yt_dlp/extractor/iprima.py

   1 import re
   2 import time
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     js_to_json,
   8     urlencode_postdata,
   9     ExtractorError,
  10     parse_qs,
  11     traverse_obj
  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': 'md5:859d53beae4609e6dd7796413f1b6cac',
  29             'upload_date': '20201103',
  30             'timestamp': 1604437480,
  31         },
  32         'params': {
  33             'skip_download': True,  # m3u8 download
  34         },
  35     }, {
  36         'url': 'http://play.iprima.cz/particka/particka-92',
  37         'only_matching': True,
  38     }, {
  39         # geo restricted
  40         'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
  41         'only_matching': True,
  42     }, {
  43         'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
  44         'only_matching': True,
  45     }, {
  46         'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
  47         'only_matching': True,
  48     }, {
  49         'url': 'http://www.iprima.cz/filmy/desne-rande',
  50         'only_matching': True,
  51     }, {
  52         'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
  53         'only_matching': True,
  54     }, {
  55         'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
  56         'only_matching': True,
  57     }, {
  58         'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
  59         'only_matching': True,
  60     }, {
  61         'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
  62         'only_matching': True,
  63     }]
  64
  65     def _perform_login(self, username, password):
  66         if self.access_token:
  67             return
  68
  69         login_page = self._download_webpage(
  70             f'{self._AUTH_ROOT}/oauth2/login', None, note='Downloading login page',
  71             errnote='Downloading login page failed')
  72
  73         login_form = self._hidden_inputs(login_page)
  74
  75         login_form.update({
  76             '_email': username,
  77             '_password': password})
  78
  79         profile_select_html, login_handle = self._download_webpage_handle(
  80             f'{self._AUTH_ROOT}/oauth2/login', None, data=urlencode_postdata(login_form),
  81             note='Logging in')
  82
  83         # a profile may need to be selected first, even when there is only a single one
  84         if '/profile-select' in login_handle.geturl():
  85             profile_id = self._search_regex(
  86                 r'data-identifier\s*=\s*["\']?(\w+)', profile_select_html, 'profile id')
  87
  88             login_handle = self._request_webpage(
  89                 f'{self._AUTH_ROOT}/user/profile-select-perform/{profile_id}', None,
  90                 query={'continueUrl': '/user/login?redirect_uri=/user/'}, note='Selecting profile')
  91
  92         code = traverse_obj(login_handle.geturl(), ({parse_qs}, 'code', 0))
  93         if not code:
  94             raise ExtractorError('Login failed', expected=True)
  95
  96         token_request_data = {
  97             'scope': 'openid+email+profile+phone+address+offline_access',
  98             'client_id': 'prima_sso',
  99             'grant_type': 'authorization_code',
 100             'code': code,
 101             'redirect_uri': f'{self._AUTH_ROOT}/sso/auth-check'}
 102
 103         token_data = self._download_json(
 104             f'{self._AUTH_ROOT}/oauth2/token', None,
 105             note='Downloading token', errnote='Downloading token failed',
 106             data=urlencode_postdata(token_request_data))
 107
 108         self.access_token = token_data.get('access_token')
 109         if self.access_token is None:
 110             raise ExtractorError('Getting token failed', expected=True)
 111
 112     def _real_initialize(self):
 113         if not self.access_token:
 114             self.raise_login_required('Login is required to access any iPrima content', method='password')
 115
 116     def _raise_access_error(self, error_code):
 117         if error_code == 'PLAY_GEOIP_DENIED':
 118             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 119         elif error_code is not None:
 120             self.raise_no_formats('Access to stream infos forbidden', expected=True)
 121
 122     def _real_extract(self, url):
 123         video_id = self._match_id(url)
 124
 125         webpage = self._download_webpage(url, video_id)
 126
 127         title = self._html_extract_title(webpage) or self._html_search_meta(
 128             ['og:title', 'twitter:title'],
 129             webpage, 'title', default=None)
 130
 131         video_id = self._search_regex((
 132             r'productId\s*=\s*([\'"])(?P<id>p\d+)\1',
 133             r'pproduct_id\s*=\s*([\'"])(?P<id>p\d+)\1',
 134         ), webpage, 'real id', group='id', default=None)
 135
 136         if not video_id:
 137             nuxt_data = self._search_nuxt_data(webpage, video_id, traverse='data')
 138             video_id = traverse_obj(
 139                 nuxt_data, (..., 'content', 'additionals', 'videoPlayId', {str}), get_all=False)
 140
 141         if not video_id:
 142             self.raise_no_formats('Unable to extract video ID from webpage')
 143
 144         metadata = self._download_json(
 145             f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
 146             video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
 147             headers={'X-OTT-Access-Token': self.access_token},
 148             expected_status=403)
 149
 150         self._raise_access_error(metadata.get('errorCode'))
 151
 152         stream_infos = metadata.get('streamInfos')
 153         formats = []
 154         if stream_infos is None:
 155             self.raise_no_formats('Reading stream infos failed', expected=True)
 156         else:
 157             for manifest in stream_infos:
 158                 manifest_type = manifest.get('type')
 159                 manifest_url = manifest.get('url')
 160                 ext = determine_ext(manifest_url)
 161                 if manifest_type == 'HLS' or ext == 'm3u8':
 162                     formats += self._extract_m3u8_formats(
 163                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 164                         m3u8_id='hls', fatal=False)
 165                 elif manifest_type == 'DASH' or ext == 'mpd':
 166                     formats += self._extract_mpd_formats(
 167                         manifest_url, video_id, mpd_id='dash', fatal=False)
 168
 169         final_result = self._search_json_ld(webpage, video_id, default={})
 170         final_result.update({
 171             'id': video_id,
 172             'title': title,
 173             'thumbnail': self._html_search_meta(
 174                 ['thumbnail', 'og:image', 'twitter:image'],
 175                 webpage, 'thumbnail', default=None),
 176             'formats': formats,
 177             'description': self._html_search_meta(
 178                 ['description', 'og:description', 'twitter:description'],
 179                 webpage, 'description', default=None)})
 180
 181         return final_result
 182
 183
 184 class IPrimaCNNIE(InfoExtractor):
 185     _VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 186     _GEO_BYPASS = False
 187
 188     _TESTS = [{
 189         'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
 190         'info_dict': {
 191             'id': 'p716177',
 192             'ext': 'mp4',
 193             'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
 194         },
 195         'params': {
 196             'skip_download': 'm3u8'
 197         }
 198     }]
 199
 200     def _real_extract(self, url):
 201         video_id = self._match_id(url)
 202
 203         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
 204
 205         webpage = self._download_webpage(url, video_id)
 206
 207         title = self._og_search_title(
 208             webpage, default=None) or self._search_regex(
 209             r'<h1>([^<]+)', webpage, 'title')
 210
 211         video_id = self._search_regex(
 212             (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
 213              r'data-product="([^"]+)">',
 214              r'id=["\']player-(p\d+)"',
 215              r'playerId\s*:\s*["\']player-(p\d+)',
 216              r'\bvideos\s*=\s*["\'](p\d+)'),
 217             webpage, 'real id')
 218
 219         playerpage = self._download_webpage(
 220             'http://play.iprima.cz/prehravac/init',
 221             video_id, note='Downloading player', query={
 222                 '_infuse': 1,
 223                 '_ts': round(time.time()),
 224                 'productId': video_id,
 225             }, headers={'Referer': url})
 226
 227         formats = []
 228
 229         def extract_formats(format_url, format_key=None, lang=None):
 230             ext = determine_ext(format_url)
 231             new_formats = []
 232             if format_key == 'hls' or ext == 'm3u8':
 233                 new_formats = self._extract_m3u8_formats(
 234                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 235                     m3u8_id='hls', fatal=False)
 236             elif format_key == 'dash' or ext == 'mpd':
 237                 return
 238                 new_formats = self._extract_mpd_formats(
 239                     format_url, video_id, mpd_id='dash', fatal=False)
 240             if lang:
 241                 for f in new_formats:
 242                     if not f.get('language'):
 243                         f['language'] = lang
 244             formats.extend(new_formats)
 245
 246         options = self._parse_json(
 247             self._search_regex(
 248                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
 249                 playerpage, 'player options', default='{}'),
 250             video_id, transform_source=js_to_json, fatal=False)
 251         if options:
 252             for key, tracks in options.get('tracks', {}).items():
 253                 if not isinstance(tracks, list):
 254                     continue
 255                 for track in tracks:
 256                     src = track.get('src')
 257                     if src:
 258                         extract_formats(src, key.lower(), track.get('lang'))
 259
 260         if not formats:
 261             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
 262                 extract_formats(src)
 263
 264         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
 265             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 266
 267         return {
 268             'id': video_id,
 269             'title': title,
 270             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 271             'formats': formats,
 272             'description': self._og_search_description(webpage, default=None),
 273         }