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 )
  12
  13
  14 class IPrimaIE(InfoExtractor):
  15     _VALID_URL = r'https?://(?!cnn)(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  16     _GEO_BYPASS = False
  17     _NETRC_MACHINE = 'iprima'
  18     _LOGIN_URL = 'https://auth.iprima.cz/oauth2/login'
  19     _TOKEN_URL = 'https://auth.iprima.cz/oauth2/token'
  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             self._LOGIN_URL, 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         _, login_handle = self._download_webpage_handle(
  80             self._LOGIN_URL, None, data=urlencode_postdata(login_form),
  81             note='Logging in')
  82
  83         code = parse_qs(login_handle.geturl()).get('code')[0]
  84         if not code:
  85             raise ExtractorError('Login failed', expected=True)
  86
  87         token_request_data = {
  88             'scope': 'openid+email+profile+phone+address+offline_access',
  89             'client_id': 'prima_sso',
  90             'grant_type': 'authorization_code',
  91             'code': code,
  92             'redirect_uri': 'https://auth.iprima.cz/sso/auth-check'}
  93
  94         token_data = self._download_json(
  95             self._TOKEN_URL, None,
  96             note='Downloading token', errnote='Downloading token failed',
  97             data=urlencode_postdata(token_request_data))
  98
  99         self.access_token = token_data.get('access_token')
 100         if self.access_token is None:
 101             raise ExtractorError('Getting token failed', expected=True)
 102
 103     def _real_initialize(self):
 104         if not self.access_token:
 105             self.raise_login_required('Login is required to access any iPrima content', method='password')
 106
 107     def _raise_access_error(self, error_code):
 108         if error_code == 'PLAY_GEOIP_DENIED':
 109             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 110         elif error_code is not None:
 111             self.raise_no_formats('Access to stream infos forbidden', expected=True)
 112
 113     def _real_extract(self, url):
 114         video_id = self._match_id(url)
 115
 116         webpage = self._download_webpage(url, video_id)
 117
 118         title = self._html_search_meta(
 119             ['og:title', 'twitter:title'],
 120             webpage, 'title', default=None)
 121
 122         video_id = self._search_regex((
 123             r'productId\s*=\s*([\'"])(?P<id>p\d+)\1',
 124             r'pproduct_id\s*=\s*([\'"])(?P<id>p\d+)\1'),
 125             webpage, 'real id', group='id')
 126
 127         metadata = self._download_json(
 128             f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
 129             video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
 130             headers={'X-OTT-Access-Token': self.access_token},
 131             expected_status=403)
 132
 133         self._raise_access_error(metadata.get('errorCode'))
 134
 135         stream_infos = metadata.get('streamInfos')
 136         formats = []
 137         if stream_infos is None:
 138             self.raise_no_formats('Reading stream infos failed', expected=True)
 139         else:
 140             for manifest in stream_infos:
 141                 manifest_type = manifest.get('type')
 142                 manifest_url = manifest.get('url')
 143                 ext = determine_ext(manifest_url)
 144                 if manifest_type == 'HLS' or ext == 'm3u8':
 145                     formats += self._extract_m3u8_formats(
 146                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 147                         m3u8_id='hls', fatal=False)
 148                 elif manifest_type == 'DASH' or ext == 'mpd':
 149                     formats += self._extract_mpd_formats(
 150                         manifest_url, video_id, mpd_id='dash', fatal=False)
 151
 152         final_result = self._search_json_ld(webpage, video_id, default={})
 153         final_result.update({
 154             'id': video_id,
 155             'title': title,
 156             'thumbnail': self._html_search_meta(
 157                 ['thumbnail', 'og:image', 'twitter:image'],
 158                 webpage, 'thumbnail', default=None),
 159             'formats': formats,
 160             'description': self._html_search_meta(
 161                 ['description', 'og:description', 'twitter:description'],
 162                 webpage, 'description', default=None)})
 163
 164         return final_result
 165
 166
 167 class IPrimaCNNIE(InfoExtractor):
 168     _VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 169     _GEO_BYPASS = False
 170
 171     _TESTS = [{
 172         'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
 173         'info_dict': {
 174             'id': 'p716177',
 175             'ext': 'mp4',
 176             'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
 177         },
 178         'params': {
 179             'skip_download': 'm3u8'
 180         }
 181     }]
 182
 183     def _real_extract(self, url):
 184         video_id = self._match_id(url)
 185
 186         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
 187
 188         webpage = self._download_webpage(url, video_id)
 189
 190         title = self._og_search_title(
 191             webpage, default=None) or self._search_regex(
 192             r'<h1>([^<]+)', webpage, 'title')
 193
 194         video_id = self._search_regex(
 195             (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
 196              r'data-product="([^"]+)">',
 197              r'id=["\']player-(p\d+)"',
 198              r'playerId\s*:\s*["\']player-(p\d+)',
 199              r'\bvideos\s*=\s*["\'](p\d+)'),
 200             webpage, 'real id')
 201
 202         playerpage = self._download_webpage(
 203             'http://play.iprima.cz/prehravac/init',
 204             video_id, note='Downloading player', query={
 205                 '_infuse': 1,
 206                 '_ts': round(time.time()),
 207                 'productId': video_id,
 208             }, headers={'Referer': url})
 209
 210         formats = []
 211
 212         def extract_formats(format_url, format_key=None, lang=None):
 213             ext = determine_ext(format_url)
 214             new_formats = []
 215             if format_key == 'hls' or ext == 'm3u8':
 216                 new_formats = self._extract_m3u8_formats(
 217                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 218                     m3u8_id='hls', fatal=False)
 219             elif format_key == 'dash' or ext == 'mpd':
 220                 return
 221                 new_formats = self._extract_mpd_formats(
 222                     format_url, video_id, mpd_id='dash', fatal=False)
 223             if lang:
 224                 for f in new_formats:
 225                     if not f.get('language'):
 226                         f['language'] = lang
 227             formats.extend(new_formats)
 228
 229         options = self._parse_json(
 230             self._search_regex(
 231                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
 232                 playerpage, 'player options', default='{}'),
 233             video_id, transform_source=js_to_json, fatal=False)
 234         if options:
 235             for key, tracks in options.get('tracks', {}).items():
 236                 if not isinstance(tracks, list):
 237                     continue
 238                 for track in tracks:
 239                     src = track.get('src')
 240                     if src:
 241                         extract_formats(src, key.lower(), track.get('lang'))
 242
 243         if not formats:
 244             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
 245                 extract_formats(src)
 246
 247         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
 248             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 249
 250         return {
 251             'id': video_id,
 252             'title': title,
 253             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 254             'formats': formats,
 255             'description': self._og_search_description(webpage, default=None),
 256         }