yt_dlp/extractor/maariv.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     int_or_none,
   4     parse_resolution,
   5     unified_timestamp,
   6     url_or_none,
   7 )
   8 from ..utils.traversal import traverse_obj
   9
  10
  11 class MaarivIE(InfoExtractor):
  12     IE_NAME = 'maariv.co.il'
  13     _VALID_URL = r'https?://player\.maariv\.co\.il/public/player\.html\?(?:[^#]+&)?media=(?P<id>\d+)'
  14     _EMBED_REGEX = [rf'<iframe[^>]+\bsrc=[\'"](?P<url>{_VALID_URL})']
  15     _TESTS = [{
  16         'url': 'https://player.maariv.co.il/public/player.html?player=maariv-desktop&media=3611585',
  17         'info_dict': {
  18             'id': '3611585',
  19             'duration': 75,
  20             'ext': 'mp4',
  21             'upload_date': '20231009',
  22             'title': 'מבצע חרבות ברזל',
  23             'timestamp': 1696851301,
  24         },
  25     }]
  26     _WEBPAGE_TESTS = [{
  27         'url': 'https://www.maariv.co.il/news/law/Article-1044008',
  28         'info_dict': {
  29             'id': '3611585',
  30             'duration': 75,
  31             'ext': 'mp4',
  32             'upload_date': '20231009',
  33             'title': 'מבצע חרבות ברזל',
  34             'timestamp': 1696851301,
  35         },
  36     }]
  37
  38     def _real_extract(self, url):
  39         video_id = self._match_id(url)
  40         data = self._download_json(
  41             f'https://dal.walla.co.il/media/{video_id}?origin=player.maariv.co.il', video_id)['data']
  42
  43         formats = []
  44         if hls_url := traverse_obj(data, ('video', 'url', {url_or_none})):
  45             formats.extend(self._extract_m3u8_formats(hls_url, video_id, m3u8_id='hls', fatal=False))
  46
  47         for http_format in traverse_obj(data, ('video', 'stream_urls', ..., 'stream_url', {url_or_none})):
  48             formats.append({
  49                 'url': http_format,
  50                 'format_id': 'http',
  51                 **parse_resolution(http_format),
  52             })
  53
  54         return {
  55             'id': video_id,
  56             **traverse_obj(data, {
  57                 'title': 'title',
  58                 'duration': ('video', 'duration', {int_or_none}),
  59                 'timestamp': ('upload_date', {unified_timestamp}),
  60             }),
  61             'formats': formats,
  62         }