yt_dlp/extractor/kompas.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     clean_html,
   4     float_or_none,
   5     traverse_obj,
   6     try_call,
   7 )
   8
   9 # Video from www.kompas.tv and video.kompas.com seems use jixie player
  10 # see [1] https://jixie.atlassian.net/servicedesk/customer/portal/2/article/1339654214?src=-1456335525,
  11 # [2] https://scripts.jixie.media/jxvideo.3.1.min.js for more info
  12
  13
  14 class KompasVideoIE(InfoExtractor):
  15     _VALID_URL = r'https?://video\.kompas\.com/\w+/(?P<id>\d+)/(?P<slug>[\w-]+)'
  16     _TESTS = [{
  17         'url': 'https://video.kompas.com/watch/164474/kim-jong-un-siap-kirim-nuklir-lawan-as-dan-korsel',
  18         'info_dict': {
  19             'id': '164474',
  20             'ext': 'mp4',
  21             'title': 'Kim Jong Un Siap Kirim Nuklir Lawan AS dan Korsel',
  22             'description': 'md5:262530c4fb7462398235f9a5dba92456',
  23             'uploader_id': '9262bf2590d558736cac4fff7978fcb1',
  24             'display_id': 'kim-jong-un-siap-kirim-nuklir-lawan-as-dan-korsel',
  25             'duration': 85.066667,
  26             'categories': ['news'],
  27             'thumbnail': 'https://video.jixie.media/1001/164474/164474_1280x720.jpg',
  28             'tags': 'count:9',
  29         }
  30     }]
  31
  32     def _real_extract(self, url):
  33         video_id, display_id = self._match_valid_url(url).group('id', 'slug')
  34         webpage = self._download_webpage(url, display_id)
  35
  36         json_data = self._download_json(
  37             'https://apidam.jixie.io/api/public/stream', display_id,
  38             query={'metadata': 'full', 'video_id': video_id})['data']
  39
  40         formats, subtitles = [], {}
  41         for stream in json_data['streams']:
  42             if stream.get('type') == 'HLS':
  43                 fmt, sub = self._extract_m3u8_formats_and_subtitles(stream.get('url'), display_id, ext='mp4')
  44                 formats.extend(fmt)
  45                 self._merge_subtitles(sub, target=subtitles)
  46             else:
  47                 formats.append({
  48                     'url': stream.get('url'),
  49                     'width': stream.get('width'),
  50                     'height': stream.get('height'),
  51                     'ext': 'mp4',
  52                 })
  53
  54         self._sort_formats(formats)
  55         return {
  56             'id': video_id,
  57             'display_id': display_id,
  58             'formats': formats,
  59             'subtitles': subtitles,
  60             'title': json_data.get('title') or self._html_search_meta(['og:title', 'twitter:title'], webpage),
  61             'description': (clean_html(traverse_obj(json_data, ('metadata', 'description')))
  62                             or self._html_search_meta(['description', 'og:description', 'twitter:description'], webpage)),
  63             'thumbnails': traverse_obj(json_data, ('metadata', 'thumbnails')),
  64             'duration': float_or_none(traverse_obj(json_data, ('metadata', 'duration'))),
  65             'tags': try_call(lambda: json_data['metadata']['keywords'].split(',')),
  66             'categories': try_call(lambda: json_data['metadata']['categories'].split(',')),
  67             'uploader_id': json_data.get('owner_id'),
  68         }