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(?:tn)?\.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         'url': 'https://mediatn.cms.nova.cz/embed/EU5ELEsmOHt?autoplay=1',
  42         'info_dict': {
  43             'id': 'EU5ELEsmOHt',
  44             'ext': 'mp4',
  45             'title': 'Haptické křeslo, bionická ruka nebo roboti. Reportérka se podívala na Týden inovací',
  46             'thumbnail': r're:^https?://.*\.jpg',
  47             'duration': 1780,
  48         },
  49         'params': {'skip_download': 'm3u8'},
  50     }]
  51
  52     def _real_extract(self, url):
  53         video_id = self._match_id(url)
  54
  55         webpage = self._download_webpage(url, video_id)
  56
  57         has_drm = False
  58         duration = None
  59         formats = []
  60
  61         def process_format_list(format_list, format_id=""):
  62             nonlocal formats, has_drm
  63             if not isinstance(format_list, list):
  64                 format_list = [format_list]
  65             for format_dict in format_list:
  66                 if not isinstance(format_dict, dict):
  67                     continue
  68                 if (not self.get_param('allow_unplayable_formats')
  69                         and traverse_obj(format_dict, ('drm', 'keySystem'))):
  70                     has_drm = True
  71                     continue
  72                 format_url = url_or_none(format_dict.get('src'))
  73                 format_type = format_dict.get('type')
  74                 ext = determine_ext(format_url)
  75                 if (format_type == 'application/x-mpegURL'
  76                         or format_id == 'HLS' or ext == 'm3u8'):
  77                     formats.extend(self._extract_m3u8_formats(
  78                         format_url, video_id, 'mp4',
  79                         entry_protocol='m3u8_native', m3u8_id='hls',
  80                         fatal=False))
  81                 elif (format_type == 'application/dash+xml'
  82                       or format_id == 'DASH' or ext == 'mpd'):
  83                     formats.extend(self._extract_mpd_formats(
  84                         format_url, video_id, mpd_id='dash', fatal=False))
  85                 else:
  86                     formats.append({
  87                         'url': format_url,
  88                     })
  89
  90         player = self._search_json(
  91             r'player:', webpage, 'player', video_id, fatal=False, end_pattern=r';\s*</script>')
  92         if player:
  93             for src in traverse_obj(player, ('lib', 'source', 'sources', ...)):
  94                 process_format_list(src)
  95             duration = traverse_obj(player, ('sourceInfo', 'duration', {int_or_none}))
  96         if not formats and not has_drm:
  97             # older code path, in use before August 2023
  98             player = self._parse_json(
  99                 self._search_regex(
 100                     (r'(?:(?:replacePlaceholders|processAdTagModifier).*?:\s*)?(?:replacePlaceholders|processAdTagModifier)\s*\(\s*(?P<json>{.*?})\s*\)(?:\s*\))?\s*,',
 101                      r'Player\.init\s*\([^,]+,(?P<cndn>\s*\w+\s*\?)?\s*(?P<json>{(?(cndn).+?|.+)})\s*(?(cndn):|,\s*{.+?}\s*\)\s*;)'),
 102                     webpage, 'player', group='json'), video_id)
 103             if player:
 104                 for format_id, format_list in player['tracks'].items():
 105                     process_format_list(format_list, format_id)
 106                 duration = int_or_none(player.get('duration'))
 107
 108         if not formats and has_drm:
 109             self.report_drm(video_id)
 110
 111         title = self._og_search_title(
 112             webpage, default=None) or self._search_regex(
 113             (r'<value>(?P<title>[^<]+)',
 114              r'videoTitle\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1'), webpage,
 115             'title', group='value')
 116         thumbnail = self._og_search_thumbnail(
 117             webpage, default=None) or self._search_regex(
 118             r'poster\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1', webpage,
 119             'thumbnail', fatal=False, group='value')
 120         duration = int_or_none(self._search_regex(
 121             r'videoDuration\s*:\s*(\d+)', webpage, 'duration',
 122             default=duration))
 123
 124         return {
 125             'id': video_id,
 126             'title': title,
 127             'thumbnail': thumbnail,
 128             'duration': duration,
 129             'formats': formats,
 130         }
 131
 132
 133 class NovaIE(InfoExtractor):
 134     IE_DESC = 'TN.cz, Prásk.tv, Nova.cz, Novaplus.cz, FANDA.tv, Krásná.cz and Doma.cz'
 135     _VALID_URL = r'https?://(?:[^.]+\.)?(?P<site>tv(?:noviny)?|tn|novaplus|vymena|fanda|krasna|doma|prask)\.nova\.cz/(?:[^/]+/)+(?P<id>[^/]+?)(?:\.html|/|$)'
 136     _TESTS = [{
 137         'url': 'http://tn.nova.cz/clanek/tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci.html#player_13260',
 138         'md5': '249baab7d0104e186e78b0899c7d5f28',
 139         'info_dict': {
 140             'id': '1757139',
 141             'display_id': 'tajemstvi-ukryte-v-podzemi-specialni-nemocnice-v-prazske-krci',
 142             'ext': 'mp4',
 143             'title': 'Podzemní nemocnice v pražské Krči',
 144             'description': 'md5:f0a42dd239c26f61c28f19e62d20ef53',
 145             'thumbnail': r're:^https?://.*\.(?:jpg)',
 146         }
 147     }, {
 148         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 149         'info_dict': {
 150             'id': '1753621',
 151             'ext': 'mp4',
 152             'title': 'Zaklínač 3: Divoký hon',
 153             'description': 're:.*Pokud se stejně jako my nemůžete.*',
 154             'thumbnail': r're:https?://.*\.jpg(\?.*)?',
 155             'upload_date': '20150521',
 156         },
 157         'params': {
 158             # rtmp download
 159             'skip_download': True,
 160         },
 161         'skip': 'gone',
 162     }, {
 163         # media.cms.nova.cz embed
 164         'url': 'https://novaplus.nova.cz/porad/ulice/epizoda/18760-2180-dil',
 165         'info_dict': {
 166             'id': '8o0n0r',
 167             'ext': 'mp4',
 168             'title': '2180. díl',
 169             'thumbnail': r're:^https?://.*\.jpg',
 170             'duration': 2578,
 171         },
 172         'params': {
 173             'skip_download': True,
 174         },
 175         'add_ie': [NovaEmbedIE.ie_key()],
 176         'skip': 'CHYBA 404: STRÁNKA NENALEZENA',
 177     }, {
 178         'url': 'http://sport.tn.nova.cz/clanek/sport/hokej/nhl/zivot-jde-dal-hodnotil-po-vyrazeni-z-playoff-jiri-sekac.html',
 179         'only_matching': True,
 180     }, {
 181         'url': 'http://fanda.nova.cz/clanek/fun-and-games/krvavy-epos-zaklinac-3-divoky-hon-vychazi-vyhrajte-ho-pro-sebe.html',
 182         'only_matching': True,
 183     }, {
 184         'url': 'http://doma.nova.cz/clanek/zdravi/prijdte-se-zapsat-do-registru-kostni-drene-jiz-ve-stredu-3-cervna.html',
 185         'only_matching': True,
 186     }, {
 187         'url': 'http://prask.nova.cz/clanek/novinky/co-si-na-sobe-nase-hvezdy-nechaly-pojistit.html',
 188         'only_matching': True,
 189     }, {
 190         'url': 'http://tv.nova.cz/clanek/novinky/zivot-je-zivot-bondovsky-trailer.html',
 191         'only_matching': True,
 192     }]
 193
 194     def _real_extract(self, url):
 195         mobj = self._match_valid_url(url)
 196         display_id = mobj.group('id')
 197         site = mobj.group('site')
 198
 199         webpage = self._download_webpage(url, display_id)
 200
 201         description = clean_html(self._og_search_description(webpage, default=None))
 202         if site == 'novaplus':
 203             upload_date = unified_strdate(self._search_regex(
 204                 r'(\d{1,2}-\d{1,2}-\d{4})$', display_id, 'upload date', default=None))
 205         elif site == 'fanda':
 206             upload_date = unified_strdate(self._search_regex(
 207                 r'<span class="date_time">(\d{1,2}\.\d{1,2}\.\d{4})', webpage, 'upload date', default=None))
 208         else:
 209             upload_date = None
 210
 211         # novaplus
 212         embed_id = self._search_regex(
 213             r'<iframe[^>]+\bsrc=["\'](?:https?:)?//media\.cms\.nova\.cz/embed/([^/?#&]+)',
 214             webpage, 'embed url', default=None)
 215         if embed_id:
 216             return {
 217                 '_type': 'url_transparent',
 218                 'url': 'https://media.cms.nova.cz/embed/%s' % embed_id,
 219                 'ie_key': NovaEmbedIE.ie_key(),
 220                 'id': embed_id,
 221                 'description': description,
 222                 'upload_date': upload_date
 223             }
 224
 225         video_id = self._search_regex(
 226             [r"(?:media|video_id)\s*:\s*'(\d+)'",
 227              r'media=(\d+)',
 228              r'id="article_video_(\d+)"',
 229              r'id="player_(\d+)"'],
 230             webpage, 'video id')
 231
 232         config_url = self._search_regex(
 233             r'src="(https?://(?:tn|api)\.nova\.cz/bin/player/videojs/config\.php\?[^"]+)"',
 234             webpage, 'config url', default=None)
 235         config_params = {}
 236
 237         if not config_url:
 238             player = self._parse_json(
 239                 self._search_regex(
 240                     r'(?s)Player\s*\(.+?\s*,\s*({.+?\bmedia\b["\']?\s*:\s*["\']?\d+.+?})\s*\)', webpage,
 241                     'player', default='{}'),
 242                 video_id, transform_source=js_to_json, fatal=False)
 243             if player:
 244                 config_url = url_or_none(player.get('configUrl'))
 245                 params = player.get('configParams')
 246                 if isinstance(params, dict):
 247                     config_params = params
 248
 249         if not config_url:
 250             DEFAULT_SITE_ID = '23000'
 251             SITES = {
 252                 'tvnoviny': DEFAULT_SITE_ID,
 253                 'novaplus': DEFAULT_SITE_ID,
 254                 'vymena': DEFAULT_SITE_ID,
 255                 'krasna': DEFAULT_SITE_ID,
 256                 'fanda': '30',
 257                 'tn': '30',
 258                 'doma': '30',
 259             }
 260
 261             site_id = self._search_regex(
 262                 r'site=(\d+)', webpage, 'site id', default=None) or SITES.get(
 263                 site, DEFAULT_SITE_ID)
 264
 265             config_url = 'https://api.nova.cz/bin/player/videojs/config.php'
 266             config_params = {
 267                 'site': site_id,
 268                 'media': video_id,
 269                 'quality': 3,
 270                 'version': 1,
 271             }
 272
 273         config = self._download_json(
 274             config_url, display_id,
 275             'Downloading config JSON', query=config_params,
 276             transform_source=lambda s: s[s.index('{'):s.rindex('}') + 1])
 277
 278         mediafile = config['mediafile']
 279         video_url = mediafile['src']
 280
 281         m = re.search(r'^(?P<url>rtmpe?://[^/]+/(?P<app>[^/]+?))/&*(?P<playpath>.+)$', video_url)
 282         if m:
 283             formats = [{
 284                 'url': m.group('url'),
 285                 'app': m.group('app'),
 286                 'play_path': m.group('playpath'),
 287                 'player_path': 'http://tvnoviny.nova.cz/static/shared/app/videojs/video-js.swf',
 288                 'ext': 'flv',
 289             }]
 290         else:
 291             formats = [{
 292                 'url': video_url,
 293             }]
 294
 295         title = mediafile.get('meta', {}).get('title') or self._og_search_title(webpage)
 296         thumbnail = config.get('poster')
 297
 298         return {
 299             'id': video_id,
 300             'display_id': display_id,
 301             'title': title,
 302             'description': description,
 303             'upload_date': upload_date,
 304             'thumbnail': thumbnail,
 305             'formats': formats,
 306         }