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