yt_dlp/extractor/iprima.py

   1 import re
   2 import time
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     determine_ext,
   8     js_to_json,
   9     parse_qs,
  10     traverse_obj,
  11     urlencode_postdata,
  12 )
  13
  14
  15 class IPrimaIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?!cnn)(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  17     _GEO_BYPASS = False
  18     _NETRC_MACHINE = 'iprima'
  19     _AUTH_ROOT = 'https://auth.iprima.cz'
  20     access_token = None
  21
  22     _TESTS = [{
  23         'url': 'https://prima.iprima.cz/particka/92-epizoda',
  24         'info_dict': {
  25             'id': 'p51388',
  26             'ext': 'mp4',
  27             'title': 'Partička (92)',
  28             'description': 'md5:57943f6a50d6188288c3a579d2fd5f01',
  29             'episode': 'Partička (92)',
  30             'season': 'Partička',
  31             'series': 'Prima Partička',
  32             'episode_number': 92,
  33             'thumbnail': 'https://d31b9s05ygj54s.cloudfront.net/prima-plus/image/video-ef6cf9de-c980-4443-92e4-17fe8bccd45c-16x9.jpeg',
  34         },
  35         'params': {
  36             'skip_download': True,  # m3u8 download
  37         },
  38     }, {
  39         'url': 'https://zoom.iprima.cz/porady/krasy-kanarskych-ostrovu/tenerife-v-risi-ohne',
  40         'info_dict': {
  41             'id': 'p1412199',
  42             'ext': 'mp4',
  43             'episode_number': 3,
  44             'episode': 'Tenerife: V říši ohně',
  45             'description': 'md5:4b4a05c574b5eaef130e68d4811c3f2c',
  46             'duration': 3111.0,
  47             'thumbnail': 'https://d31b9s05ygj54s.cloudfront.net/prima-plus/image/video-f66dd7fb-c1a0-47d1-b3bc-7db328d566c5-16x9-1711636518.jpg/t_16x9_medium_1366_768',
  48             'title': 'Tenerife: V říši ohně',
  49             'timestamp': 1711825800,
  50             'upload_date': '20240330',
  51         },
  52         'params': {
  53             'skip_download': True,  # m3u8 download
  54         },
  55     }, {
  56         'url': 'http://play.iprima.cz/particka/particka-92',
  57         'only_matching': True,
  58     }, {
  59         # geo restricted
  60         'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
  61         'only_matching': True,
  62     }, {
  63         'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
  64         'only_matching': True,
  65     }, {
  66         'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
  67         'only_matching': True,
  68     }, {
  69         'url': 'http://www.iprima.cz/filmy/desne-rande',
  70         'only_matching': True,
  71     }, {
  72         'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
  73         'only_matching': True,
  74     }, {
  75         'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
  76         'only_matching': True,
  77     }, {
  78         'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
  79         'only_matching': True,
  80     }, {
  81         'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
  82         'only_matching': True,
  83     }]
  84
  85     def _perform_login(self, username, password):
  86         if self.access_token:
  87             return
  88
  89         login_page = self._download_webpage(
  90             f'{self._AUTH_ROOT}/oauth2/login', None, note='Downloading login page',
  91             errnote='Downloading login page failed')
  92
  93         login_form = self._hidden_inputs(login_page)
  94
  95         login_form.update({
  96             '_email': username,
  97             '_password': password})
  98
  99         profile_select_html, login_handle = self._download_webpage_handle(
 100             f'{self._AUTH_ROOT}/oauth2/login', None, data=urlencode_postdata(login_form),
 101             note='Logging in')
 102
 103         # a profile may need to be selected first, even when there is only a single one
 104         if '/profile-select' in login_handle.url:
 105             profile_id = self._search_regex(
 106                 r'data-identifier\s*=\s*["\']?(\w+)', profile_select_html, 'profile id')
 107
 108             login_handle = self._request_webpage(
 109                 f'{self._AUTH_ROOT}/user/profile-select-perform/{profile_id}', None,
 110                 query={'continueUrl': '/user/login?redirect_uri=/user/'}, note='Selecting profile')
 111
 112         code = traverse_obj(login_handle.url, ({parse_qs}, 'code', 0))
 113         if not code:
 114             raise ExtractorError('Login failed', expected=True)
 115
 116         token_request_data = {
 117             'scope': 'openid+email+profile+phone+address+offline_access',
 118             'client_id': 'prima_sso',
 119             'grant_type': 'authorization_code',
 120             'code': code,
 121             'redirect_uri': f'{self._AUTH_ROOT}/sso/auth-check'}
 122
 123         token_data = self._download_json(
 124             f'{self._AUTH_ROOT}/oauth2/token', None,
 125             note='Downloading token', errnote='Downloading token failed',
 126             data=urlencode_postdata(token_request_data))
 127
 128         self.access_token = token_data.get('access_token')
 129         if self.access_token is None:
 130             raise ExtractorError('Getting token failed', expected=True)
 131
 132     def _real_initialize(self):
 133         if not self.access_token:
 134             self.raise_login_required('Login is required to access any iPrima content', method='password')
 135
 136     def _raise_access_error(self, error_code):
 137         if error_code == 'PLAY_GEOIP_DENIED':
 138             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 139         elif error_code is not None:
 140             self.raise_no_formats('Access to stream infos forbidden', expected=True)
 141
 142     def _real_extract(self, url):
 143         video_id = self._match_id(url)
 144
 145         webpage = self._download_webpage(url, video_id)
 146
 147         title = self._html_extract_title(webpage) or self._html_search_meta(
 148             ['og:title', 'twitter:title'],
 149             webpage, 'title', default=None)
 150
 151         video_id = self._search_regex((
 152             r'productId\s*=\s*([\'"])(?P<id>p\d+)\1',
 153             r'pproduct_id\s*=\s*([\'"])(?P<id>p\d+)\1',
 154             r'let\s+videos\s*=\s*([\'"])(?P<id>p\d+)\1',
 155         ), webpage, 'real id', group='id', default=None)
 156
 157         if not video_id:
 158             nuxt_data = self._search_nuxt_data(webpage, video_id, traverse='data', fatal=False)
 159             video_id = traverse_obj(
 160                 nuxt_data, (..., 'content', 'additionals', 'videoPlayId', {str}), get_all=False)
 161
 162         if not video_id:
 163             nuxt_data = self._search_json(
 164                 r'<script[^>]+\bid=["\']__NUXT_DATA__["\'][^>]*>',
 165                 webpage, 'nuxt data', None, end_pattern=r'</script>', contains_pattern=r'\[(?s:.+)\]')
 166
 167             video_id = traverse_obj(nuxt_data, lambda _, v: re.fullmatch(r'p\d+', v), get_all=False)
 168
 169         if not video_id:
 170             self.raise_no_formats('Unable to extract video ID from webpage')
 171
 172         metadata = self._download_json(
 173             f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
 174             video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
 175             headers={'X-OTT-Access-Token': self.access_token},
 176             expected_status=403)
 177
 178         self._raise_access_error(metadata.get('errorCode'))
 179
 180         stream_infos = metadata.get('streamInfos')
 181         formats = []
 182         if stream_infos is None:
 183             self.raise_no_formats('Reading stream infos failed', expected=True)
 184         else:
 185             for manifest in stream_infos:
 186                 manifest_type = manifest.get('type')
 187                 manifest_url = manifest.get('url')
 188                 ext = determine_ext(manifest_url)
 189                 if manifest_type == 'HLS' or ext == 'm3u8':
 190                     formats += self._extract_m3u8_formats(
 191                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 192                         m3u8_id='hls', fatal=False)
 193                 elif manifest_type == 'DASH' or ext == 'mpd':
 194                     formats += self._extract_mpd_formats(
 195                         manifest_url, video_id, mpd_id='dash', fatal=False)
 196
 197         final_result = self._search_json_ld(webpage, video_id, default={})
 198         final_result.update({
 199             'id': video_id,
 200             'title': final_result.get('title') or title,
 201             'thumbnail': self._html_search_meta(
 202                 ['thumbnail', 'og:image', 'twitter:image'],
 203                 webpage, 'thumbnail', default=None),
 204             'formats': formats,
 205             'description': self._html_search_meta(
 206                 ['description', 'og:description', 'twitter:description'],
 207                 webpage, 'description', default=None)})
 208
 209         return final_result
 210
 211
 212 class IPrimaCNNIE(InfoExtractor):
 213     _VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 214     _GEO_BYPASS = False
 215
 216     _TESTS = [{
 217         'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
 218         'info_dict': {
 219             'id': 'p716177',
 220             'ext': 'mp4',
 221             'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
 222         },
 223         'params': {
 224             'skip_download': 'm3u8',
 225         },
 226     }]
 227
 228     def _real_extract(self, url):
 229         video_id = self._match_id(url)
 230
 231         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
 232
 233         webpage = self._download_webpage(url, video_id)
 234
 235         title = self._og_search_title(
 236             webpage, default=None) or self._search_regex(
 237             r'<h1>([^<]+)', webpage, 'title')
 238
 239         video_id = self._search_regex(
 240             (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
 241              r'data-product="([^"]+)">',
 242              r'id=["\']player-(p\d+)"',
 243              r'playerId\s*:\s*["\']player-(p\d+)',
 244              r'\bvideos\s*=\s*["\'](p\d+)'),
 245             webpage, 'real id')
 246
 247         playerpage = self._download_webpage(
 248             'http://play.iprima.cz/prehravac/init',
 249             video_id, note='Downloading player', query={
 250                 '_infuse': 1,
 251                 '_ts': round(time.time()),
 252                 'productId': video_id,
 253             }, headers={'Referer': url})
 254
 255         formats = []
 256
 257         def extract_formats(format_url, format_key=None, lang=None):
 258             ext = determine_ext(format_url)
 259             new_formats = []
 260             if format_key == 'hls' or ext == 'm3u8':
 261                 new_formats = self._extract_m3u8_formats(
 262                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 263                     m3u8_id='hls', fatal=False)
 264             elif format_key == 'dash' or ext == 'mpd':
 265                 return
 266                 new_formats = self._extract_mpd_formats(
 267                     format_url, video_id, mpd_id='dash', fatal=False)
 268             if lang:
 269                 for f in new_formats:
 270                     if not f.get('language'):
 271                         f['language'] = lang
 272             formats.extend(new_formats)
 273
 274         options = self._parse_json(
 275             self._search_regex(
 276                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
 277                 playerpage, 'player options', default='{}'),
 278             video_id, transform_source=js_to_json, fatal=False)
 279         if options:
 280             for key, tracks in options.get('tracks', {}).items():
 281                 if not isinstance(tracks, list):
 282                     continue
 283                 for track in tracks:
 284                     src = track.get('src')
 285                     if src:
 286                         extract_formats(src, key.lower(), track.get('lang'))
 287
 288         if not formats:
 289             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
 290                 extract_formats(src)
 291
 292         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
 293             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 294
 295         return {
 296             'id': video_id,
 297             'title': title,
 298             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 299             'formats': formats,
 300             'description': self._og_search_description(webpage, default=None),
 301         }