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/series/great-outdoors/video/ice-climber-is-in-for-a-shock',
  15         'md5': 'c4cbe74c9c17c5676b704b950b73dd92',
  16         'info_dict': {
  17             'id': 'cc82397e-cc3f-4d11-9390-a785add090e8',
  18             'ext': 'mp4',
  19             'title': 'Ice Climber Is In For A Shock',
  20             'description': 'md5:55606ce1378d4c72e6545e160c9d9695',
  21             'uploader': 'TWC - Digital (No Distro)',
  22             'uploader_id': '6ccd5455-16bb-46f2-9c57-ff858bb9f62c',
  23             'upload_date': '20160720',
  24             'timestamp': 1469018835,
  25         }
  26     }, {
  27         'url': 'https://weather.com/en-CA/international/videos/video/unidentified-object-falls-from-sky-in-india',
  28         'only_matching': True,
  29     }]
  30
  31     def _real_extract(self, url):
  32         asset_name, locale, display_id = self._match_valid_url(url).groups()
  33         if not locale:
  34             locale = 'en-US'
  35         video_data = list(self._download_json(
  36             'https://weather.com/api/v1/p/redux-dal', display_id, data=json.dumps([{
  37                 'name': 'getCMSAssetsUrlConfig',
  38                 'params': {
  39                     'language': locale.replace('-', '_'),
  40                     'query': {
  41                         'assetName': {
  42                             '$in': asset_name,
  43                         },
  44                     },
  45                 }
  46             }]).encode(), headers={
  47                 'Content-Type': 'application/json',
  48             })['dal']['getCMSAssetsUrlConfig'].values())[0]['data'][0]
  49         video_id = video_data['id']
  50         seo_meta = video_data.get('seometa', {})
  51         title = video_data.get('title') or seo_meta['title']
  52
  53         urls = []
  54         thumbnails = []
  55         formats = []
  56         for variant_id, variant_url in video_data.get('variants', []).items():
  57             variant_url = variant_url.strip()
  58             if not variant_url or variant_url in urls:
  59                 continue
  60             urls.append(variant_url)
  61             ext = determine_ext(variant_url)
  62             if ext == 'jpg':
  63                 thumbnails.append({
  64                     'url': variant_url,
  65                     'id': variant_id,
  66                 })
  67             elif ThePlatformIE.suitable(variant_url):
  68                 tp_formats, _ = self._extract_theplatform_smil(variant_url, video_id)
  69                 formats.extend(tp_formats)
  70             elif ext == 'm3u8':
  71                 formats.extend(self._extract_m3u8_formats(
  72                     variant_url, video_id, 'mp4', 'm3u8_native',
  73                     m3u8_id=variant_id, fatal=False))
  74             elif ext == 'f4m':
  75                 formats.extend(self._extract_f4m_formats(
  76                     variant_url, video_id, f4m_id=variant_id, fatal=False))
  77             else:
  78                 formats.append({
  79                     'url': variant_url,
  80                     'format_id': variant_id,
  81                 })
  82
  83         cc_url = video_data.get('cc_url')
  84
  85         return {
  86             'id': video_id,
  87             'display_id': display_id,
  88             'title': title,
  89             'description': video_data.get('description') or seo_meta.get('description') or seo_meta.get('og:description'),
  90             'duration': parse_duration(video_data.get('duration')),
  91             'uploader': video_data.get('providername'),
  92             'uploader_id': video_data.get('providerid'),
  93             'timestamp': parse_iso8601(video_data.get('publishdate')),
  94             'subtitles': {locale[:2]: [{'url': cc_url}]} if cc_url else None,
  95             'thumbnails': thumbnails,
  96             'formats': formats,
  97         }