yt_dlp/extractor/tass.py

   1 import json
   2
   3 from .common import InfoExtractor
   4 from ..utils import (
   5     js_to_json,
   6     qualities,
   7 )
   8
   9
  10 class TassIE(InfoExtractor):
  11     _WORKING = False
  12     _VALID_URL = r'https?://(?:tass\.ru|itar-tass\.com)/[^/]+/(?P<id>\d+)'
  13     _TESTS = [
  14         {
  15             'url': 'http://tass.ru/obschestvo/1586870',
  16             'md5': '3b4cdd011bc59174596b6145cda474a4',
  17             'info_dict': {
  18                 'id': '1586870',
  19                 'ext': 'mp4',
  20                 'title': 'Посетителям московского зоопарка показали красную панду',
  21                 'description': 'Приехавшую из Дублина Зейну можно увидеть в павильоне "Кошки тропиков"',
  22                 'thumbnail': r're:^https?://.*\.jpg$',
  23             },
  24         },
  25         {
  26             'url': 'http://itar-tass.com/obschestvo/1600009',
  27             'only_matching': True,
  28         },
  29     ]
  30
  31     def _real_extract(self, url):
  32         video_id = self._match_id(url)
  33
  34         webpage = self._download_webpage(url, video_id)
  35
  36         sources = json.loads(js_to_json(self._search_regex(
  37             r'(?s)sources\s*:\s*(\[.+?\])', webpage, 'sources')))
  38
  39         quality = qualities(['sd', 'hd'])
  40
  41         formats = []
  42         for source in sources:
  43             video_url = source.get('file')
  44             if not video_url or not video_url.startswith('http') or not video_url.endswith('.mp4'):
  45                 continue
  46             label = source.get('label')
  47             formats.append({
  48                 'url': video_url,
  49                 'format_id': label,
  50                 'quality': quality(label),
  51             })
  52
  53         return {
  54             'id': video_id,
  55             'title': self._og_search_title(webpage),
  56             'description': self._og_search_description(webpage),
  57             'thumbnail': self._og_search_thumbnail(webpage),
  58             'formats': formats,
  59         }