yt_dlp/extractor/nova.py

   1 import re
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     clean_html,
   6     determine_ext,
   7     int_or_none,
   8     js_to_json,
   9     traverse_obj,
  10     unified_strdate,
  11     url_or_none,
  12 )
  13
  14
  15 class NovaEmbedIE(InfoExtractor):
  16     _VALID_URL = r'https?://media\.cms\.nova\.cz/embed/(?P<id>[^/?#&]+)'
  17     _TESTS = [{
  18         'url': 'https://media.cms.nova.cz/embed/8o0n0r?autoplay=1',
  19         'info_dict': {
  20             'id': '8o0n0r',
  21             'title': '2180. díl',
  22             'thumbnail': r're:^https?://.*\.jpg',
  23             'duration': 2578,
  24         },
  25         'params': {
  26             'skip_download': True,
  27             'ignore_no_formats_error': True,
  28         },
  29         'expected_warnings': ['DRM protected', 'Requested format is not available'],
  30     }, {
  31         'url': 'https://media.cms.nova.cz/embed/KybpWYvcgOa',
  32         'info_dict': {
  33             'id': 'KybpWYvcgOa',
  34             'ext': 'mp4',
  35             'title': 'Borhyová oslavila 60? Soutěžící z pořadu odboural moderátora Ondřeje Sokola',
  36             'thumbnail': r're:^https?://.*\.jpg',
  37             'duration': 114,
  38         },
  39         'params': {'skip_download': 'm3u8'},
  40     }]
  41
  42     def _real_extract(self, url):
  43         video_id = self._match_id(url)
  44
  45         webpage = self._download_webpage(url, video_id)
  46
  47         has_drm = False
  48         duration = None
  49         formats = []
  50
  51         def process_format_list(format_list, format_id=""):
  52             nonlocal formats, has_drm
  53             if not isinstance(format_list, list):
  54                 format_list = [format_list]
  55             for format_dict in format_list:
  56                 if not isinstance(format_dict, dict):
  57                     continue
  58                 if (not self.get_param('allow_unplayable_formats')
  59                         and traverse_obj(format_dict, ('drm', 'keySystem'))):
  60                     has_drm = True
  61                     continue
  62                 format_url = url_or_none(format_dict.get('src'))
  63                 format_type = format_dict.get('type')
  64                 ext = determine_ext(format_url)
  65                 if (format_type == 'application/x-mpegURL'
  66                         or format_id == 'HLS' or ext == 'm3u8'):
  67                     formats.extend(self._extract_m3u8_formats(
  68                         format_url, video_id, 'mp4',
  69                         entry_protocol='m3u8_native', m3u8_id='hls',
  70                         fatal=False))
  71                 elif (format_type == 'application/dash+xml'
  72                       or format_id == 'DASH' or ext == 'mpd'):
  73                     formats.extend(self._extract_mpd_formats(
  74                         format_url, video_id, mpd_id='dash', fatal=False))
  75                 else:
  76                     formats.append({
  77                         'url': format_url,
  78                     })
  79
  80         player = self._search_json(
  81             r'player:', webpage, 'player', video_id, fatal=False, end_pattern=r';\s*</script>')
  82         if player:
  83             for src in traverse_obj(player, ('lib', 'source', 'sources', ...)):
  84                 process_format_list(src)
  85             duration = traverse_obj(player, ('sourceInfo', 'duration', {int_or_none}))
  86         if not formats and not has_drm:
  87             # older code path, in use before August 2023
  88             player = self._parse_json(
  89                 self._search_regex(
  90                     (r'(?:(?:replacePlaceholders|processAdTagModifier).*?:\s*)?(?:replacePlaceholders|processAdTagModifier)\s*\(\s*(?P<json>{.*?})\s*\)(?:\s*\))?\s*,',
  91                      r'Player\.init\s*\([^,]+,(?P<cndn>\s*\w+\s*\?)?\s*(?P<json>{(?(cndn).+?|.+)})\s*(?(cndn):|,\s*{.+?}\s*\)\s*;)'),
  92                     webpage, 'player', group='json'), video_id)
  93             if player:
  94                 for format_id, format_list in player['tracks'].items():
  95                     process_format_list(format_list, format_id)
  96                 duration = int_or_none(player.get('duration'))
  97
  98         if not formats and has_drm:
  99             self.report_drm(video_id)
 100
 101         title = self._og_search_title(
 102             webpage, default=None) or self._search_regex(
 103             (r'<value>(?P<title>[^<]+)',
 104              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
 105             'title', group='value')
 106         thumbnail = self._og_search_thumbnail(
 107             webpage, default=None) or self._search_regex(
 108             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
 109             'thumbnail', fatal=False, group='value')
 110         duration = int_or_none(self._search_regex(
 111             r'videoDuration\s*:\s*(\d+)', webpage, 'duration',
 112             default=duration))
 113
 114         return {
 115             'id': video_id,
 116             'title': title,
 117             'thumbnail': thumbnail,
 118             'duration': duration,
 119             'formats': formats,
 120         }
 121
 122
 123 class NovaIE(InfoExtractor):
 124     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
 125     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
 126     _TESTS = [{
 127         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 128         'md5': '249baab7d0104e186e78b0899c7d5f28',
 129         'info_dict': {
 130             'id': '1757139',
 131             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 132             'ext': 'mp4',
 133             'title': 'Podzemní nemocnice v pražské Krči',
 134             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 135             'thumbnail': r're:^https?://.*\.(?:jpg)',
 136         }
 137     }, {
 138         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 139         'info_dict': {
 140             'id': '1753621',
 141             'ext': 'mp4',
 142             'title': 'Zaklínač 3: Divoký hon',
 143             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 144             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 145             'upload_date': '20150521',
 146         },
 147         'params': {
 148             # rtmp download
 149             'skip_download': True,
 150         },
 151         'skip': 'gone',
 152     }, {
 153         # media.cms.nova.cz embed
 154         'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
 155         'info_dict': {
 156             'id': '8o0n0r',
 157             'ext': 'mp4',
 158             'title': '2180. díl',
 159             'thumbnail': r're:^https?://.*\.jpg',
 160             'duration': 2578,
 161         },
 162         'params': {
 163             'skip_download': True,
 164         },
 165         'add_ie': [NovaEmbedIE.ie_key()],
 166         'skip': 'CHYBA 404: STRÁNKA NENALEZENA',
 167     }, {
 168         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 169         'only_matching': True,
 170     }, {
 171         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 172         'only_matching': True,
 173     }, {
 174         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 175         'only_matching': True,
 176     }, {
 177         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 178         'only_matching': True,
 179     }, {
 180         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 181         'only_matching': True,
 182     }]
 183
 184     def _real_extract(self, url):
 185         mobj = self._match_valid_url(url)
 186         display_id = mobj.group('id')
 187         site = mobj.group('site')
 188
 189         webpage = self._download_webpage(url, display_id)
 190
 191         description = clean_html(self._og_search_description(webpage, default=None))
 192         if site == 'novaplus':
 193             upload_date = unified_strdate(self._search_regex(
 194                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 195         elif site == 'fanda':
 196             upload_date = unified_strdate(self._search_regex(
 197                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 198         else:
 199             upload_date = None
 200
 201         # novaplus
 202         embed_id = self._search_regex(
 203             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 204             webpage, 'embed url', default=None)
 205         if embed_id:
 206             return {
 207                 '_type': 'url_transparent',
 208                 'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
 209                 'ie_key': NovaEmbedIE.ie_key(),
 210                 'id': embed_id,
 211                 'description': description,
 212                 'upload_date': upload_date
 213             }
 214
 215         video_id = self._search_regex(
 216             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 217              r'media=(\d+)',
 218              r'id="article_video_(\d+)"',
 219              r'id="player_(\d+)"'],
 220             webpage, 'video id')
 221
 222         config_url = self._search_regex(
 223             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 224             webpage, 'config url', default=None)
 225         config_params = {}
 226
 227         if not config_url:
 228             player = self._parse_json(
 229                 self._search_regex(
 230                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 231                     'player', default='{}'),
 232                 video_id, transform_source=js_to_json, fatal=False)
 233             if player:
 234                 config_url = url_or_none(player.get('configUrl'))
 235                 params = player.get('configParams')
 236                 if isinstance(params, dict):
 237                     config_params = params
 238
 239         if not config_url:
 240             DEFAULT_SITE_ID = '23000'
 241             SITES = {
 242                 'tvnoviny': DEFAULT_SITE_ID,
 243                 'novaplus': DEFAULT_SITE_ID,
 244                 'vymena': DEFAULT_SITE_ID,
 245                 'krasna': DEFAULT_SITE_ID,
 246                 'fanda': '30',
 247                 'tn': '30',
 248                 'doma': '30',
 249             }
 250
 251             site_id = self._search_regex(
 252                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 253                 site, DEFAULT_SITE_ID)
 254
 255             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 256             config_params = {
 257                 'site': site_id,
 258                 'media': video_id,
 259                 'quality': 3,
 260                 'version': 1,
 261             }
 262
 263         config = self._download_json(
 264             config_url, display_id,
 265             'Downloading config JSON', query=config_params,
 266             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 267
 268         mediafile = config['mediafile']
 269         video_url = mediafile['src']
 270
 271         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 272         if m:
 273             formats = [{
 274                 'url': m.group('url'),
 275                 'app': m.group('app'),
 276                 'play_path': m.group('playpath'),
 277                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 278                 'ext': 'flv',
 279             }]
 280         else:
 281             formats = [{
 282                 'url': video_url,
 283             }]
 284
 285         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 286         thumbnail = config.get('poster')
 287
 288         return {
 289             'id': video_id,
 290             'display_id': display_id,
 291             'title': title,
 292             'description': description,
 293             'upload_date': upload_date,
 294             'thumbnail': thumbnail,
 295             'formats': formats,
 296         }