yt_dlp/extractor/viu.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_kwargs,
  10     compat_str,
  11     compat_urlparse,
  12     compat_urllib_request,
  13 )
  14 from ..utils import (
  15     ExtractorError,
  16     int_or_none,
  17     try_get,
  18     smuggle_url,
  19     unsmuggle_url,
  20 )
  21
  22
  23 class ViuBaseIE(InfoExtractor):
  24     def _real_initialize(self):
  25         viu_auth_res = self._request_webpage(
  26             'https://www.viu.com/api/apps/v2/authenticate', None,
  27             'Requesting Viu auth', query={
  28                 'acct': 'test',
  29                 'appid': 'viu_desktop',
  30                 'fmt': 'json',
  31                 'iid': 'guest',
  32                 'languageid': 'default',
  33                 'platform': 'desktop',
  34                 'userid': 'guest',
  35                 'useridtype': 'guest',
  36                 'ver': '1.0'
  37             }, headers=self.geo_verification_headers())
  38         self._auth_token = viu_auth_res.info()['X-VIU-AUTH']
  39
  40     def _call_api(self, path, *args, **kwargs):
  41         headers = self.geo_verification_headers()
  42         headers.update({
  43             'X-VIU-AUTH': self._auth_token
  44         })
  45         headers.update(kwargs.get('headers', {}))
  46         kwargs['headers'] = headers
  47         response = self._download_json(
  48             'https://www.viu.com/api/' + path, *args,
  49             **compat_kwargs(kwargs))['response']
  50         if response.get('status') != 'success':
  51             raise ExtractorError('%s said: %s' % (
  52                 self.IE_NAME, response['message']), expected=True)
  53         return response
  54
  55
  56 class ViuIE(ViuBaseIE):
  57     _VALID_URL = r'(?:viu:|https?://[^/]+\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
  58     _TESTS = [{
  59         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
  60         'info_dict': {
  61             'id': '1116705532',
  62             'ext': 'mp4',
  63             'title': 'Citizen Khan - Ep 1',
  64             'description': 'md5:d7ea1604f49e5ba79c212c551ce2110e',
  65         },
  66         'params': {
  67             'skip_download': 'm3u8 download',
  68         },
  69         'skip': 'Geo-restricted to India',
  70     }, {
  71         'url': 'https://www.viu.com/en/media/1130599965',
  72         'info_dict': {
  73             'id': '1130599965',
  74             'ext': 'mp4',
  75             'title': 'Jealousy Incarnate - Episode 1',
  76             'description': 'md5:d3d82375cab969415d2720b6894361e9',
  77         },
  78         'params': {
  79             'skip_download': 'm3u8 download',
  80         },
  81         'skip': 'Geo-restricted to Indonesia',
  82     }, {
  83         'url': 'https://india.viu.com/en/media/1126286865',
  84         'only_matching': True,
  85     }]
  86
  87     def _real_extract(self, url):
  88         video_id = self._match_id(url)
  89
  90         video_data = self._call_api(
  91             'clip/load', video_id, 'Downloading video data', query={
  92                 'appid': 'viu_desktop',
  93                 'fmt': 'json',
  94                 'id': video_id
  95             })['item'][0]
  96
  97         title = video_data['title']
  98
  99         m3u8_url = None
 100         url_path = video_data.get('urlpathd') or video_data.get('urlpath')
 101         tdirforwhole = video_data.get('tdirforwhole')
 102         # #EXT-X-BYTERANGE is not supported by native hls downloader
 103         # and ffmpeg (#10955)
 104         # hls_file = video_data.get('hlsfile')
 105         hls_file = video_data.get('jwhlsfile')
 106         if url_path and tdirforwhole and hls_file:
 107             m3u8_url = '%s/%s/%s' % (url_path, tdirforwhole, hls_file)
 108         else:
 109             # m3u8_url = re.sub(
 110             #     r'(/hlsc_)[a-z]+(\d+\.m3u8)',
 111             #     r'\1whe\2', video_data['href'])
 112             m3u8_url = video_data['href']
 113         formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
 114         self._sort_formats(formats)
 115
 116         subtitles = {}
 117         for key, value in video_data.items():
 118             mobj = re.match(r'^subtitle_(?P<lang>[^_]+)_(?P<ext>(vtt|srt))', key)
 119             if not mobj:
 120                 continue
 121             subtitles.setdefault(mobj.group('lang'), []).append({
 122                 'url': value,
 123                 'ext': mobj.group('ext')
 124             })
 125
 126         return {
 127             'id': video_id,
 128             'title': title,
 129             'description': video_data.get('description'),
 130             'series': video_data.get('moviealbumshowname'),
 131             'episode': title,
 132             'episode_number': int_or_none(video_data.get('episodeno')),
 133             'duration': int_or_none(video_data.get('duration')),
 134             'formats': formats,
 135             'subtitles': subtitles,
 136         }
 137
 138
 139 class ViuPlaylistIE(ViuBaseIE):
 140     IE_NAME = 'viu:playlist'
 141     _VALID_URL = r'https?://www\.viu\.com/[^/]+/listing/playlist-(?P<id>\d+)'
 142     _TEST = {
 143         'url': 'https://www.viu.com/en/listing/playlist-22461380',
 144         'info_dict': {
 145             'id': '22461380',
 146             'title': 'The Good Wife',
 147         },
 148         'playlist_count': 16,
 149         'skip': 'Geo-restricted to Indonesia',
 150     }
 151
 152     def _real_extract(self, url):
 153         playlist_id = self._match_id(url)
 154         playlist_data = self._call_api(
 155             'container/load', playlist_id,
 156             'Downloading playlist info', query={
 157                 'appid': 'viu_desktop',
 158                 'fmt': 'json',
 159                 'id': 'playlist-' + playlist_id
 160             })['container']
 161
 162         entries = []
 163         for item in playlist_data.get('item', []):
 164             item_id = item.get('id')
 165             if not item_id:
 166                 continue
 167             item_id = compat_str(item_id)
 168             entries.append(self.url_result(
 169                 'viu:' + item_id, 'Viu', item_id))
 170
 171         return self.playlist_result(
 172             entries, playlist_id, playlist_data.get('title'))
 173
 174
 175 class ViuOTTIE(InfoExtractor):
 176     IE_NAME = 'viu:ott'
 177     _NETRC_MACHINE = 'viu'
 178     _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+)'
 179     _TESTS = [{
 180         'url': 'http://www.viu.com/ott/sg/en-us/vod/3421/The%20Prime%20Minister%20and%20I',
 181         'info_dict': {
 182             'id': '3421',
 183             'ext': 'mp4',
 184             'title': 'A New Beginning',
 185             'description': 'md5:1e7486a619b6399b25ba6a41c0fe5b2c',
 186         },
 187         'params': {
 188             'skip_download': 'm3u8 download',
 189             'noplaylist': True,
 190         },
 191         'skip': 'Geo-restricted to Singapore',
 192     }, {
 193         'url': 'http://www.viu.com/ott/hk/zh-hk/vod/7123/%E5%A4%A7%E4%BA%BA%E5%A5%B3%E5%AD%90',
 194         'info_dict': {
 195             'id': '7123',
 196             'ext': 'mp4',
 197             'title': '這就是我的生活之道',
 198             'description': 'md5:4eb0d8b08cf04fcdc6bbbeb16043434f',
 199         },
 200         'params': {
 201             'skip_download': 'm3u8 download',
 202             'noplaylist': True,
 203         },
 204         'skip': 'Geo-restricted to Hong Kong',
 205     }, {
 206         'url': 'https://www.viu.com/ott/hk/zh-hk/vod/68776/%E6%99%82%E5%B0%9A%E5%AA%BD%E5%92%AA',
 207         'playlist_count': 12,
 208         'info_dict': {
 209             'id': '3916',
 210             'title': '時尚媽咪',
 211         },
 212         'params': {
 213             'skip_download': 'm3u8 download',
 214             'noplaylist': False,
 215         },
 216         'skip': 'Geo-restricted to Hong Kong',
 217     }]
 218
 219     _AREA_ID = {
 220         'HK': 1,
 221         'SG': 2,
 222         'TH': 4,
 223         'PH': 5,
 224     }
 225     _LANGUAGE_FLAG = {
 226         'zh-hk': 1,
 227         'zh-cn': 2,
 228         'en-us': 3,
 229     }
 230     _user_info = None
 231
 232     def _detect_error(self, response):
 233         code = response.get('status', {}).get('code')
 234         if code > 0:
 235             message = try_get(response, lambda x: x['status']['message'])
 236             raise ExtractorError('%s said: %s (%s)' % (
 237                 self.IE_NAME, message, code), expected=True)
 238         return response['data']
 239
 240     def _raise_login_required(self):
 241         raise ExtractorError(
 242             'This video requires login. '
 243             'Specify --username and --password or --netrc (machine: %s) '
 244             'to provide account credentials.' % self._NETRC_MACHINE,
 245             expected=True)
 246
 247     def _login(self, country_code, video_id):
 248         if not self._user_info:
 249             username, password = self._get_login_info()
 250             if username is None or password is None:
 251                 return
 252
 253             data = self._download_json(
 254                 compat_urllib_request.Request(
 255                     'https://www.viu.com/ott/%s/index.php' % country_code, method='POST'),
 256                 video_id, 'Logging in', errnote=False, fatal=False,
 257                 query={'r': 'user/login'},
 258                 data=json.dumps({
 259                     'username': username,
 260                     'password': password,
 261                     'platform_flag_label': 'web',
 262                 }).encode())
 263             self._user_info = self._detect_error(data)['user']
 264
 265         return self._user_info
 266
 267     def _real_extract(self, url):
 268         url, idata = unsmuggle_url(url, {})
 269         country_code, lang_code, video_id = self._match_valid_url(url).groups()
 270
 271         query = {
 272             'r': 'vod/ajax-detail',
 273             'platform_flag_label': 'web',
 274             'product_id': video_id,
 275         }
 276
 277         area_id = self._AREA_ID.get(country_code.upper())
 278         if area_id:
 279             query['area_id'] = area_id
 280
 281         product_data = self._download_json(
 282             'http://www.viu.com/ott/%s/index.php' % country_code, video_id,
 283             'Downloading video info', query=query)['data']
 284
 285         video_data = product_data.get('current_product')
 286         if not video_data:
 287             raise ExtractorError('This video is not available in your region.', expected=True)
 288
 289         series_id = video_data.get('series_id')
 290         if self._yes_playlist(series_id, video_id, idata):
 291             series = product_data.get('series', {})
 292             product = series.get('product')
 293             if product:
 294                 entries = []
 295                 for entry in sorted(product, key=lambda x: int_or_none(x.get('number', 0))):
 296                     item_id = entry.get('product_id')
 297                     if not item_id:
 298                         continue
 299                     item_id = compat_str(item_id)
 300                     entries.append(self.url_result(
 301                         smuggle_url(
 302                             'http://www.viu.com/ott/%s/%s/vod/%s/' % (country_code, lang_code, item_id),
 303                             {'force_noplaylist': True}),  # prevent infinite recursion
 304                         'ViuOTT',
 305                         item_id,
 306                         entry.get('synopsis', '').strip()))
 307
 308                 return self.playlist_result(entries, series_id, series.get('name'), series.get('description'))
 309
 310         duration_limit = False
 311         query = {
 312             'ccs_product_id': video_data['ccs_product_id'],
 313             'language_flag_id': self._LANGUAGE_FLAG.get(lang_code.lower()) or '3',
 314         }
 315         headers = {
 316             'Referer': url,
 317             'Origin': url,
 318         }
 319         try:
 320             stream_data = self._download_json(
 321                 'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
 322                 video_id, 'Downloading stream info', query=query, headers=headers)
 323             stream_data = self._detect_error(stream_data)['stream']
 324         except (ExtractorError, KeyError):
 325             stream_data = None
 326             if video_data.get('user_level', 0) > 0:
 327                 user = self._login(country_code, video_id)
 328                 if user:
 329                     query['identity'] = user['identity']
 330                     stream_data = self._download_json(
 331                         'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
 332                         video_id, 'Downloading stream info', query=query, headers=headers)
 333                     stream_data = self._detect_error(stream_data).get('stream')
 334                 else:
 335                     # preview is limited to 3min for non-members
 336                     # try to bypass the duration limit
 337                     duration_limit = True
 338                     query['duration'] = '180'
 339                     stream_data = self._download_json(
 340                         'https://d1k2us671qcoau.cloudfront.net/distribute_web_%s.php' % country_code,
 341                         video_id, 'Downloading stream info', query=query, headers=headers)
 342                     try:
 343                         stream_data = self._detect_error(stream_data)['stream']
 344                     except (ExtractorError, KeyError):  # if still not working, give up
 345                         self._raise_login_required()
 346
 347         if not stream_data:
 348             raise ExtractorError('Cannot get stream info', expected=True)
 349
 350         stream_sizes = stream_data.get('size', {})
 351         formats = []
 352         for vid_format, stream_url in stream_data.get('url', {}).items():
 353             height = int_or_none(self._search_regex(
 354                 r's(\d+)p', vid_format, 'height', default=None))
 355
 356             # bypass preview duration limit
 357             if duration_limit:
 358                 stream_url = compat_urlparse.urlparse(stream_url)
 359                 query = dict(compat_urlparse.parse_qsl(stream_url.query, keep_blank_values=True))
 360                 time_duration = int_or_none(video_data.get('time_duration'))
 361                 query.update({
 362                     'duration': time_duration if time_duration > 0 else '9999999',
 363                     'duration_start': '0',
 364                 })
 365                 stream_url = stream_url._replace(query=compat_urlparse.urlencode(query)).geturl()
 366
 367             formats.append({
 368                 'format_id': vid_format,
 369                 'url': stream_url,
 370                 'height': height,
 371                 'ext': 'mp4',
 372                 'filesize': int_or_none(stream_sizes.get(vid_format))
 373             })
 374         self._sort_formats(formats)
 375
 376         subtitles = {}
 377         for sub in video_data.get('subtitle', []):
 378             sub_url = sub.get('url')
 379             if not sub_url:
 380                 continue
 381             subtitles.setdefault(sub.get('name'), []).append({
 382                 'url': sub_url,
 383                 'ext': 'srt',
 384             })
 385
 386         title = video_data['synopsis'].strip()
 387
 388         return {
 389             'id': video_id,
 390             'title': title,
 391             'description': video_data.get('description'),
 392             'series': product_data.get('series', {}).get('name'),
 393             'episode': title,
 394             'episode_number': int_or_none(video_data.get('number')),
 395             'duration': int_or_none(stream_data.get('duration')),
 396             'thumbnail': video_data.get('cover_image_url'),
 397             'formats': formats,
 398             'subtitles': subtitles,
 399         }