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
  85     def _real_extract(self, url):
  86         video_id = self._match_id(url)
  87
  88         webpage = self._download_webpage(url, video_id)
  89
  90         if '>Programmet er ikke længere tilgængeligt' in webpage:
  91             raise ExtractorError(
  92                 'Video %s is not available' % video_id, expected=True)
  93
  94         video_id = self._search_regex(
  95             (r'data-(?:material-identifier|episode-slug)="([^"]+)"',
  96              r'data-resource="[^>"]+mu/programcard/expanded/([^"]+)"'),
  97             webpage, 'video id', default=None)
  98
  99         if not video_id:
 100             video_id = compat_urllib_parse_unquote(self._search_regex(
 101                 r'(urn(?:%3A|:)dr(?:%3A|:)mu(?:%3A|:)programcard(?:%3A|:)[\da-f]+)',
 102                 webpage, 'urn'))
 103
 104         data = self._download_json(
 105             'https://www.dr.dk/mu-online/api/1.4/programcard/%s' % video_id,
 106             video_id, 'Downloading video JSON', query={'expanded': 'true'})
 107
 108         title = str_or_none(data.get('Title')) or re.sub(
 109             r'\s*\|\s*(?:TV\s*\|\s*DR|DRTV)$', '',
 110             self._og_search_title(webpage))
 111         description = self._og_search_description(
 112             webpage, default=None) or data.get('Description')
 113
 114         timestamp = unified_timestamp(
 115             data.get('PrimaryBroadcastStartTime') or data.get('SortDateTime'))
 116
 117         thumbnail = None
 118         duration = None
 119
 120         restricted_to_denmark = False
 121
 122         formats = []
 123         subtitles = {}
 124
 125         assets = []
 126         primary_asset = data.get('PrimaryAsset')
 127         if isinstance(primary_asset, dict):
 128             assets.append(primary_asset)
 129         secondary_assets = data.get('SecondaryAssets')
 130         if isinstance(secondary_assets, list):
 131             for secondary_asset in secondary_assets:
 132                 if isinstance(secondary_asset, dict):
 133                     assets.append(secondary_asset)
 134
 135         def hex_to_bytes(hex):
 136             return binascii.a2b_hex(hex.encode('ascii'))
 137
 138         def decrypt_uri(e):
 139             n = int(e[2:10], 16)
 140             a = e[10 + n:]
 141             data = bytes_to_intlist(hex_to_bytes(e[10:10 + n]))
 142             key = bytes_to_intlist(hashlib.sha256(
 143                 ('%s:sRBzYNXBzkKgnjj8pGtkACch' % a).encode('utf-8')).digest())
 144             iv = bytes_to_intlist(hex_to_bytes(a))
 145             decrypted = aes_cbc_decrypt(data, key, iv)
 146             return intlist_to_bytes(
 147                 decrypted[:-decrypted[-1]]).decode('utf-8').split('?')[0]
 148
 149         for asset in assets:
 150             kind = asset.get('Kind')
 151             if kind == 'Image':
 152                 thumbnail = url_or_none(asset.get('Uri'))
 153             elif kind in ('VideoResource', 'AudioResource'):
 154                 duration = float_or_none(asset.get('DurationInMilliseconds'), 1000)
 155                 restricted_to_denmark = asset.get('RestrictedToDenmark')
 156                 asset_target = asset.get('Target')
 157                 for link in asset.get('Links', []):
 158                     uri = link.get('Uri')
 159                     if not uri:
 160                         encrypted_uri = link.get('EncryptedUri')
 161                         if not encrypted_uri:
 162                             continue
 163                         try:
 164                             uri = decrypt_uri(encrypted_uri)
 165                         except Exception:
 166                             self.report_warning(
 167                                 'Unable to decrypt EncryptedUri', video_id)
 168                             continue
 169                     uri = url_or_none(uri)
 170                     if not uri:
 171                         continue
 172                     target = link.get('Target')
 173                     format_id = target or ''
 174                     if asset_target in ('SpokenSubtitles', 'SignLanguage', 'VisuallyInterpreted'):
 175                         preference = -1
 176                         format_id += '-%s' % asset_target
 177                     elif asset_target == 'Default':
 178                         preference = 1
 179                     else:
 180                         preference = None
 181                     if target == 'HDS':
 182                         f4m_formats = self._extract_f4m_formats(
 183                             uri + '?hdcore=3.3.0&plugin=aasp-3.3.0.99.43',
 184                             video_id, preference, f4m_id=format_id, fatal=False)
 185                         if kind == 'AudioResource':
 186                             for f in f4m_formats:
 187                                 f['vcodec'] = 'none'
 188                         formats.extend(f4m_formats)
 189                     elif target == 'HLS':
 190                         formats.extend(self._extract_m3u8_formats(
 191                             uri, video_id, 'mp4', entry_protocol='m3u8_native',
 192                             preference=preference, m3u8_id=format_id,
 193                             fatal=False))
 194                     else:
 195                         bitrate = link.get('Bitrate')
 196                         if bitrate:
 197                             format_id += '-%s' % bitrate
 198                         formats.append({
 199                             'url': uri,
 200                             'format_id': format_id,
 201                             'tbr': int_or_none(bitrate),
 202                             'ext': link.get('FileFormat'),
 203                             'vcodec': 'none' if kind == 'AudioResource' else None,
 204                             'preference': preference,
 205                         })
 206             subtitles_list = asset.get('SubtitlesList') or asset.get('Subtitleslist')
 207             if isinstance(subtitles_list, list):
 208                 LANGS = {
 209                     'Danish': 'da',
 210                 }
 211                 for subs in subtitles_list:
 212                     if not isinstance(subs, dict):
 213                         continue
 214                     sub_uri = url_or_none(subs.get('Uri'))
 215                     if not sub_uri:
 216                         continue
 217                     lang = subs.get('Language') or 'da'
 218                     subtitles.setdefault(LANGS.get(lang, lang), []).append({
 219                         'url': sub_uri,
 220                         'ext': mimetype2ext(subs.get('MimeType')) or 'vtt'
 221                     })
 222
 223         if not formats and restricted_to_denmark:
 224             self.raise_geo_restricted(
 225                 'Unfortunately, DR is not allowed to show this program outside Denmark.',
 226                 countries=self._GEO_COUNTRIES)
 227
 228         self._sort_formats(formats)
 229
 230         return {
 231             'id': video_id,
 232             'title': title,
 233             'description': description,
 234             'thumbnail': thumbnail,
 235             'timestamp': timestamp,
 236             'duration': duration,
 237             'formats': formats,
 238             'subtitles': subtitles,
 239             'series': str_or_none(data.get('SeriesTitle')),
 240             'season': str_or_none(data.get('SeasonTitle')),
 241             'season_number': int_or_none(data.get('SeasonNumber')),
 242             'season_id': str_or_none(data.get('SeasonUrn')),
 243             'episode': str_or_none(data.get('EpisodeTitle')),
 244             'episode_number': int_or_none(data.get('EpisodeNumber')),
 245             'release_year': int_or_none(data.get('ProductionYear')),
 246         }
 247
 248
 249 class DRTVLiveIE(InfoExtractor):
 250     IE_NAME = 'drtv:live'
 251     _VALID_URL = r'https?://(?:www\.)?dr\.dk/(?:tv|TV)/live/(?P<id>[\da-z-]+)'
 252     _GEO_COUNTRIES = ['DK']
 253     _TEST = {
 254         'url': 'https://www.dr.dk/tv/live/dr1',
 255         'info_dict': {
 256             'id': 'dr1',
 257             'ext': 'mp4',
 258             'title': 're:^DR1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
 259         },
 260         'params': {
 261             # m3u8 download
 262             'skip_download': True,
 263         },
 264     }
 265
 266     def _real_extract(self, url):
 267         channel_id = self._match_id(url)
 268         channel_data = self._download_json(
 269             'https://www.dr.dk/mu-online/api/1.0/channel/' + channel_id,
 270             channel_id)
 271         title = self._live_title(channel_data['Title'])
 272
 273         formats = []
 274         for streaming_server in channel_data.get('StreamingServers', []):
 275             server = streaming_server.get('Server')
 276             if not server:
 277                 continue
 278             link_type = streaming_server.get('LinkType')
 279             for quality in streaming_server.get('Qualities', []):
 280                 for stream in quality.get('Streams', []):
 281                     stream_path = stream.get('Stream')
 282                     if not stream_path:
 283                         continue
 284                     stream_url = update_url_query(
 285                         '%s/%s' % (server, stream_path), {'b': ''})
 286                     if link_type == 'HLS':
 287                         formats.extend(self._extract_m3u8_formats(
 288                             stream_url, channel_id, 'mp4',
 289                             m3u8_id=link_type, fatal=False, live=True))
 290                     elif link_type == 'HDS':
 291                         formats.extend(self._extract_f4m_formats(update_url_query(
 292                             '%s/%s' % (server, stream_path), {'hdcore': '3.7.0'}),
 293                             channel_id, f4m_id=link_type, fatal=False))
 294         self._sort_formats(formats)
 295
 296         return {
 297             'id': channel_id,
 298             'title': title,
 299             'thumbnail': channel_data.get('PrimaryImageUri'),
 300             'formats': formats,
 301             'is_live': True,
 302         }