yt_dlp/extractor/iprima.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6
   7 from .common import InfoExtractor
   8 from ..utils import (
   9     determine_ext,
  10     js_to_json,
  11     urlencode_postdata,
  12     ExtractorError,
  13     parse_qs
  14 )
  15
  16
  17 class IPrimaIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?!cnn)(?:[^/]+)\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
  19     _GEO_BYPASS = False
  20     _NETRC_MACHINE = 'iprima'
  21     _LOGIN_URL = 'https://auth.iprima.cz/oauth2/login'
  22     _TOKEN_URL = 'https://auth.iprima.cz/oauth2/token'
  23     access_token = None
  24
  25     _TESTS = [{
  26         'url': 'https://prima.iprima.cz/particka/92-epizoda',
  27         'info_dict': {
  28             'id': 'p51388',
  29             'ext': 'mp4',
  30             'title': 'Partička (92)',
  31             'description': 'md5:859d53beae4609e6dd7796413f1b6cac',
  32             'upload_date': '20201103',
  33             'timestamp': 1604437480,
  34         },
  35         'params': {
  36             'skip_download': True,  # m3u8 download
  37         },
  38     }, {
  39         'url': 'http://play.iprima.cz/particka/particka-92',
  40         'only_matching': True,
  41     }, {
  42         # geo restricted
  43         'url': 'http://play.iprima.cz/closer-nove-pripady/closer-nove-pripady-iv-1',
  44         'only_matching': True,
  45     }, {
  46         'url': 'https://prima.iprima.cz/my-little-pony/mapa-znameni-2-2',
  47         'only_matching': True,
  48     }, {
  49         'url': 'https://prima.iprima.cz/porady/jak-se-stavi-sen/rodina-rathousova-praha',
  50         'only_matching': True,
  51     }, {
  52         'url': 'http://www.iprima.cz/filmy/desne-rande',
  53         'only_matching': True,
  54     }, {
  55         'url': 'https://zoom.iprima.cz/10-nejvetsich-tajemstvi-zahad/posvatna-mista-a-stavby',
  56         'only_matching': True,
  57     }, {
  58         'url': 'https://krimi.iprima.cz/mraz-0/sebevrazdy',
  59         'only_matching': True,
  60     }, {
  61         'url': 'https://cool.iprima.cz/derava-silnice-nevadi',
  62         'only_matching': True,
  63     }, {
  64         'url': 'https://love.iprima.cz/laska-az-za-hrob/slib-dany-bratrovi',
  65         'only_matching': True,
  66     }]
  67
  68     def _perform_login(self, username, password):
  69         if self.access_token:
  70             return
  71
  72         login_page = self._download_webpage(
  73             self._LOGIN_URL, None, note='Downloading login page',
  74             errnote='Downloading login page failed')
  75
  76         login_form = self._hidden_inputs(login_page)
  77
  78         login_form.update({
  79             '_email': username,
  80             '_password': password})
  81
  82         _, login_handle = self._download_webpage_handle(
  83             self._LOGIN_URL, None, data=urlencode_postdata(login_form),
  84             note='Logging in')
  85
  86         code = parse_qs(login_handle.geturl()).get('code')[0]
  87         if not code:
  88             raise ExtractorError('Login failed', expected=True)
  89
  90         token_request_data = {
  91             'scope': 'openid+email+profile+phone+address+offline_access',
  92             'client_id': 'prima_sso',
  93             'grant_type': 'authorization_code',
  94             'code': code,
  95             'redirect_uri': 'https://auth.iprima.cz/sso/auth-check'}
  96
  97         token_data = self._download_json(
  98             self._TOKEN_URL, None,
  99             note='Downloading token', errnote='Downloading token failed',
 100             data=urlencode_postdata(token_request_data))
 101
 102         self.access_token = token_data.get('access_token')
 103         if self.access_token is None:
 104             raise ExtractorError('Getting token failed', expected=True)
 105
 106     def _real_initialize(self):
 107         if not self.access_token:
 108             self.raise_login_required('Login is required to access any iPrima content', method='password')
 109
 110     def _raise_access_error(self, error_code):
 111         if error_code == 'PLAY_GEOIP_DENIED':
 112             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 113         elif error_code is not None:
 114             self.raise_no_formats('Access to stream infos forbidden', expected=True)
 115
 116     def _real_extract(self, url):
 117         video_id = self._match_id(url)
 118
 119         webpage = self._download_webpage(url, video_id)
 120
 121         title = self._html_search_meta(
 122             ['og:title', 'twitter:title'],
 123             webpage, 'title', default=None)
 124
 125         video_id = self._search_regex((
 126             r'productId\s*=\s*([\'"])(?P<id>p\d+)\1',
 127             r'pproduct_id\s*=\s*([\'"])(?P<id>p\d+)\1'),
 128             webpage, 'real id', group='id')
 129
 130         metadata = self._download_json(
 131             f'https://api.play-backend.iprima.cz/api/v1//products/id-{video_id}/play',
 132             video_id, note='Getting manifest URLs', errnote='Failed to get manifest URLs',
 133             headers={'X-OTT-Access-Token': self.access_token},
 134             expected_status=403)
 135
 136         self._raise_access_error(metadata.get('errorCode'))
 137
 138         stream_infos = metadata.get('streamInfos')
 139         formats = []
 140         if stream_infos is None:
 141             self.raise_no_formats('Reading stream infos failed', expected=True)
 142         else:
 143             for manifest in stream_infos:
 144                 manifest_type = manifest.get('type')
 145                 manifest_url = manifest.get('url')
 146                 ext = determine_ext(manifest_url)
 147                 if manifest_type == 'HLS' or ext == 'm3u8':
 148                     formats += self._extract_m3u8_formats(
 149                         manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
 150                         m3u8_id='hls', fatal=False)
 151                 elif manifest_type == 'DASH' or ext == 'mpd':
 152                     formats += self._extract_mpd_formats(
 153                         manifest_url, video_id, mpd_id='dash', fatal=False)
 154             self._sort_formats(formats)
 155
 156         final_result = self._search_json_ld(webpage, video_id) or {}
 157         final_result.update({
 158             'id': video_id,
 159             'title': title,
 160             'thumbnail': self._html_search_meta(
 161                 ['thumbnail', 'og:image', 'twitter:image'],
 162                 webpage, 'thumbnail', default=None),
 163             'formats': formats,
 164             'description': self._html_search_meta(
 165                 ['description', 'og:description', 'twitter:description'],
 166                 webpage, 'description', default=None)})
 167
 168         return final_result
 169
 170
 171 class IPrimaCNNIE(InfoExtractor):
 172     _VALID_URL = r'https?://cnn\.iprima\.cz/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 173     _GEO_BYPASS = False
 174
 175     _TESTS = [{
 176         'url': 'https://cnn.iprima.cz/porady/strunc/24072020-koronaviru-mam-plne-zuby-strasit-druhou-vlnou-je-absurdni-rika-senatorka-dernerova',
 177         'info_dict': {
 178             'id': 'p716177',
 179             'ext': 'mp4',
 180             'title': 'md5:277c6b1ed0577e51b40ddd35602ff43e',
 181         },
 182         'params': {
 183             'skip_download': 'm3u8'
 184         }
 185     }]
 186
 187     def _real_extract(self, url):
 188         video_id = self._match_id(url)
 189
 190         self._set_cookie('play.iprima.cz', 'ott_adult_confirmed', '1')
 191
 192         webpage = self._download_webpage(url, video_id)
 193
 194         title = self._og_search_title(
 195             webpage, default=None) or self._search_regex(
 196             r'<h1>([^<]+)', webpage, 'title')
 197
 198         video_id = self._search_regex(
 199             (r'<iframe[^>]+\bsrc=["\'](?:https?:)?//(?:api\.play-backend\.iprima\.cz/prehravac/embedded|prima\.iprima\.cz/[^/]+/[^/]+)\?.*?\bid=(p\d+)',
 200              r'data-product="([^"]+)">',
 201              r'id=["\']player-(p\d+)"',
 202              r'playerId\s*:\s*["\']player-(p\d+)',
 203              r'\bvideos\s*=\s*["\'](p\d+)'),
 204             webpage, 'real id')
 205
 206         playerpage = self._download_webpage(
 207             'http://play.iprima.cz/prehravac/init',
 208             video_id, note='Downloading player', query={
 209                 '_infuse': 1,
 210                 '_ts': round(time.time()),
 211                 'productId': video_id,
 212             }, headers={'Referer': url})
 213
 214         formats = []
 215
 216         def extract_formats(format_url, format_key=None, lang=None):
 217             ext = determine_ext(format_url)
 218             new_formats = []
 219             if format_key == 'hls' or ext == 'm3u8':
 220                 new_formats = self._extract_m3u8_formats(
 221                     format_url, video_id, 'mp4', entry_protocol='m3u8_native',
 222                     m3u8_id='hls', fatal=False)
 223             elif format_key == 'dash' or ext == 'mpd':
 224                 return
 225                 new_formats = self._extract_mpd_formats(
 226                     format_url, video_id, mpd_id='dash', fatal=False)
 227             if lang:
 228                 for f in new_formats:
 229                     if not f.get('language'):
 230                         f['language'] = lang
 231             formats.extend(new_formats)
 232
 233         options = self._parse_json(
 234             self._search_regex(
 235                 r'(?s)(?:TDIPlayerOptions|playerOptions)\s*=\s*({.+?});\s*\]\]',
 236                 playerpage, 'player options', default='{}'),
 237             video_id, transform_source=js_to_json, fatal=False)
 238         if options:
 239             for key, tracks in options.get('tracks', {}).items():
 240                 if not isinstance(tracks, list):
 241                     continue
 242                 for track in tracks:
 243                     src = track.get('src')
 244                     if src:
 245                         extract_formats(src, key.lower(), track.get('lang'))
 246
 247         if not formats:
 248             for _, src in re.findall(r'src["\']\s*:\s*(["\'])(.+?)\1', playerpage):
 249                 extract_formats(src)
 250
 251         if not formats and '>GEO_IP_NOT_ALLOWED<' in playerpage:
 252             self.raise_geo_restricted(countries=['CZ'], metadata_available=True)
 253
 254         self._sort_formats(formats)
 255
 256         return {
 257             'id': video_id,
 258             'title': title,
 259             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 260             'formats': formats,
 261             'description': self._og_search_description(webpage, default=None),
 262         }