_ 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/\d+/videos/byid/|
  19                             nexx:(?:\d+:)?|
  20                             https?://arc\.nexx\.cloud/api/video/
  21                         )
  22                         (?P<id>\d+)
  23                     '''
  24     _TESTS = [{
  25         # movie
  26         'url': 'https://api.nexx.cloud/v3/748/videos/byid/128907',
  27         'md5': '828cea195be04e66057b846288295ba1',
  28         'info_dict': {
  29             'id': '128907',
  30             'ext': 'mp4',
  31             'title': 'Stiftung Warentest',
  32             'alt_title': 'Wie ein Test abläuft',
  33             'description': 'md5:d1ddb1ef63de721132abd38639cc2fd2',
  34             'release_year': 2013,
  35             'creator': 'SPIEGEL TV',
  36             'thumbnail': r're:^https?://.*\.jpg$',
  37             'duration': 2509,
  38             'timestamp': 1384264416,
  39             'upload_date': '20131112',
  40         },
  41     }, {
  42         # episode
  43         'url': 'https://api.nexx.cloud/v3/741/videos/byid/247858',
  44         'info_dict': {
  45             'id': '247858',
  46             'ext': 'mp4',
  47             'title': 'Return of the Golden Child (OV)',
  48             'description': 'md5:5d969537509a92b733de21bae249dc63',
  49             'release_year': 2017,
  50             'thumbnail': r're:^https?://.*\.jpg$',
  51             'duration': 1397,
  52             'timestamp': 1495033267,
  53             'upload_date': '20170517',
  54             'episode_number': 2,
  55             'season_number': 2,
  56         },
  57         'params': {
  58             'skip_download': True,
  59         },
  60     }, {
  61         'url': 'https://api.nexxcdn.com/v3/748/videos/byid/128907',
  62         'only_matching': True,
  63     }, {
  64         'url': 'nexx:748:128907',
  65         'only_matching': True,
  66     }, {
  67         'url': 'nexx:128907',
  68         'only_matching': True,
  69     }, {
  70         'url': 'https://arc.nexx.cloud/api/video/128907.json',
  71         'only_matching': True,
  72     }]
  73
  74     @staticmethod
  75     def _extract_domain_id(webpage):
  76         mobj = re.search(
  77             r'<script\b[^>]+\bsrc=["\'](?:https?:)?//require\.nexx(?:\.cloud|cdn\.com)/(?P<id>\d+)',
  78             webpage)
  79         return mobj.group('id') if mobj else None
  80
  81     @staticmethod
  82     def _extract_urls(webpage):
  83         # Reference:
  84         # 1. https://nx-s.akamaized.net/files/201510/44.pdf
  85
  86         entries = []
  87
  88         # JavaScript Integration
  89         domain_id = NexxIE._extract_domain_id(webpage)
  90         if domain_id:
  91             for video_id in re.findall(
  92                     r'(?is)onPLAYReady.+?_play\.init\s*\(.+?\s*,\s*["\']?(\d+)',
  93                     webpage):
  94                 entries.append(
  95                     'https://api.nexx.cloud/v3/%s/videos/byid/%s'
  96                     % (domain_id, video_id))
  97
  98         # TODO: support more embed formats
  99
 100         return entries
 101
 102     @staticmethod
 103     def _extract_url(webpage):
 104         return NexxIE._extract_urls(webpage)[0]
 105
 106     def _real_extract(self, url):
 107         video_id = self._match_id(url)
 108
 109         video = self._download_json(
 110             'https://arc.nexx.cloud/api/video/%s.json' % video_id,
 111             video_id)['result']
 112
 113         general = video['general']
 114         title = general['title']
 115
 116         stream_data = video['streamdata']
 117         language = general.get('language_raw') or ''
 118
 119         # TODO: reverse more cdns
 120
 121         cdn = stream_data['cdnType']
 122         assert cdn == 'azure'
 123
 124         azure_locator = stream_data['azureLocator']
 125
 126         AZURE_URL = 'http://nx%s%02d.akamaized.net/'
 127
 128         def get_cdn_shield_base(shield_type='', prefix='-p'):
 129             for secure in ('', 's'):
 130                 cdn_shield = stream_data.get('cdnShield%sHTTP%s' % (shield_type, secure.upper()))
 131                 if cdn_shield:
 132                     return 'http%s://%s' % (secure, cdn_shield)
 133             else:
 134                 return AZURE_URL % (prefix, int(stream_data['azureAccount'].replace('nexxplayplus', '')))
 135
 136         azure_stream_base = get_cdn_shield_base()
 137         is_ml = ',' in language
 138         azure_manifest_url = '%s%s/%s_src%s.ism/Manifest' % (
 139             azure_stream_base, azure_locator, video_id, ('_manifest' if is_ml else '')) + '%s'
 140
 141         protection_token = try_get(
 142             video, lambda x: x['protectiondata']['token'], compat_str)
 143         if protection_token:
 144             azure_manifest_url += '?hdnts=%s' % protection_token
 145
 146         formats = self._extract_m3u8_formats(
 147             azure_manifest_url % '(format=m3u8-aapl)',
 148             video_id, 'mp4', 'm3u8_native',
 149             m3u8_id='%s-hls' % cdn, fatal=False)
 150         formats.extend(self._extract_mpd_formats(
 151             azure_manifest_url % '(format=mpd-time-csf)',
 152             video_id, mpd_id='%s-dash' % cdn, fatal=False))
 153         formats.extend(self._extract_ism_formats(
 154             azure_manifest_url % '', video_id, ism_id='%s-mss' % cdn, fatal=False))
 155
 156         azure_progressive_base = get_cdn_shield_base('Prog', '-d')
 157         azure_file_distribution = stream_data.get('azureFileDistribution')
 158         if azure_file_distribution:
 159             fds = azure_file_distribution.split(',')
 160             if fds:
 161                 for fd in fds:
 162                     ss = fd.split(':')
 163                     if len(ss) == 2:
 164                         tbr = int_or_none(ss[0])
 165                         if tbr:
 166                             f = {
 167                                 'url': '%s%s/%s_src_%s_%d.mp4' % (
 168                                     azure_progressive_base, azure_locator, video_id, ss[1], tbr),
 169                                 'format_id': '%s-http-%d' % (cdn, tbr),
 170                                 'tbr': tbr,
 171                             }
 172                             width_height = ss[1].split('x')
 173                             if len(width_height) == 2:
 174                                 f.update({
 175                                     'width': int_or_none(width_height[0]),
 176                                     'height': int_or_none(width_height[1]),
 177                                 })
 178                             formats.append(f)
 179
 180         self._sort_formats(formats)
 181
 182         return {
 183             'id': video_id,
 184             'title': title,
 185             'alt_title': general.get('subtitle'),
 186             'description': general.get('description'),
 187             'release_year': int_or_none(general.get('year')),
 188             'creator': general.get('studio') or general.get('studio_adref'),
 189             'thumbnail': try_get(
 190                 video, lambda x: x['imagedata']['thumb'], compat_str),
 191             'duration': parse_duration(general.get('runtime')),
 192             'timestamp': int_or_none(general.get('uploaded')),
 193             'episode_number': int_or_none(try_get(
 194                 video, lambda x: x['episodedata']['episode'])),
 195             'season_number': int_or_none(try_get(
 196                 video, lambda x: x['episodedata']['season'])),
 197             'formats': formats,
 198         }
 199
 200
 201 class NexxEmbedIE(InfoExtractor):
 202     _VALID_URL = r'https?://embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?P<id>[^/?#&]+)'
 203     _TEST = {
 204         'url': 'http://embed.nexx.cloud/748/KC1614647Z27Y7T?autoplay=1',
 205         'md5': '16746bfc28c42049492385c989b26c4a',
 206         'info_dict': {
 207             'id': '161464',
 208             'ext': 'mp4',
 209             'title': 'Nervenkitzel Achterbahn',
 210             'alt_title': 'Karussellbauer in Deutschland',
 211             'description': 'md5:ffe7b1cc59a01f585e0569949aef73cc',
 212             'release_year': 2005,
 213             'creator': 'SPIEGEL TV',
 214             'thumbnail': r're:^https?://.*\.jpg$',
 215             'duration': 2761,
 216             'timestamp': 1394021479,
 217             'upload_date': '20140305',
 218         },
 219         'params': {
 220             'format': 'bestvideo',
 221             'skip_download': True,
 222         },
 223     }
 224
 225     @staticmethod
 226     def _extract_urls(webpage):
 227         # Reference:
 228         # 1. https://nx-s.akamaized.net/files/201510/44.pdf
 229
 230         # iFrame Embed Integration
 231         return [mobj.group('url') for mobj in re.finditer(
 232             r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.nexx(?:\.cloud|cdn\.com)/\d+/(?:(?!\1).)+)\1',
 233             webpage)]
 234
 235     def _real_extract(self, url):
 236         embed_id = self._match_id(url)
 237
 238         webpage = self._download_webpage(url, embed_id)
 239
 240         return self.url_result(NexxIE._extract_url(webpage), ie=NexxIE.ie_key())