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|lyd)(?:/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         'skip': 'this video has been removed',
  55     }, {
  56         # embed
  57         'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
  58         'info_dict': {
  59             'id': 'urn:dr:mu:programcard:57c926176187a50a9c6e83c6',
  60             'ext': 'mp4',
  61             'title': 'christiania pusher street ryddes drdkrjpo',
  62             'description': 'md5:2a71898b15057e9b97334f61d04e6eb5',
  63             'timestamp': 1472800279,
  64             'upload_date': '20160902',
  65             'duration': 131.4,
  66         },
  67         'params': {
  68             'skip_download': True,
  69         },
  70         'expected_warnings': ['Unable to download f4m manifest'],
  71     }, {
  72         # with SignLanguage formats
  73         'url': 'https://www.dr.dk/tv/se/historien-om-danmark/-/historien-om-danmark-stenalder',
  74         'info_dict': {
  75             'id': '00831690010',
  76             'ext': 'mp4',
  77             'title': 'Historien om Danmark: Stenalder',
  78             'description': 'md5:8c66dcbc1669bbc6f873879880f37f2a',
  79             'timestamp': 1546628400,
  80             'upload_date': '20190104',
  81             'duration': 3504.618,
  82             'formats': 'mincount:20',
  83             'release_year': 2017,
  84             'season_id': 'urn:dr:mu:bundle:5afc03ad6187a4065ca5fd35',
  85             'season_number': 1,
  86             'season': 'Historien om Danmark',
  87             'series': 'Historien om Danmark',
  88         },
  89         'params': {
  90             'skip_download': True,
  91         },
  92     }, {
  93         'url': 'https://www.dr.dk/lyd/p4kbh/regionale-nyheder-kh4/p4-nyheder-2019-06-26-17-30-9',
  94         'only_matching': True,
  95     }, {
  96         'url': 'https://www.dr.dk/drtv/se/bonderoeven_71769',
  97         'info_dict': {
  98             'id': '00951930010',
  99             'ext': 'mp4',
 100             'title': 'Bonderøven 2019 (1:8)',
 101             'description': 'md5:b6dcfe9b6f0bea6703e9a0092739a5bd',
 102             'timestamp': 1603188600,
 103             'upload_date': '20201020',
 104             'duration': 2576.6,
 105             'season': 'Bonderøven 2019',
 106             'season_id': 'urn:dr:mu:bundle:5c201667a11fa01ca4528ce5',
 107             'release_year': 2019,
 108             'season_number': 2019,
 109             'series': 'Frank & Kastaniegaarden'
 110         },
 111         'params': {
 112             'skip_download': True,
 113         },
 114     }, {
 115         'url': 'https://www.dr.dk/drtv/episode/bonderoeven_71769',
 116         'only_matching': True,
 117     }, {
 118         'url': 'https://dr-massive.com/drtv/se/bonderoeven_71769',
 119         'only_matching': True,
 120     }, {
 121         'url': 'https://www.dr.dk/drtv/program/jagten_220924',
 122         'only_matching': True,
 123     }, {
 124         'url': 'https://www.dr.dk/lyd/p4aarhus/regionale-nyheder-ar4/regionale-nyheder-2022-05-05-12-30-3',
 125         'info_dict': {
 126             'id': 'urn:dr:mu:programcard:6265cb2571401424d0360113',
 127             'title': "Regionale nyheder",
 128             'ext': 'mp4',
 129             'duration': 120.043,
 130             'series': 'P4 Østjylland regionale nyheder',
 131             'timestamp': 1651746600,
 132             'season': 'Regionale nyheder',
 133             'release_year': 0,
 134             'season_id': 'urn:dr:mu:bundle:61c26889539f0201586b73c5',
 135             'description': '',
 136             'upload_date': '20220505',
 137         },
 138         'params': {
 139             'skip_download': True,
 140         },
 141     }]
 142
 143     def _real_extract(self, url):
 144         video_id = self._match_id(url)
 145
 146         webpage = self._download_webpage(url, video_id)
 147
 148         if '>Programmet er ikke længere tilgængeligt' in webpage:
 149             raise ExtractorError(
 150                 'Video %s is not available' % video_id, expected=True)
 151
 152         video_id = self._search_regex(
 153             (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
 154              r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
 155             webpage, 'video id', default=None)
 156
 157         if not video_id:
 158             video_id = self._search_regex(
 159                 r'(urn(?:%3A|:)dr(?:%3A|:)mu(?:%3A|:)programcard(?:%3A|:)[\da-f]+)',
 160                 webpage, 'urn', default=None)
 161             if video_id:
 162                 video_id = compat_urllib_parse_unquote(video_id)
 163
 164         _PROGRAMCARD_BASE = 'https://www.dr.dk/mu-online/api/1.4/programcard'
 165         query = {'expanded': 'true'}
 166
 167         if video_id:
 168             programcard_url = '%s/%s' % (_PROGRAMCARD_BASE, video_id)
 169         else:
 170             programcard_url = _PROGRAMCARD_BASE
 171             page = self._parse_json(
 172                 self._search_regex(
 173                     r'data\s*=\s*({.+?})\s*(?:;|</script)', webpage,
 174                     'data'), '1')['cache']['page']
 175             page = page[list(page.keys())[0]]
 176             item = try_get(
 177                 page, (lambda x: x['item'], lambda x: x['entries'][0]['item']),
 178                 dict)
 179             video_id = item['customId'].split(':')[-1]
 180             query['productionnumber'] = video_id
 181
 182         data = self._download_json(
 183             programcard_url, video_id, 'Downloading video JSON', query=query)
 184
 185         title = str_or_none(data.get('Title')) or re.sub(
 186             r'\s*\|\s*(?:TV\s*\|\s*DR|DRTV)$', '',
 187             self._og_search_title(webpage))
 188         description = self._og_search_description(
 189             webpage, default=None) or data.get('Description')
 190
 191         timestamp = unified_timestamp(
 192             data.get('PrimaryBroadcastStartTime') or data.get('SortDateTime'))
 193
 194         thumbnail = None
 195         duration = None
 196
 197         restricted_to_denmark = False
 198
 199         formats = []
 200         subtitles = {}
 201
 202         assets = []
 203         primary_asset = data.get('PrimaryAsset')
 204         if isinstance(primary_asset, dict):
 205             assets.append(primary_asset)
 206         secondary_assets = data.get('SecondaryAssets')
 207         if isinstance(secondary_assets, list):
 208             for secondary_asset in secondary_assets:
 209                 if isinstance(secondary_asset, dict):
 210                     assets.append(secondary_asset)
 211
 212         def hex_to_bytes(hex):
 213             return binascii.a2b_hex(hex.encode('ascii'))
 214
 215         def decrypt_uri(e):
 216             n = int(e[2:10], 16)
 217             a = e[10 + n:]
 218             data = hex_to_bytes(e[10:10 + n])
 219             key = hashlib.sha256(('%s:sRBzYNXBzkKgnjj8pGtkACch' % a).encode('utf-8')).digest()
 220             iv = hex_to_bytes(a)
 221             decrypted = unpad_pkcs7(aes_cbc_decrypt_bytes(data, key, iv))
 222             return decrypted.decode('utf-8').split('?')[0]
 223
 224         for asset in assets:
 225             kind = asset.get('Kind')
 226             if kind == 'Image':
 227                 thumbnail = url_or_none(asset.get('Uri'))
 228             elif kind in ('VideoResource', 'AudioResource'):
 229                 duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
 230                 restricted_to_denmark = asset.get('RestrictedToDenmark')
 231                 asset_target = asset.get('Target')
 232                 for link in asset.get('Links', []):
 233                     uri = link.get('Uri')
 234                     if not uri:
 235                         encrypted_uri = link.get('EncryptedUri')
 236                         if not encrypted_uri:
 237                             continue
 238                         try:
 239                             uri = decrypt_uri(encrypted_uri)
 240                         except Exception:
 241                             self.report_warning(
 242                                 'Unable to decrypt EncryptedUri', video_id)
 243                             continue
 244                     uri = url_or_none(uri)
 245                     if not uri:
 246                         continue
 247                     target = link.get('Target')
 248                     format_id = target or ''
 249                     if asset_target in ('SpokenSubtitles', 'SignLanguage', 'VisuallyInterpreted'):
 250                         preference = -1
 251                         format_id += '-%s' % asset_target
 252                     elif asset_target == 'Default':
 253                         preference = 1
 254                     else:
 255                         preference = None
 256                     if target == 'HDS':
 257                         f4m_formats = self._extract_f4m_formats(
 258                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
 259                             video_id, preference, f4m_id=format_id, fatal=False)
 260                         if kind == 'AudioResource':
 261                             for f in f4m_formats:
 262                                 f['vcodec'] = 'none'
 263                         formats.extend(f4m_formats)
 264                     elif target == 'HLS':
 265                         formats.extend(self._extract_m3u8_formats(
 266                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
 267                             quality=preference, m3u8_id=format_id,
 268                             fatal=False))
 269                     else:
 270                         bitrate = link.get('Bitrate')
 271                         if bitrate:
 272                             format_id += '-%s' % bitrate
 273                         formats.append({
 274                             'url': uri,
 275                             'format_id': format_id,
 276                             'tbr': int_or_none(bitrate),
 277                             'ext': link.get('FileFormat'),
 278                             'vcodec': 'none' if kind == 'AudioResource' else None,
 279                             'quality': preference,
 280                         })
 281             subtitles_list = asset.get('SubtitlesList') or asset.get('Subtitleslist')
 282             if isinstance(subtitles_list, list):
 283                 LANGS = {
 284                     'Danish': 'da',
 285                 }
 286                 for subs in subtitles_list:
 287                     if not isinstance(subs, dict):
 288                         continue
 289                     sub_uri = url_or_none(subs.get('Uri'))
 290                     if not sub_uri:
 291                         continue
 292                     lang = subs.get('Language') or 'da'
 293                     subtitles.setdefault(LANGS.get(lang, lang), []).append({
 294                         'url': sub_uri,
 295                         'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
 296                     })
 297
 298         if not formats and restricted_to_denmark:
 299             self.raise_geo_restricted(
 300                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
 301                 countries=self._GEO_COUNTRIES)
 302
 303         self._sort_formats(formats)
 304
 305         return {
 306             'id': video_id,
 307             'title': title,
 308             'description': description,
 309             'thumbnail': thumbnail,
 310             'timestamp': timestamp,
 311             'duration': duration,
 312             'formats': formats,
 313             'subtitles': subtitles,
 314             'series': str_or_none(data.get('SeriesTitle')),
 315             'season': str_or_none(data.get('SeasonTitle')),
 316             'season_number': int_or_none(data.get('SeasonNumber')),
 317             'season_id': str_or_none(data.get('SeasonUrn')),
 318             'episode': str_or_none(data.get('EpisodeTitle')),
 319             'episode_number': int_or_none(data.get('EpisodeNumber')),
 320             'release_year': int_or_none(data.get('ProductionYear')),
 321         }
 322
 323
 324 class DRTVLiveIE(InfoExtractor):
 325     IE_NAME = 'drtv:live'
 326     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
 327     _GEO_COUNTRIES = ['DK']
 328     _TEST = {
 329         'url': 'https://www.dr.dk/tv/live/dr1',
 330         'info_dict': {
 331             'id': 'dr1',
 332             'ext': 'mp4',
 333             'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 334         },
 335         'params': {
 336             # m3u8 download
 337             'skip_download': True,
 338         },
 339     }
 340
 341     def _real_extract(self, url):
 342         channel_id = self._match_id(url)
 343         channel_data = self._download_json(
 344             'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
 345             channel_id)
 346         title = channel_data['Title']
 347
 348         formats = []
 349         for streaming_server in channel_data.get('StreamingServers', []):
 350             server = streaming_server.get('Server')
 351             if not server:
 352                 continue
 353             link_type = streaming_server.get('LinkType')
 354             for quality in streaming_server.get('Qualities', []):
 355                 for stream in quality.get('Streams', []):
 356                     stream_path = stream.get('Stream')
 357                     if not stream_path:
 358                         continue
 359                     stream_url = update_url_query(
 360                         '%s/%s' % (server, stream_path), {'b': ''})
 361                     if link_type == 'HLS':
 362                         formats.extend(self._extract_m3u8_formats(
 363                             stream_url, channel_id, 'mp4',
 364                             m3u8_id=link_type, fatal=False, live=True))
 365                     elif link_type == 'HDS':
 366                         formats.extend(self._extract_f4m_formats(update_url_query(
 367                             '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
 368                             channel_id, f4m_id=link_type, fatal=False))
 369         self._sort_formats(formats)
 370
 371         return {
 372             'id': channel_id,
 373             'title': title,
 374             'thumbnail': channel_data.get('PrimaryImageUri'),
 375             'formats': formats,
 376             'is_live': True,
 377         }