_ Git - youtube-dl/blob - youtube_dl/extractor/theweatherchannel.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .theplatform import ThePlatformIE
   5 from ..utils import (
   6     determine_ext,
   7     parse_duration,
   8 )
   9
  10
  11 class TheWeatherChannelIE(ThePlatformIE):
  12     _VALID_URL = r'https?://(?:www\.)?weather\.com/(?:[^/]+/)*video/(?P<id>[^/?#]+)'
  13     _TESTS = [{
  14         'url': 'https://weather.com/series/great-outdoors/video/ice-climber-is-in-for-a-shock',
  15         'md5': 'ab924ac9574e79689c24c6b95e957def',
  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         }
  24     }]
  25
  26     def _real_extract(self, url):
  27         display_id = self._match_id(url)
  28         webpage = self._download_webpage(url, display_id)
  29         drupal_settings = self._parse_json(self._search_regex(
  30             r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
  31             webpage, 'drupal settings'), display_id)
  32         video_id = drupal_settings['twc']['contexts']['node']['uuid']
  33         video_data = self._download_json(
  34             'https://dsx.weather.com/cms/v4/asset-collection/en_US/' + video_id, video_id)
  35         seo_meta = video_data.get('seometa', {})
  36         title = video_data.get('title') or seo_meta['title']
  37
  38         urls = []
  39         thumbnails = []
  40         formats = []
  41         for variant_id, variant_url in video_data.get('variants', []).items():
  42             variant_url = variant_url.strip()
  43             if not variant_url or variant_url in urls:
  44                 continue
  45             urls.append(variant_url)
  46             ext = determine_ext(variant_url)
  47             if ext == 'jpg':
  48                 thumbnails.append({
  49                     'url': variant_url,
  50                     'id': variant_id,
  51                 })
  52             elif ThePlatformIE.suitable(variant_url):
  53                 tp_formats, _ = self._extract_theplatform_smil(variant_url, video_id)
  54                 formats.extend(tp_formats)
  55             elif ext == 'm3u8':
  56                 formats.extend(self._extract_m3u8_formats(
  57                     variant_url, video_id, 'mp4', 'm3u8_native',
  58                     m3u8_id=variant_id, fatal=False))
  59             elif ext == 'f4m':
  60                 formats.extend(self._extract_f4m_formats(
  61                     variant_url, video_id, f4m_id=variant_id, fatal=False))
  62             else:
  63                 formats.append({
  64                     'url': variant_url,
  65                     'format_id': variant_id,
  66                 })
  67         self._sort_formats(formats)
  68
  69         return {
  70             'id': video_id,
  71             'display_id': display_id,
  72             'title': title,
  73             'description': video_data.get('description') or seo_meta.get('description') or seo_meta.get('og:description'),
  74             'duration': parse_duration(video_data.get('duration')),
  75             'uploader': video_data.get('providername'),
  76             'uploader_id': video_data.get('providerid'),
  77             'thumbnails': thumbnails,
  78             'formats': formats,
  79         }