_ Git - youtube-dl/blob - youtube_dl/extractor/nexx.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_str
   8 from ..utils import (
   9     int_or_none,
  10     parse_duration,
  11     try_get,
  12 )
  13
  14
  15 class NexxIE(InfoExtractor):
  16     _VALID_URL = r'''(?x)
  17                         (?:
  18                             https?://api\.nexx(?:\.cloud|cdn\.com)/v3/(?P<domain_id>\d+)/videos/byid/|
  19                             nexx:(?P<domain_id_s>\d+):
  20                         )
  21                         (?P<id>\d+)
  22                     '''
  23     _TESTS = [{
  24         # movie
  25         'url': 'https://api.nexx.cloud/v3/748/videos/byid/128907',
  26         'md5': '828cea195be04e66057b846288295ba1',
  27         'info_dict': {
  28             'id': '128907',
  29             'ext': 'mp4',
  30             'title': 'Stiftung Warentest',
  31             'alt_title': 'Wie ein Test abläuft',
  32             'description': 'md5:d1ddb1ef63de721132abd38639cc2fd2',
  33             'release_year': 2013,
  34             'creator': 'SPIEGEL TV',
  35             'thumbnail': r're:^https?://.*\.jpg$',
  36             'duration': 2509,
  37             'timestamp': 1384264416,
  38             'upload_date': '20131112',
  39         },
  40     }, {
  41         # episode
  42         'url': 'https://api.nexx.cloud/v3/741/videos/byid/247858',
  43         'info_dict': {
  44             'id': '247858',
  45             'ext': 'mp4',
  46             'title': 'Return of the Golden Child (OV)',
  47             'description': 'md5:5d969537509a92b733de21bae249dc63',
  48             'release_year': 2017,
  49             'thumbnail': r're:^https?://.*\.jpg$',
  50             'duration': 1397,
  51             'timestamp': 1495033267,
  52             'upload_date': '20170517',
  53             'episode_number': 2,
  54             'season_number': 2,
  55         },
  56         'params': {
  57             'skip_download': True,
  58         },
  59     }, {
  60         'url': 'https://api.nexxcdn.com/v3/748/videos/byid/128907',
  61         'only_matching': True,
  62     }, {
  63         'url': 'nexx:748:128907',
  64         'only_matching': True,
  65     }]
  66
  67     @staticmethod
  68     def _extract_domain_id(webpage):
  69         mobj = re.search(
  70             r'<script\b[^>]+\bsrc=["\'](?:https?:)?//require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
  71             webpage)
  72         return mobj.group('id') if mobj else None
  73
  74     @staticmethod
  75     def _extract_urls(webpage):
  76         # Reference:
  77         # 1. https://nx-s.akamaized.net/files/201510/44.pdf
  78
  79         entries = []
  80
  81         # JavaScript Integration
  82         domain_id = NexxIE._extract_domain_id(webpage)
  83         if domain_id:
  84             for video_id in re.findall(
  85                     r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
  86                     webpage):
  87                 entries.append(
  88                     'https://api.nexx.cloud/v3/%s/videos/byid/%s'
  89                     % (domain_id, video_id))
  90
  91         # TODO: support more embed formats
  92
  93         return entries
  94
  95     @staticmethod
  96     def _extract_url(webpage):
  97         return NexxIE._extract_urls(webpage)[0]
  98
  99     def _real_extract(self, url):
 100         video_id = self._match_id(url)
 101
 102         video = self._download_json(
 103             'https://arc.nexx.cloud/api/video/%s.json' % video_id,
 104             video_id)['result']
 105
 106         general = video['general']
 107         title = general['title']
 108
 109         stream_data = video['streamdata']
 110         language = general.get('language_raw') or ''
 111
 112         # TODO: reverse more cdns
 113
 114         cdn = stream_data['cdnType']
 115         assert cdn == 'azure'
 116
 117         azure_locator = stream_data['azureLocator']
 118
 119         AZURE_URL = 'http://nx%s%02d.akamaized.net/'
 120
 121         def get_cdn_shield_base(shield_type='', prefix='-p'):
 122             for secure in ('', 's'):
 123                 cdn_shield = stream_data.get('cdnShield%sHTTP%s' % (shield_type, secure.upper()))
 124                 if cdn_shield:
 125                     return 'http%s://%s' % (secure, cdn_shield)
 126             else:
 127                 return AZURE_URL % (prefix, int(stream_data['azureAccount'].replace('nexxplayplus', '')))
 128
 129         azure_stream_base = get_cdn_shield_base()
 130         is_ml = ',' in language
 131         azure_manifest_url = '%s%s/%s_src%s.ism/Manifest' % (
 132             azure_stream_base, azure_locator, video_id, ('_manifest' if is_ml else '')) + '%s'
 133
 134         protection_token = try_get(
 135             video, lambda x: x['protectiondata']['token'], compat_str)
 136         if protection_token:
 137             azure_manifest_url += '?hdnts=%s' % protection_token
 138
 139         formats = self._extract_m3u8_formats(
 140             azure_manifest_url % '(format=m3u8-aapl)',
 141             video_id, 'mp4', 'm3u8_native',
 142             m3u8_id='%s-hls' % cdn, fatal=False)
 143         formats.extend(self._extract_mpd_formats(
 144             azure_manifest_url % '(format=mpd-time-csf)',
 145             video_id, mpd_id='%s-dash' % cdn, fatal=False))
 146         formats.extend(self._extract_ism_formats(
 147             azure_manifest_url % '', video_id, ism_id='%s-mss' % cdn, fatal=False))
 148
 149         azure_progressive_base = get_cdn_shield_base('Prog', '-d')
 150         azure_file_distribution = stream_data.get('azureFileDistribution')
 151         if azure_file_distribution:
 152             fds = azure_file_distribution.split(',')
 153             if fds:
 154                 for fd in fds:
 155                     ss = fd.split(':')
 156                     if len(ss) == 2:
 157                         tbr = int_or_none(ss[0])
 158                         if tbr:
 159                             f = {
 160                                 'url': '%s%s/%s_src_%s_%d.mp4' % (
 161                                     azure_progressive_base, azure_locator, video_id, ss[1], tbr),
 162                                 'format_id': '%s-http-%d' % (cdn, tbr),
 163                                 'tbr': tbr,
 164                             }
 165                             width_height = ss[1].split('x')
 166                             if len(width_height) == 2:
 167                                 f.update({
 168                                     'width': int_or_none(width_height[0]),
 169                                     'height': int_or_none(width_height[1]),
 170                                 })
 171                             formats.append(f)
 172
 173         self._sort_formats(formats)
 174
 175         return {
 176             'id': video_id,
 177             'title': title,
 178             'alt_title': general.get('subtitle'),
 179             'description': general.get('description'),
 180             'release_year': int_or_none(general.get('year')),
 181             'creator': general.get('studio') or general.get('studio_adref'),
 182             'thumbnail': try_get(
 183                 video, lambda x: x['imagedata']['thumb'], compat_str),
 184             'duration': parse_duration(general.get('runtime')),
 185             'timestamp': int_or_none(general.get('uploaded')),
 186             'episode_number': int_or_none(try_get(
 187                 video, lambda x: x['episodedata']['episode'])),
 188             'season_number': int_or_none(try_get(
 189                 video, lambda x: x['episodedata']['season'])),
 190             'formats': formats,
 191         }
 192
 193
 194 class NexxEmbedIE(InfoExtractor):
 195     _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
 196     _TEST = {
 197         'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
 198         'md5': '16746bfc28c42049492385c989b26c4a',
 199         'info_dict': {
 200             'id': '161464',
 201             'ext': 'mp4',
 202             'title': 'Nervenkitzel Achterbahn',
 203             'alt_title': 'Karussellbauer in Deutschland',
 204             'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
 205             'release_year': 2005,
 206             'creator': 'SPIEGEL TV',
 207             'thumbnail': r're:^https?://.*\.jpg$',
 208             'duration': 2761,
 209             'timestamp': 1394021479,
 210             'upload_date': '20140305',
 211         },
 212         'params': {
 213             'format': 'bestvideo',
 214             'skip_download': True,
 215         },
 216     }
 217
 218     @staticmethod
 219     def _extract_urls(webpage):
 220         # Reference:
 221         # 1. https://nx-s.akamaized.net/files/201510/44.pdf
 222
 223         # iFrame Embed Integration
 224         return [mobj.group('url') for mobj in re.finditer(
 225             r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
 226             webpage)]
 227
 228     def _real_extract(self, url):
 229         embed_id = self._match_id(url)
 230
 231         webpage = self._download_webpage(url, embed_id)
 232
 233         return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())