yt_dlp/extractor/newstube.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import base64
   5 import hashlib
   6
   7 from .common import InfoExtractor
   8 from ..aes import aes_cbc_decrypt_bytes, unpad_pkcs7
   9 from ..utils import (
  10     int_or_none,
  11     parse_codecs,
  12     parse_duration,
  13 )
  14
  15
  16 class NewstubeIE(InfoExtractor):
  17     _VALID_URL = r'https?://(?:www\.)?newstube\.ru/media/(?P<id>.+)'
  18     _TEST = {
  19         'url': 'http://www.newstube.ru/media/telekanal-cnn-peremestil-gorod-slavyansk-v-krym',
  20         'md5': '9d10320ad473444352f72f746ccb8b8c',
  21         'info_dict': {
  22             'id': '728e0ef2-e187-4012-bac0-5a081fdcb1f6',
  23             'ext': 'mp4',
  24             'title': 'Телеканал CNN переместил город Славянск в Крым',
  25             'description': 'md5:419a8c9f03442bc0b0a794d689360335',
  26             'duration': 31.05,
  27         },
  28     }
  29
  30     def _real_extract(self, url):
  31         video_id = self._match_id(url)
  32
  33         page = self._download_webpage(url, video_id)
  34         title = self._html_search_meta(['og:title', 'twitter:title'], page, fatal=True)
  35
  36         video_guid = self._html_search_regex(
  37             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})',
  38             page, 'video GUID')
  39
  40         enc_data = base64.b64decode(self._download_webpage(
  41             'https://www.newstube.ru/embed/api/player/getsources2',
  42             video_guid, query={
  43                 'guid': video_guid,
  44                 'ff': 3,
  45             }))
  46         key = hashlib.pbkdf2_hmac(
  47             'sha1', video_guid.replace('-', '').encode(), enc_data[:16], 1)[:16]
  48         dec_data = unpad_pkcs7(aes_cbc_decrypt_bytes(enc_data[32:], key, enc_data[16:32]))
  49         sources = self._parse_json(dec_data, video_guid)
  50
  51         formats = []
  52         for source in sources:
  53             source_url = source.get('Src')
  54             if not source_url:
  55                 continue
  56             height = int_or_none(source.get('Height'))
  57             f = {
  58                 'format_id': 'http' + ('-%dp' % height if height else ''),
  59                 'url': source_url,
  60                 'width': int_or_none(source.get('Width')),
  61                 'height': height,
  62             }
  63             source_type = source.get('Type')
  64             if source_type:
  65                 f.update(parse_codecs(self._search_regex(
  66                     r'codecs="([^"]+)"', source_type, 'codecs', fatal=False)))
  67             formats.append(f)
  68
  69         self._check_formats(formats, video_guid)
  70         self._sort_formats(formats)
  71
  72         return {
  73             'id': video_guid,
  74             'title': title,
  75             'description': self._html_search_meta(['description', 'og:description'], page),
  76             'thumbnail': self._html_search_meta(['og:image:secure_url', 'og:image', 'twitter:image'], page),
  77             'duration': parse_duration(self._html_search_meta('duration', page)),
  78             'formats': formats,
  79         }