yt_dlp/extractor/rtvnh.py

   1 from .common import InfoExtractor
   2 from ..utils import ExtractorError
   3
   4
   5 class RTVNHIE(InfoExtractor):
   6     _VALID_URL = r'https?://(?:www\.)?rtvnh\.nl/video/(?P<id>[0-9]+)'
   7     _TEST = {
   8         'url': 'http://www.rtvnh.nl/video/131946',
   9         'md5': 'cdbec9f44550763c8afc96050fa747dc',
  10         'info_dict': {
  11             'id': '131946',
  12             'ext': 'mp4',
  13             'title': 'Grote zoektocht in zee bij Zandvoort naar vermiste vrouw',
  14             'thumbnail': r're:^https?:.*\.jpg$'
  15         }
  16     }
  17
  18     def _real_extract(self, url):
  19         video_id = self._match_id(url)
  20
  21         meta = self._parse_json(self._download_webpage(
  22             'http://www.rtvnh.nl/video/json?m=' + video_id, video_id), video_id)
  23
  24         status = meta.get('status')
  25         if status != 200:
  26             raise ExtractorError(
  27                 '%s returned error code %d' % (self.IE_NAME, status), expected=True)
  28
  29         formats = []
  30         rtmp_formats = self._extract_smil_formats(
  31             'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id)
  32         formats.extend(rtmp_formats)
  33
  34         for rtmp_format in rtmp_formats:
  35             rtmp_url = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
  36             rtsp_format = rtmp_format.copy()
  37             del rtsp_format['play_path']
  38             del rtsp_format['ext']
  39             rtsp_format.update({
  40                 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
  41                 'url': rtmp_url.replace('rtmp://', 'rtsp://'),
  42                 'protocol': 'rtsp',
  43             })
  44             formats.append(rtsp_format)
  45             http_base_url = rtmp_url.replace('rtmp://', 'http://')
  46             formats.extend(self._extract_m3u8_formats(
  47                 http_base_url + '/playlist.m3u8', video_id, 'mp4',
  48                 'm3u8_native', m3u8_id='hls', fatal=False))
  49             formats.extend(self._extract_f4m_formats(
  50                 http_base_url + '/manifest.f4m',
  51                 video_id, f4m_id='hds', fatal=False))
  52
  53         return {
  54             'id': video_id,
  55             'title': meta['title'].strip(),
  56             'thumbnail': meta.get('image'),
  57             'formats': formats
  58         }