yt_dlp/extractor/theweatherchannel.py

   1 import json
   2
   3 from .theplatform import ThePlatformIE
   4 from ..utils import (
   5     determine_ext,
   6     parse_duration,
   7     parse_iso8601,
   8 )
   9
  10
  11 class TheWeatherChannelIE(ThePlatformIE):  # XXX: Do not subclass from concrete IE
  12     _VALID_URL = r'https?://(?:www\.)?weather\.com(?P<asset_name>(?:/(?P<locale>[a-z]{2}-[A-Z]{2}))?/(?:[^/]+/)*video/(?P<id>[^/?#]+))'
  13     _TESTS = [{
  14         'url': 'https://weather.com/storms/hurricane/video/invest-95l-in-atlantic-has-a-medium-chance-of-development',
  15         'md5': '68f0cf616435683f27ce36bd9c927394',
  16         'info_dict': {
  17             'id': '81acef2d-ee8c-4545-ba83-bff3cc80db97',
  18             'ext': 'mp4',
  19             'title': 'Invest 95L In Atlantic Has A Medium Chance Of Development',
  20             'description': 'md5:0de720fd5f0d0e32207bd4c270fff824',
  21             'uploader': 'TWC - Digital',
  22             'uploader_id': 'b5a999e0-9e04-11e1-9ee2-001d092f5a10',
  23             'upload_date': '20230721',
  24             'timestamp': 1689967343,
  25             'display_id': 'invest-95l-in-atlantic-has-a-medium-chance-of-development',
  26             'duration': 34.0,
  27         },
  28     }, {
  29         'url': 'https://weather.com/en-CA/international/videos/video/unidentified-object-falls-from-sky-in-india',
  30         'only_matching': True,
  31     }]
  32
  33     def _real_extract(self, url):
  34         asset_name, locale, display_id = self._match_valid_url(url).groups()
  35         if not locale:
  36             locale = 'en-US'
  37         video_data = next(iter(self._download_json(
  38             'https://weather.com/api/v1/p/redux-dal', display_id, data=json.dumps([{
  39                 'name': 'getCMSAssetsUrlConfig',
  40                 'params': {
  41                     'language': locale.replace('-', '_'),
  42                     'query': {
  43                         'assetName': {
  44                             '$in': asset_name,
  45                         },
  46                     },
  47                 },
  48             }]).encode(), headers={
  49                 'Content-Type': 'application/json',
  50             })['dal']['getCMSAssetsUrlConfig'].values()))['data'][0]
  51         video_id = video_data['id']
  52         seo_meta = video_data.get('seometa', {})
  53         title = video_data.get('title') or seo_meta['title']
  54
  55         urls = []
  56         thumbnails = []
  57         formats = []
  58         for variant_id, variant_url in video_data.get('variants', []).items():
  59             variant_url = variant_url.strip()
  60             if not variant_url or variant_url in urls:
  61                 continue
  62             urls.append(variant_url)
  63             ext = determine_ext(variant_url)
  64             if ext == 'jpg':
  65                 thumbnails.append({
  66                     'url': variant_url,
  67                     'id': variant_id,
  68                 })
  69             elif ThePlatformIE.suitable(variant_url):
  70                 tp_formats, _ = self._extract_theplatform_smil(variant_url, video_id)
  71                 formats.extend(tp_formats)
  72             elif ext == 'm3u8':
  73                 formats.extend(self._extract_m3u8_formats(
  74                     variant_url, video_id, 'mp4', 'm3u8_native',
  75                     m3u8_id=variant_id, fatal=False))
  76             elif ext == 'f4m':
  77                 formats.extend(self._extract_f4m_formats(
  78                     variant_url, video_id, f4m_id=variant_id, fatal=False))
  79             else:
  80                 formats.append({
  81                     'url': variant_url,
  82                     'format_id': variant_id,
  83                 })
  84
  85         cc_url = video_data.get('cc_url')
  86
  87         return {
  88             'id': video_id,
  89             'display_id': display_id,
  90             'title': title,
  91             'description': video_data.get('description') or seo_meta.get('description') or seo_meta.get('og:description'),
  92             'duration': parse_duration(video_data.get('duration')),
  93             'uploader': video_data.get('providername'),
  94             'uploader_id': video_data.get('providerid'),
  95             'timestamp': parse_iso8601(video_data.get('publishdate')),
  96             'subtitles': {locale[:2]: [{'url': cc_url}]} if cc_url else None,
  97             'thumbnails': thumbnails,
  98             'formats': formats,
  99         }