get('description'),
 104             'series': video_data.get('moviealbumshowname'),
 105             'episode': title,
 106             'episode_number': int_or_none(video_data.get('episodeno')),
 107             'duration': int_or_none(video_data.get('duration')),
 108             'formats': formats,
 109             'subtitles': subtitles,
 110         }
 111
 112
 113 class ViuPlaylistIE(ViuBaseIE):
 114     IE_NAME = 'viu:playlist'
 115     _VALID_URL = r'https?://www\.viu\.com/[^/]+/listing/playlist-(?P<id>\d+)'
 116     _TEST = {
 117         'url': 'https://www.viu.com/en/listing/playlist-22461380',
 118         'info_dict': {
 119             'id': '22461380',
 120             'title': 'The Good Wife',
 121         },
 122         'playlist_count': 16,
 123         'skip': 'Geo-restricted to Indonesia',
 124     }
 125
 126     def _real_extract(self, url):
 127         playlist_id = self._match_id(url)
 128         playlist_data = self._call_api(
 129             'container/load', playlist_id,
 130             'Downloading playlist info', query={
 131                 'appid': 'viu_desktop',
 132                 'fmt': 'json',
 133                 'id': 'playlist-' + playlist_id
 134             })['container']
 135
 136         entries = []
 137         for item in playlist_data.get('item', []):
 138             item_id = item.get('id')
 139             if not item_id:
 140                 continue
 141             item_id = compat_str(item_id)
 142             entries.append(self.url_result(
 143                 'viu:' + item_id, 'Viu', item_id))
 144
 145         return self.playlist_result(
 146             entries, playlist_id, playlist_data.get('title'))
 147
 148
 149 class ViuOTTIE(InfoExtractor):
 150     IE_NAME = 'viu:ott'
 151     _NETRC_MACHINE = 'viu'
 152     _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+)'
 153     _TESTS = [{
 154         'url': 'http://www.viu.com/ott/sg/en-us/vod/3421/The%20Prime%20Minister%20and%20I',
 155         'info_dict': {
 156             'id': '3421',
 157             'ext': 'mp4',
 158             'title': 'A New Beginning',
 159             'description': 'md5:1e7486a619b6399b25ba6a41c0fe5b2c',
 160         },
 161         'params': {
 162             'skip_download': 'm3u8 download',
 163             'noplaylist': True,
 164         },
 165         'skip': 'Geo-restricted to Singapore',
 166     }, {
 167         'url': 'https://www.viu.com/ott/hk/zh-hk/vod/430078/%E7%AC%AC%E5%85%AD%E6%84%9F-3',
 168         'info_dict': {
 169             'id': '430078',
 170             'ext': 'mp4',
 171             'title': '大韓民國的1%',
 172             'description': 'md5:74d6db47ddd9ddb9c89a05739103ccdb',
 173             'episode_number': 1,
 174             'duration': 6614,
 175             'episode': '大韓民國的1%',
 176             'series': '第六感 3',
 177             'thumbnail': 'https://d2anahhhmp1ffz.cloudfront.net/1313295781/d2b14f48d008ef2f3a9200c98d8e9b63967b9cc2',
 178         },
 179         'params': {
 180             'skip_download': 'm3u8 download',
 181             'noplaylist': True,
 182         },
 183         'skip': 'Geo-restricted to Hong Kong',
 184     }, {
 185         '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',
 186         'playlist_count': 16,
 187         'info_dict': {
 188             'id': '23807',
 189             'title': '我的室友是九尾狐',
 190             'description': 'md5:b42c95f2b4a316cdd6ae14ca695f33b9',
 191         },
 192         'params': {
 193             'skip_download': 'm3u8 download',
 194             'noplaylist': False,
 195         },
 196         'skip': 'Geo-restricted to Hong Kong',
 197     }]
 198
 199     _AREA_ID = {
 200         'HK': 1,
 201         'SG': 2,
 202         'TH': 4,
 203         'PH': 5,
 204     }
 205     _LANGUAGE_FLAG = {
 206         'zh-hk': 1,
 207         'zh-cn': 2,
 208         'en-us': 3,
 209     }
 210
 211     _user_token = None
 212     _auth_codes = {}
 213
 214     def _detect_error(self, response):
 215         code = try_get(response, lambda x: x['status']['code'])
 216         if code and code > 0:
 217             message = try_get(response, lambda x: x['status']['message'])
 218             raise ExtractorError(f'{self.IE_NAME} said: {message} ({code})', expected=True)
 219         return response.get('data') or {}
 220
 221     def _login(self, country_code, video_id):
 222         if self._user_token is None:
 223             username, password = self._get_login_info()
 224             if username is None:
 225                 return
 226             headers = {
 227                 'Authorization': f'Bearer {self._auth_codes[country_code]}',
 228                 'Content-Type': 'application/json'
 229             }
 230             data = self._download_json(
 231                 'https://api-gateway-global.viu.com/api/account/validate',
 232                 video_id, 'Validating email address', headers=headers,
 233                 data=json.dumps({
 234                     'principal': username,
 235                     'provider': 'email'
 236                 }).encode())
 237             if not data.get('exists'):
 238                 raise ExtractorError('Invalid email address')
 239
 240             data = self._download_json(
 241                 'https://api-gateway-global.viu.com/api/auth/login',
 242                 video_id, 'Logging in', headers=headers,
 243                 data=json.dumps({
 244                     'email': username,
 245                     'password': password,
 246                     'provider': 'email',
 247                 }).encode())
 248             self._detect_error(data)
 249             self._user_token = data.get('identity')
 250             # need to update with valid user's token else will throw an error again
 251             self._auth_codes[country_code] = data.get('token')
 252         return self._user_token
 253
 254     def _get_token(self, country_code, video_id):
 255         rand = ''.join(random.choice('0123456789') for _ in range(10))
 256         return self._download_json(
 257             f'https://api-gateway-global.viu.com/api/auth/token?v={rand}000', video_id,
 258             headers={'Content-Type': 'application/json'}, note='Getting bearer token',
 259             data=json.dumps({
 260                 'countryCode': country_code.upper(),
 261                 'platform': 'browser',
 262                 'platformFlagLabel': 'web',
 263                 'language': 'en',
 264                 'uuid': str(uuid.uuid4()),
 265                 'carrierId': '0'
 266             }).encode('utf-8'))['token']
 267
 268     def _real_extract(self, url):
 269         url, idata = unsmuggle_url(url, {})
 270         country_code, lang_code, video_id = self._match_valid_url(url).groups()
 271
 272         query = {
 273             'r': 'vod/ajax-detail',
 274             'platform_flag_label': 'web',
 275             'product_id': video_id,
 276         }
 277
 278         area_id = self._AREA_ID.get(country_code.upper())
 279         if area_id:
 280             query['area_id'] = area_id
 281
 282         product_data = self._download_json(
 283             f'http://www.viu.com/ott/{country_code}/index.php', video_id,
 284             'Downloading video info', query=query)['data']
 285
 286         video_data = product_data.get('current_product')
 287         if not video_data:
 288             self.raise_geo_restricted()
 289
 290         series_id = video_data.get('series_id')
 291         if self._yes_playlist(series_id, video_id, idata):
 292             series = product_data.get('series') or {}
 293             product = series.get('product')
 294             if product:
 295                 entries = []
 296                 for entry in sorted(product, key=lambda x: int_or_none(x.get('number', 0))):
 297                     item_id = entry.get('product_id')
 298                     if not item_id:
 299                         continue
 300                     entries.append(self.url_result(
 301                         smuggle_url(f'http://www.viu.com/ott/{country_code}/{lang_code}/vod/{item_id}/',
 302                                     {'force_noplaylist': True}),
 303                         ViuOTTIE, str(item_id), entry.get('synopsis', '').strip()))
 304
 305                 return self.playlist_result(entries, series_id, series.get('name'), series.get('description'))
 306
 307         duration_limit = False
 308         query = {
 309             'ccs_product_id': video_data['ccs_product_id'],
 310             'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or '3',
 311         }
 312
 313         def download_playback():
 314             stream_data = self._download_json(
 315                 'https://api-gateway-global.viu.com/api/playback/distribute',
 316                 video_id=video_id, query=query, fatal=False, note='Downloading stream info',
 317                 headers={
 318                     'Authorization': f'Bearer {self._auth_codes[country_code]}',
 319                     'Referer': url,
 320                     'Origin': url
 321                 })
 322             return self._detect_error(stream_data).get('stream')
 323
 324         if not self._auth_codes.get(country_code):
 325             self._auth_codes[country_code] = self._get_token(country_code, video_id)
 326
 327         stream_data = None
 328         try:
 329             stream_data = download_playback()
 330         except (ExtractorError, KeyError):
 331             token = self._login(country_code, video_id)
 332             if token is not None:
 333                 query['identity'] = token
 334             else:
 335                 # The content is Preview or for VIP only.
 336                 # We can try to bypass the duration which is limited to 3mins only
 337                 duration_limit, query['duration'] = True, '180'
 338             try:
 339                 stream_data = download_playback()
 340             except (ExtractorError, KeyError):
 341                 if token is not None:
 342                     raise
 343                 self.raise_login_required(method='password')
 344         if not stream_data:
 345             raise ExtractorError('Cannot get stream info', expected=True)
 346
 347         formats = []
 348         for vid_format, stream_url in (stream_data.get('url') or {}).items():
 349             height = int(self._search_regex(r's(\d+)p', vid_format, 'height', default=None))
 350
 351             # bypass preview duration limit
 352             if duration_limit:
 353                 old_stream_url = urllib.parse.urlparse(stream_url)
 354                 query = dict(urllib.parse.parse_qsl(old_stream_url.query, keep_blank_values=True))
 355                 query.update({
 356                     'duration': video_data.get('time_duration') or '9999999',
 357                     'duration_start': '0',
 358                 })
 359                 stream_url = old_stream_url._replace(query=urllib.parse.urlencode(query)).geturl()
 360
 361             formats.append({
 362                 'format_id': vid_format,
 363                 'url': stream_url,
 364                 'height': height,
 365                 'ext': 'mp4',
 366                 'filesize': try_get(stream_data, lambda x: x['size'][vid_format], int)
 367             })
 368         self._sort_formats(formats)
 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         }