yt_dlp/extractor/newstube.py

   1 import base64
   2 import hashlib
   3
   4 from .common import InfoExtractor
   5 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
   6 from ..utils import (
   7     int_or_none,
   8     parse_codecs,
   9     parse_duration,
  10 )
  11
  12
  13 class NewstubeIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  15     _TEST = {
  16         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
  17         'md5': '9d10320ad473444352f72f746ccb8b8c',
  18         'info_dict': {
  19             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
  20             'ext': 'mp4',
  21             'title': 'Телеканал CNN переместил город Славянск в Крым',
  22             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
  23             'duration': 31.05,
  24         },
  25     }
  26
  27     def _real_extract(self, url):
  28         video_id = self._match_id(url)
  29
  30         page = self._download_webpage(url, video_id)
  31         title = self._html_search_meta(['og:title', 'twitter:title'], page, fatal=True)
  32
  33         video_guid = self._html_search_regex(
  34             r'<meta\s+property="og:video(?::(?:(?:secure_)?url|iframe))?"\s+content="https?://(?:www\.)?newstube\.ru/embed/(?P<guid>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
  35             page, 'video GUID')
  36
  37         enc_data = base64.b64decode(self._download_webpage(
  38             'https://www.newstube.ru/embed/api/player/getsources2',
  39             video_guid, query={
  40                 'guid': video_guid,
  41                 'ff': 3,
  42             }))
  43         key = hashlib.pbkdf2_hmac(
  44             'sha1', video_guid.replace('-', '').encode(), enc_data[:16], 1)[:16]
  45         dec_data = unpad_pkcs7(aes_cbc_decrypt_bytes(enc_data[32:], key, enc_data[16:32]))
  46         sources = self._parse_json(dec_data, video_guid)
  47
  48         formats = []
  49         for source in sources:
  50             source_url = source.get('Src')
  51             if not source_url:
  52                 continue
  53             height = int_or_none(source.get('Height'))
  54             f = {
  55                 'format_id': 'http' + ('-%dp' % height if height else ''),
  56                 'url': source_url,
  57                 'width': int_or_none(source.get('Width')),
  58                 'height': height,
  59             }
  60             source_type = source.get('Type')
  61             if source_type:
  62                 f.update(parse_codecs(self._search_regex(
  63                     r'codecs="([^"]+)"', source_type, 'codecs', fatal=False)))
  64             formats.append(f)
  65
  66         self._check_formats(formats, video_guid)
  67
  68         return {
  69             'id': video_guid,
  70             'title': title,
  71             'description': self._html_search_meta(['description', 'og:description'], page),
  72             'thumbnail': self._html_search_meta(['og:image:secure_url', 'og:image', 'twitter:image'], page),
  73             'duration': parse_duration(self._html_search_meta('duration', page)),
  74             'formats': formats,
  75         }