youtube_dl/extractor/telegraaf.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     determine_ext,
   7     int_or_none,
   8     parse_iso8601,
   9     try_get,
  10 )
  11
  12
  13 class TelegraafIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/video/(?P<id>\d+)'
  15     _TEST = {
  16         'url': 'https://www.telegraaf.nl/video/734366489/historisch-scheepswrak-slaat-na-100-jaar-los',
  17         'info_dict': {
  18             'id': 'gaMItuoSeUg2',
  19             'ext': 'mp4',
  20             'title': 'Historisch scheepswrak slaat na 100 jaar los',
  21             'description': 'md5:6f53b7c4f55596722ac24d6c0ec00cfb',
  22             'thumbnail': r're:^https?://.*\.jpg',
  23             'duration': 55,
  24             'timestamp': 1572805527,
  25             'upload_date': '20191103',
  26         },
  27         'params': {
  28             # m3u8 download
  29             'skip_download': True,
  30         },
  31     }
  32
  33     def _real_extract(self, url):
  34         article_id = self._match_id(url)
  35
  36         video_id = self._download_json(
  37             'https://www.telegraaf.nl/graphql', article_id, query={
  38                 'query': '''{
  39   article(uid: %s) {
  40     videos {
  41       videoId
  42     }
  43   }
  44 }''' % article_id,
  45             })['data']['article']['videos'][0]['videoId']
  46
  47         item = self._download_json(
  48             'https://content.tmgvideo.nl/playlist/item=%s/playlist.json' % video_id,
  49             video_id)['items'][0]
  50         title = item['title']
  51
  52         formats = []
  53         locations = item.get('locations') or {}
  54         for location in locations.get('adaptive', []):
  55             manifest_url = location.get('src')
  56             if not manifest_url:
  57                 continue
  58             ext = determine_ext(manifest_url)
  59             if ext == 'm3u8':
  60                 formats.extend(self._extract_m3u8_formats(
  61                     manifest_url, video_id, ext='mp4', m3u8_id='hls', fatal=False))
  62             elif ext == 'mpd':
  63                 formats.extend(self._extract_mpd_formats(
  64                     manifest_url, video_id, mpd_id='dash', fatal=False))
  65             else:
  66                 self.report_warning('Unknown adaptive format %s' % ext)
  67         for location in locations.get('progressive', []):
  68             src = try_get(location, lambda x: x['sources'][0]['src'])
  69             if not src:
  70                 continue
  71             label = location.get('label')
  72             formats.append({
  73                 'url': src,
  74                 'width': int_or_none(location.get('width')),
  75                 'height': int_or_none(location.get('height')),
  76                 'format_id': 'http' + ('-%s' % label if label else ''),
  77             })
  78
  79         self._sort_formats(formats)
  80
  81         return {
  82             'id': video_id,
  83             'title': title,
  84             'description': item.get('description'),
  85             'formats': formats,
  86             'duration': int_or_none(item.get('duration')),
  87             'thumbnail': item.get('poster'),
  88             'timestamp': parse_iso8601(item.get('datecreated'), ' '),
  89         }