youtube_dl/extractor/drtv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import binascii
   5 import hashlib
   6 import re
   7
   8
   9 from .common import InfoExtractor
  10 from ..aes import aes_cbc_decrypt
  11 from ..compat import compat_urllib_parse_unquote
  12 from ..utils import (
  13     bytes_to_intlist,
  14     ExtractorError,
  15     int_or_none,
  16     intlist_to_bytes,
  17     float_or_none,
  18     mimetype2ext,
  19     str_or_none,
  20     try_get,
  21     unified_timestamp,
  22     update_url_query,
  23     url_or_none,
  24 )
  25
  26
  27 class DRTVIE(InfoExtractor):
  28     _VALID_URL = r'''(?x)
  29                     https?://
  30                         (?:
  31                             (?:www\.)?dr\.dk/(?:tv/se|nyheder|radio(?:/ondemand)?)/(?:[^/]+/)*|
  32                             (?:www\.)?(?:dr\.dk|dr-massive\.com)/drtv/(?:se|episode|program)/
  33                         )
  34                         (?P<id>[\da-z_-]+)
  35                     '''
  36     _GEO_BYPASS = False
  37     _GEO_COUNTRIES = ['DK']
  38     IE_NAME = 'drtv'
  39     _TESTS = [{
  40         'url': 'https://www.dr.dk/tv/se/boern/ultra/klassen-ultra/klassen-darlig-taber-10',
  41         'md5': '25e659cccc9a2ed956110a299fdf5983',
  42         'info_dict': {
  43             'id': 'klassen-darlig-taber-10',
  44             'ext': 'mp4',
  45             'title': 'Klassen - Dårlig taber (10)',
  46             'description': 'md5:815fe1b7fa656ed80580f31e8b3c79aa',
  47             'timestamp': 1539085800,
  48             'upload_date': '20181009',
  49             'duration': 606.84,
  50             'series': 'Klassen',
  51             'season': 'Klassen I',
  52             'season_number': 1,
  53             'season_id': 'urn:dr:mu:bundle:57d7e8216187a4031cfd6f6b',
  54             'episode': 'Episode 10',
  55             'episode_number': 10,
  56             'release_year': 2016,
  57         },
  58         'expected_warnings': ['Unable to download f4m manifest'],
  59     }, {
  60         # embed
  61         'url': 'https://www.dr.dk/nyheder/indland/live-christianias-rydning-af-pusher-street-er-i-gang',
  62         'info_dict': {
  63             'id': 'urn:dr:mu:programcard:57c926176187a50a9c6e83c6',
  64             'ext': 'mp4',
  65             'title': 'christiania pusher street ryddes drdkrjpo',
  66             'description': 'md5:2a71898b15057e9b97334f61d04e6eb5',
  67             'timestamp': 1472800279,
  68             'upload_date': '20160902',
  69             'duration': 131.4,
  70         },
  71         'params': {
  72             'skip_download': True,
  73         },
  74         'expected_warnings': ['Unable to download f4m manifest'],
  75     }, {
  76         # with SignLanguage formats
  77         'url': 'https://www.dr.dk/tv/se/historien-om-danmark/-/historien-om-danmark-stenalder',
  78         'info_dict': {
  79             'id': 'historien-om-danmark-stenalder',
  80             'ext': 'mp4',
  81             'title': 'Historien om Danmark: Stenalder',
  82             'description': 'md5:8c66dcbc1669bbc6f873879880f37f2a',
  83             'timestamp': 1546628400,
  84             'upload_date': '20190104',
  85             'duration': 3502.56,
  86             'formats': 'mincount:20',
  87         },
  88         'params': {
  89             'skip_download': True,
  90         },
  91     }, {
  92         'url': 'https://www.dr.dk/radio/p4kbh/regionale-nyheder-kh4/p4-nyheder-2019-06-26-17-30-9',
  93         'only_matching': True,
  94     }, {
  95         'url': 'https://www.dr.dk/drtv/se/bonderoeven_71769',
  96         'info_dict': {
  97             'id': '00951930010',
  98             'ext': 'mp4',
  99             'title': 'Bonderøven (1:8)',
 100             'description': 'md5:3cf18fc0d3b205745d4505f896af8121',
 101             'timestamp': 1546542000,
 102             'upload_date': '20190103',
 103             'duration': 2576.6,
 104         },
 105         'params': {
 106             'skip_download': True,
 107         },
 108     }, {
 109         'url': 'https://www.dr.dk/drtv/episode/bonderoeven_71769',
 110         'only_matching': True,
 111     }, {
 112         'url': 'https://dr-massive.com/drtv/se/bonderoeven_71769',
 113         'only_matching': True,
 114     }, {
 115         'url': 'https://www.dr.dk/drtv/program/jagten_220924',
 116         'only_matching': True,
 117     }]
 118
 119     def _real_extract(self, url):
 120         video_id = self._match_id(url)
 121
 122         webpage = self._download_webpage(url, video_id)
 123
 124         if '>Programmet er ikke længere tilgængeligt' in webpage:
 125             raise ExtractorError(
 126                 'Video %s is not available' % video_id, expected=True)
 127
 128         video_id = self._search_regex(
 129             (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
 130              r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
 131             webpage, 'video id', default=None)
 132
 133         if not video_id:
 134             video_id = self._search_regex(
 135                 r'(urn(?:%3A|:)dr(?:%3A|:)mu(?:%3A|:)programcard(?:%3A|:)[\da-f]+)',
 136                 webpage, 'urn', default=None)
 137             if video_id:
 138                 video_id = compat_urllib_parse_unquote(video_id)
 139
 140         _PROGRAMCARD_BASE = 'https://www.dr.dk/mu-online/api/1.4/programcard'
 141         query = {'expanded': 'true'}
 142
 143         if video_id:
 144             programcard_url = '%s/%s' % (_PROGRAMCARD_BASE, video_id)
 145         else:
 146             programcard_url = _PROGRAMCARD_BASE
 147             page = self._parse_json(
 148                 self._search_regex(
 149                     r'data\s*=\s*({.+?})\s*(?:;|</script)', webpage,
 150                     'data'), '1')['cache']['page']
 151             page = page[list(page.keys())[0]]
 152             item = try_get(
 153                 page, (lambda x: x['item'], lambda x: x['entries'][0]['item']),
 154                 dict)
 155             video_id = item['customId'].split(':')[-1]
 156             query['productionnumber'] = video_id
 157
 158         data = self._download_json(
 159             programcard_url, video_id, 'Downloading video JSON', query=query)
 160
 161         title = str_or_none(data.get('Title')) or re.sub(
 162             r'\s*\|\s*(?:TV\s*\|\s*DR|DRTV)$', '',
 163             self._og_search_title(webpage))
 164         description = self._og_search_description(
 165             webpage, default=None) or data.get('Description')
 166
 167         timestamp = unified_timestamp(
 168             data.get('PrimaryBroadcastStartTime') or data.get('SortDateTime'))
 169
 170         thumbnail = None
 171         duration = None
 172
 173         restricted_to_denmark = False
 174
 175         formats = []
 176         subtitles = {}
 177
 178         assets = []
 179         primary_asset = data.get('PrimaryAsset')
 180         if isinstance(primary_asset, dict):
 181             assets.append(primary_asset)
 182         secondary_assets = data.get('SecondaryAssets')
 183         if isinstance(secondary_assets, list):
 184             for secondary_asset in secondary_assets:
 185                 if isinstance(secondary_asset, dict):
 186                     assets.append(secondary_asset)
 187
 188         def hex_to_bytes(hex):
 189             return binascii.a2b_hex(hex.encode('ascii'))
 190
 191         def decrypt_uri(e):
 192             n = int(e[2:10], 16)
 193             a = e[10 + n:]
 194             data = bytes_to_intlist(hex_to_bytes(e[10:10 + n]))
 195             key = bytes_to_intlist(hashlib.sha256(
 196                 ('%s:sRBzYNXBzkKgnjj8pGtkACch' % a).encode('utf-8')).digest())
 197             iv = bytes_to_intlist(hex_to_bytes(a))
 198             decrypted = aes_cbc_decrypt(data, key, iv)
 199             return intlist_to_bytes(
 200                 decrypted[:-decrypted[-1]]).decode('utf-8').split('?')[0]
 201
 202         for asset in assets:
 203             kind = asset.get('Kind')
 204             if kind == 'Image':
 205                 thumbnail = url_or_none(asset.get('Uri'))
 206             elif kind in ('VideoResource', 'AudioResource'):
 207                 duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
 208                 restricted_to_denmark = asset.get('RestrictedToDenmark')
 209                 asset_target = asset.get('Target')
 210                 for link in asset.get('Links', []):
 211                     uri = link.get('Uri')
 212                     if not uri:
 213                         encrypted_uri = link.get('EncryptedUri')
 214                         if not encrypted_uri:
 215                             continue
 216                         try:
 217                             uri = decrypt_uri(encrypted_uri)
 218                         except Exception:
 219                             self.report_warning(
 220                                 'Unable to decrypt EncryptedUri', video_id)
 221                             continue
 222                     uri = url_or_none(uri)
 223                     if not uri:
 224                         continue
 225                     target = link.get('Target')
 226                     format_id = target or ''
 227                     if asset_target in ('SpokenSubtitles', 'SignLanguage', 'VisuallyInterpreted'):
 228                         preference = -1
 229                         format_id += '-%s' % asset_target
 230                     elif asset_target == 'Default':
 231                         preference = 1
 232                     else:
 233                         preference = None
 234                     if target == 'HDS':
 235                         f4m_formats = self._extract_f4m_formats(
 236                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
 237                             video_id, preference, f4m_id=format_id, fatal=False)
 238                         if kind == 'AudioResource':
 239                             for f in f4m_formats:
 240                                 f['vcodec'] = 'none'
 241                         formats.extend(f4m_formats)
 242                     elif target == 'HLS':
 243                         formats.extend(self._extract_m3u8_formats(
 244                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
 245                             preference=preference, m3u8_id=format_id,
 246                             fatal=False))
 247                     else:
 248                         bitrate = link.get('Bitrate')
 249                         if bitrate:
 250                             format_id += '-%s' % bitrate
 251                         formats.append({
 252                             'url': uri,
 253                             'format_id': format_id,
 254                             'tbr': int_or_none(bitrate),
 255                             'ext': link.get('FileFormat'),
 256                             'vcodec': 'none' if kind == 'AudioResource' else None,
 257                             'preference': preference,
 258                         })
 259             subtitles_list = asset.get('SubtitlesList') or asset.get('Subtitleslist')
 260             if isinstance(subtitles_list, list):
 261                 LANGS = {
 262                     'Danish': 'da',
 263                 }
 264                 for subs in subtitles_list:
 265                     if not isinstance(subs, dict):
 266                         continue
 267                     sub_uri = url_or_none(subs.get('Uri'))
 268                     if not sub_uri:
 269                         continue
 270                     lang = subs.get('Language') or 'da'
 271                     subtitles.setdefault(LANGS.get(lang, lang), []).append({
 272                         'url': sub_uri,
 273                         'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
 274                     })
 275
 276         if not formats and restricted_to_denmark:
 277             self.raise_geo_restricted(
 278                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
 279                 countries=self._GEO_COUNTRIES)
 280
 281         self._sort_formats(formats)
 282
 283         return {
 284             'id': video_id,
 285             'title': title,
 286             'description': description,
 287             'thumbnail': thumbnail,
 288             'timestamp': timestamp,
 289             'duration': duration,
 290             'formats': formats,
 291             'subtitles': subtitles,
 292             'series': str_or_none(data.get('SeriesTitle')),
 293             'season': str_or_none(data.get('SeasonTitle')),
 294             'season_number': int_or_none(data.get('SeasonNumber')),
 295             'season_id': str_or_none(data.get('SeasonUrn')),
 296             'episode': str_or_none(data.get('EpisodeTitle')),
 297             'episode_number': int_or_none(data.get('EpisodeNumber')),
 298             'release_year': int_or_none(data.get('ProductionYear')),
 299         }
 300
 301
 302 class DRTVLiveIE(InfoExtractor):
 303     IE_NAME = 'drtv:live'
 304     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
 305     _GEO_COUNTRIES = ['DK']
 306     _TEST = {
 307         'url': 'https://www.dr.dk/tv/live/dr1',
 308         'info_dict': {
 309             'id': 'dr1',
 310             'ext': 'mp4',
 311             'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 312         },
 313         'params': {
 314             # m3u8 download
 315             'skip_download': True,
 316         },
 317     }
 318
 319     def _real_extract(self, url):
 320         channel_id = self._match_id(url)
 321         channel_data = self._download_json(
 322             'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
 323             channel_id)
 324         title = self._live_title(channel_data['Title'])
 325
 326         formats = []
 327         for streaming_server in channel_data.get('StreamingServers', []):
 328             server = streaming_server.get('Server')
 329             if not server:
 330                 continue
 331             link_type = streaming_server.get('LinkType')
 332             for quality in streaming_server.get('Qualities', []):
 333                 for stream in quality.get('Streams', []):
 334                     stream_path = stream.get('Stream')
 335                     if not stream_path:
 336                         continue
 337                     stream_url = update_url_query(
 338                         '%s/%s' % (server, stream_path), {'b': ''})
 339                     if link_type == 'HLS':
 340                         formats.extend(self._extract_m3u8_formats(
 341                             stream_url, channel_id, 'mp4',
 342                             m3u8_id=link_type, fatal=False, live=True))
 343                     elif link_type == 'HDS':
 344                         formats.extend(self._extract_f4m_formats(update_url_query(
 345                             '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
 346                             channel_id, f4m_id=link_type, fatal=False))
 347         self._sort_formats(formats)
 348
 349         return {
 350             'id': channel_id,
 351             'title': title,
 352             'thumbnail': channel_data.get('PrimaryImageUri'),
 353             'formats': formats,
 354             'is_live': True,
 355         }