yt_dlp/extractor/drtv.py

   1 import binascii
   2 import hashlib
   3 import re
   4
   5
   6 from .common import InfoExtractor
   7 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
   8 from ..compat import compat_urllib_parse_unquote
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     float_or_none,
  13     mimetype2ext,
  14     str_or_none,
  15     try_get,
  16     unified_timestamp,
  17     update_url_query,
  18     url_or_none,
  19 )
  20
  21
  22 class DRTVIE(InfoExtractor):
  23     _VALID_URL = r'''(?x)
  24                     https?://
  25                         (?:
  26                             (?:www\.)?dr\.dk/(?:tv/se|nyheder|radio(?:/ondemand)?)/(?:[^/]+/)*|
  27                             (?:www\.)?(?:dr\.dk|dr-massive\.com)/drtv/(?:se|episode|program)/
  28                         )
  29                         (?P<id>[\da-z_-]+)
  30                     '''
  31     _GEO_BYPASS = False
  32     _GEO_COUNTRIES = ['DK']
  33     IE_NAME = 'drtv'
  34     _TESTS = [{
  35         'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
  36         'md5': '25e659cccc9a2ed956110a299fdf5983',
  37         'info_dict': {
  38             'id': 'klassen-darlig-taber-10',
  39             'ext': 'mp4',
  40             'title': 'Klassen - Dårlig taber (10)',
  41             'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa',
  42             'timestamp': 1539085800,
  43             'upload_date': '20181009',
  44             'duration': 606.84,
  45             'series': 'Klassen',
  46             'season': 'Klassen I',
  47             'season_number': 1,
  48             'season_id': 'urn:dr:mu:bundle:57d7e8216187a4031cfd6f6b',
  49             'episode': 'Episode 10',
  50             'episode_number': 10,
  51             'release_year': 2016,
  52         },
  53         'expected_warnings': ['Unable to download f4m manifest'],
  54     }, {
  55         # embed
  56         'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
  57         'info_dict': {
  58             'id': 'urn:dr:mu:programcard:57c926176187a50a9c6e83c6',
  59             'ext': 'mp4',
  60             'title': 'christiania pusher street ryddes drdkrjpo',
  61             'description': 'md5:2a71898b15057e9b97334f61d04e6eb5',
  62             'timestamp': 1472800279,
  63             'upload_date': '20160902',
  64             'duration': 131.4,
  65         },
  66         'params': {
  67             'skip_download': True,
  68         },
  69         'expected_warnings': ['Unable to download f4m manifest'],
  70     }, {
  71         # with SignLanguage formats
  72         'url': 'https://www.dr.dk/tv/se/historien-om-danmark/-/historien-om-danmark-stenalder',
  73         'info_dict': {
  74             'id': 'historien-om-danmark-stenalder',
  75             'ext': 'mp4',
  76             'title': 'Historien om Danmark: Stenalder',
  77             'description': 'md5:8c66dcbc1669bbc6f873879880f37f2a',
  78             'timestamp': 1546628400,
  79             'upload_date': '20190104',
  80             'duration': 3502.56,
  81             'formats': 'mincount:20',
  82         },
  83         'params': {
  84             'skip_download': True,
  85         },
  86     }, {
  87         'url': 'https://www.dr.dk/radio/p4kbh/regionale-nyheder-kh4/p4-nyheder-2019-06-26-17-30-9',
  88         'only_matching': True,
  89     }, {
  90         'url': 'https://www.dr.dk/drtv/se/bonderoeven_71769',
  91         'info_dict': {
  92             'id': '00951930010',
  93             'ext': 'mp4',
  94             'title': 'Bonderøven (1:8)',
  95             'description': 'md5:3cf18fc0d3b205745d4505f896af8121',
  96             'timestamp': 1546542000,
  97             'upload_date': '20190103',
  98             'duration': 2576.6,
  99         },
 100         'params': {
 101             'skip_download': True,
 102         },
 103     }, {
 104         'url': 'https://www.dr.dk/drtv/episode/bonderoeven_71769',
 105         'only_matching': True,
 106     }, {
 107         'url': 'https://dr-massive.com/drtv/se/bonderoeven_71769',
 108         'only_matching': True,
 109     }, {
 110         'url': 'https://www.dr.dk/drtv/program/jagten_220924',
 111         'only_matching': True,
 112     }]
 113
 114     def _real_extract(self, url):
 115         video_id = self._match_id(url)
 116
 117         webpage = self._download_webpage(url, video_id)
 118
 119         if '>Programmet er ikke længere tilgængeligt' in webpage:
 120             raise ExtractorError(
 121                 'Video %s is not available' % video_id, expected=True)
 122
 123         video_id = self._search_regex(
 124             (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
 125              r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
 126             webpage, 'video id', default=None)
 127
 128         if not video_id:
 129             video_id = self._search_regex(
 130                 r'(urn(?:%3A|:)dr(?:%3A|:)mu(?:%3A|:)programcard(?:%3A|:)[\da-f]+)',
 131                 webpage, 'urn', default=None)
 132             if video_id:
 133                 video_id = compat_urllib_parse_unquote(video_id)
 134
 135         _PROGRAMCARD_BASE = 'https://www.dr.dk/mu-online/api/1.4/programcard'
 136         query = {'expanded': 'true'}
 137
 138         if video_id:
 139             programcard_url = '%s/%s' % (_PROGRAMCARD_BASE, video_id)
 140         else:
 141             programcard_url = _PROGRAMCARD_BASE
 142             page = self._parse_json(
 143                 self._search_regex(
 144                     r'data\s*=\s*({.+?})\s*(?:;|</script)', webpage,
 145                     'data'), '1')['cache']['page']
 146             page = page[list(page.keys())[0]]
 147             item = try_get(
 148                 page, (lambda x: x['item'], lambda x: x['entries'][0]['item']),
 149                 dict)
 150             video_id = item['customId'].split(':')[-1]
 151             query['productionnumber'] = video_id
 152
 153         data = self._download_json(
 154             programcard_url, video_id, 'Downloading video JSON', query=query)
 155
 156         title = str_or_none(data.get('Title')) or re.sub(
 157             r'\s*\|\s*(?:TV\s*\|\s*DR|DRTV)$', '',
 158             self._og_search_title(webpage))
 159         description = self._og_search_description(
 160             webpage, default=None) or data.get('Description')
 161
 162         timestamp = unified_timestamp(
 163             data.get('PrimaryBroadcastStartTime') or data.get('SortDateTime'))
 164
 165         thumbnail = None
 166         duration = None
 167
 168         restricted_to_denmark = False
 169
 170         formats = []
 171         subtitles = {}
 172
 173         assets = []
 174         primary_asset = data.get('PrimaryAsset')
 175         if isinstance(primary_asset, dict):
 176             assets.append(primary_asset)
 177         secondary_assets = data.get('SecondaryAssets')
 178         if isinstance(secondary_assets, list):
 179             for secondary_asset in secondary_assets:
 180                 if isinstance(secondary_asset, dict):
 181                     assets.append(secondary_asset)
 182
 183         def hex_to_bytes(hex):
 184             return binascii.a2b_hex(hex.encode('ascii'))
 185
 186         def decrypt_uri(e):
 187             n = int(e[2:10], 16)
 188             a = e[10 + n:]
 189             data = hex_to_bytes(e[10:10 + n])
 190             key = hashlib.sha256(('%s:sRBzYNXBzkKgnjj8pGtkACch' % a).encode('utf-8')).digest()
 191             iv = hex_to_bytes(a)
 192             decrypted = unpad_pkcs7(aes_cbc_decrypt_bytes(data, key, iv))
 193             return decrypted.decode('utf-8').split('?')[0]
 194
 195         for asset in assets:
 196             kind = asset.get('Kind')
 197             if kind == 'Image':
 198                 thumbnail = url_or_none(asset.get('Uri'))
 199             elif kind in ('VideoResource', 'AudioResource'):
 200                 duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
 201                 restricted_to_denmark = asset.get('RestrictedToDenmark')
 202                 asset_target = asset.get('Target')
 203                 for link in asset.get('Links', []):
 204                     uri = link.get('Uri')
 205                     if not uri:
 206                         encrypted_uri = link.get('EncryptedUri')
 207                         if not encrypted_uri:
 208                             continue
 209                         try:
 210                             uri = decrypt_uri(encrypted_uri)
 211                         except Exception:
 212                             self.report_warning(
 213                                 'Unable to decrypt EncryptedUri', video_id)
 214                             continue
 215                     uri = url_or_none(uri)
 216                     if not uri:
 217                         continue
 218                     target = link.get('Target')
 219                     format_id = target or ''
 220                     if asset_target in ('SpokenSubtitles', 'SignLanguage', 'VisuallyInterpreted'):
 221                         preference = -1
 222                         format_id += '-%s' % asset_target
 223                     elif asset_target == 'Default':
 224                         preference = 1
 225                     else:
 226                         preference = None
 227                     if target == 'HDS':
 228                         f4m_formats = self._extract_f4m_formats(
 229                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
 230                             video_id, preference, f4m_id=format_id, fatal=False)
 231                         if kind == 'AudioResource':
 232                             for f in f4m_formats:
 233                                 f['vcodec'] = 'none'
 234                         formats.extend(f4m_formats)
 235                     elif target == 'HLS':
 236                         formats.extend(self._extract_m3u8_formats(
 237                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
 238                             quality=preference, m3u8_id=format_id,
 239                             fatal=False))
 240                     else:
 241                         bitrate = link.get('Bitrate')
 242                         if bitrate:
 243                             format_id += '-%s' % bitrate
 244                         formats.append({
 245                             'url': uri,
 246                             'format_id': format_id,
 247                             'tbr': int_or_none(bitrate),
 248                             'ext': link.get('FileFormat'),
 249                             'vcodec': 'none' if kind == 'AudioResource' else None,
 250                             'quality': preference,
 251                         })
 252             subtitles_list = asset.get('SubtitlesList') or asset.get('Subtitleslist')
 253             if isinstance(subtitles_list, list):
 254                 LANGS = {
 255                     'Danish': 'da',
 256                 }
 257                 for subs in subtitles_list:
 258                     if not isinstance(subs, dict):
 259                         continue
 260                     sub_uri = url_or_none(subs.get('Uri'))
 261                     if not sub_uri:
 262                         continue
 263                     lang = subs.get('Language') or 'da'
 264                     subtitles.setdefault(LANGS.get(lang, lang), []).append({
 265                         'url': sub_uri,
 266                         'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
 267                     })
 268
 269         if not formats and restricted_to_denmark:
 270             self.raise_geo_restricted(
 271                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
 272                 countries=self._GEO_COUNTRIES)
 273
 274         self._sort_formats(formats)
 275
 276         return {
 277             'id': video_id,
 278             'title': title,
 279             'description': description,
 280             'thumbnail': thumbnail,
 281             'timestamp': timestamp,
 282             'duration': duration,
 283             'formats': formats,
 284             'subtitles': subtitles,
 285             'series': str_or_none(data.get('SeriesTitle')),
 286             'season': str_or_none(data.get('SeasonTitle')),
 287             'season_number': int_or_none(data.get('SeasonNumber')),
 288             'season_id': str_or_none(data.get('SeasonUrn')),
 289             'episode': str_or_none(data.get('EpisodeTitle')),
 290             'episode_number': int_or_none(data.get('EpisodeNumber')),
 291             'release_year': int_or_none(data.get('ProductionYear')),
 292         }
 293
 294
 295 class DRTVLiveIE(InfoExtractor):
 296     IE_NAME = 'drtv:live'
 297     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
 298     _GEO_COUNTRIES = ['DK']
 299     _TEST = {
 300         'url': 'https://www.dr.dk/tv/live/dr1',
 301         'info_dict': {
 302             'id': 'dr1',
 303             'ext': 'mp4',
 304             'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 305         },
 306         'params': {
 307             # m3u8 download
 308             'skip_download': True,
 309         },
 310     }
 311
 312     def _real_extract(self, url):
 313         channel_id = self._match_id(url)
 314         channel_data = self._download_json(
 315             'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
 316             channel_id)
 317         title = channel_data['Title']
 318
 319         formats = []
 320         for streaming_server in channel_data.get('StreamingServers', []):
 321             server = streaming_server.get('Server')
 322             if not server:
 323                 continue
 324             link_type = streaming_server.get('LinkType')
 325             for quality in streaming_server.get('Qualities', []):
 326                 for stream in quality.get('Streams', []):
 327                     stream_path = stream.get('Stream')
 328                     if not stream_path:
 329                         continue
 330                     stream_url = update_url_query(
 331                         '%s/%s' % (server, stream_path), {'b': ''})
 332                     if link_type == 'HLS':
 333                         formats.extend(self._extract_m3u8_formats(
 334                             stream_url, channel_id, 'mp4',
 335                             m3u8_id=link_type, fatal=False, live=True))
 336                     elif link_type == 'HDS':
 337                         formats.extend(self._extract_f4m_formats(update_url_query(
 338                             '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
 339                             channel_id, f4m_id=link_type, fatal=False))
 340         self._sort_formats(formats)
 341
 342         return {
 343             'id': channel_id,
 344             'title': title,
 345             'thumbnail': channel_data.get('PrimaryImageUri'),
 346             'formats': formats,
 347             'is_live': True,
 348         }