yt_dlp/extractor/gbnews.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     ExtractorError,
   4     extract_attributes,
   5     get_elements_html_by_class,
   6     url_or_none,
   7 )
   8 from ..utils.traversal import traverse_obj
   9
  10
  11 class GBNewsIE(InfoExtractor):
  12     IE_DESC = 'GB News clips, features and live streams'
  13     _VALID_URL = r'https?://(?:www\.)?gbnews\.(?:uk|com)/(?:\w+/)?(?P<id>[^#?]+)'
  14
  15     _PLATFORM = 'safari'
  16     _SSMP_URL = 'https://mm-v2.simplestream.com/ssmp/api.php'
  17     _TESTS = [{
  18         'url': 'https://www.gbnews.com/news/bbc-claudine-gay-harvard-university-antisemitism-row',
  19         'info_dict': {
  20             'id': '52264136',
  21             'ext': 'mp4',
  22             'thumbnail': r're:https?://www\.gbnews\.\w+/.+\.(?:jpe?g|png|webp)',
  23             'display_id': 'bbc-claudine-gay-harvard-university-antisemitism-row',
  24             'description': 'The post was criticised by former employers of the broadcaster',
  25             'title': 'BBC deletes post after furious backlash over headline downplaying antisemitism',
  26         },
  27     }, {
  28         'url': 'https://www.gbnews.com/royal/prince-harry-in-love-with-kate-meghan-markle-jealous-royal',
  29         'info_dict': {
  30             'id': '52328390',
  31             'ext': 'mp4',
  32             'thumbnail': r're:https?://www\.gbnews\.\w+/.+\.(?:jpe?g|png|webp)',
  33             'display_id': 'prince-harry-in-love-with-kate-meghan-markle-jealous-royal',
  34             'description': 'Ingrid Seward has published 17 books documenting the highs and lows of the Royal Family',
  35             'title': 'Royal author claims Prince Harry was \'in love\' with Kate - Meghan was \'jealous\'',
  36         },
  37     }, {
  38         'url': 'https://www.gbnews.uk/watchlive',
  39         'info_dict': {
  40             'id': '1069',
  41             'ext': 'mp4',
  42             'thumbnail': r're:https?://www\.gbnews\.\w+/.+\.(?:jpe?g|png|webp)',
  43             'display_id': 'watchlive',
  44             'live_status': 'is_live',
  45             'title': r're:^GB News Live',
  46         },
  47         'params': {'skip_download': 'm3u8'},
  48     }]
  49     _SS_ENDPOINTS = None
  50
  51     def _get_ss_endpoint(self, data_id, data_env):
  52         if not self._SS_ENDPOINTS:
  53             self._SS_ENDPOINTS = {}
  54
  55         if not data_id:
  56             data_id = 'GB003'
  57         if not data_env:
  58             data_env = 'production'
  59         key = data_id, data_env
  60         result = self._SS_ENDPOINTS.get(key)
  61         if result:
  62             return result
  63
  64         json_data = self._download_json(
  65             self._SSMP_URL, None, 'Downloading Simplestream JSON metadata', query={
  66                 'id': data_id,
  67                 'env': data_env,
  68             })
  69         meta_url = traverse_obj(json_data, ('response', 'api_hostname', {url_or_none}))
  70         if not meta_url:
  71             raise ExtractorError('No API host found')
  72
  73         self._SS_ENDPOINTS[key] = meta_url
  74         return meta_url
  75
  76     def _real_extract(self, url):
  77         display_id = self._match_id(url).rpartition('/')[2]
  78         webpage = self._download_webpage(url, display_id)
  79
  80         video_data = None
  81         elements = get_elements_html_by_class('simplestream', webpage)
  82         for html_tag in elements:
  83             attributes = extract_attributes(html_tag)
  84             if 'sidebar' not in (attributes.get('class') or ''):
  85                 video_data = attributes
  86         if not video_data:
  87             raise ExtractorError('Could not find video element', expected=True)
  88
  89         endpoint_url = self._get_ss_endpoint(video_data.get('data-id'), video_data.get('data-env'))
  90
  91         uvid = video_data['data-uvid']
  92         video_type = video_data.get('data-type')
  93         if not video_type or video_type == 'vod':
  94             video_type = 'show'
  95         stream_data = self._download_json(
  96             f'{endpoint_url}/api/{video_type}/stream/{uvid}',
  97             uvid, 'Downloading stream JSON', query={
  98                 'key': video_data.get('data-key'),
  99                 'platform': self._PLATFORM,
 100             })
 101         if traverse_obj(stream_data, 'drm'):
 102             self.report_drm(uvid)
 103
 104         return {
 105             'id': uvid,
 106             'display_id': display_id,
 107             'title': self._og_search_title(webpage, default=None),
 108             'description': self._og_search_description(webpage, default=None),
 109             'formats': self._extract_m3u8_formats(traverse_obj(stream_data, (
 110                 'response', 'stream', {url_or_none})), uvid, 'mp4'),
 111             'thumbnail': self._og_search_thumbnail(webpage, default=None),
 112             'is_live': video_type == 'live',
 113         }