Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/rtvnh.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 from .common import InfoExtractor
   5 from ..utils import ExtractorError
   6
   7
   8 class RTVNHIE(InfoExtractor):
   9     _VALID_URL = r'https?://(?:www\.)?rtvnh\.nl/video/(?P<id>[0-9]+)'
  10     _TEST = {
  11         'url': 'http://www.rtvnh.nl/video/131946',
  12         'md5': 'cdbec9f44550763c8afc96050fa747dc',
  13         'info_dict': {
  14             'id': '131946',
  15             'ext': 'mp4',
  16             'title': 'Grote zoektocht in zee bij Zandvoort naar vermiste vrouw',
  17             'thumbnail': r're:^https?:.*\.jpg$'
  18         }
  19     }
  20
  21     def _real_extract(self, url):
  22         video_id = self._match_id(url)
  23
  24         meta = self._parse_json(self._download_webpage(
  25             'http://www.rtvnh.nl/video/json?m=' + video_id, video_id), video_id)
  26
  27         status = meta.get('status')
  28         if status != 200:
  29             raise ExtractorError(
  30                 '%s returned error code %d' % (self.IE_NAME, status), expected=True)
  31
  32         formats = []
  33         rtmp_formats = self._extract_smil_formats(
  34             'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id)
  35         formats.extend(rtmp_formats)
  36
  37         for rtmp_format in rtmp_formats:
  38             rtmp_url = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
  39             rtsp_format = rtmp_format.copy()
  40             del rtsp_format['play_path']
  41             del rtsp_format['ext']
  42             rtsp_format.update({
  43                 'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
  44                 'url': rtmp_url.replace('rtmp://', 'rtsp://'),
  45                 'protocol': 'rtsp',
  46             })
  47             formats.append(rtsp_format)
  48             http_base_url = rtmp_url.replace('rtmp://', 'http://')
  49             formats.extend(self._extract_m3u8_formats(
  50                 http_base_url + '/playlist.m3u8', video_id, 'mp4',
  51                 'm3u8_native', m3u8_id='hls', fatal=False))
  52             formats.extend(self._extract_f4m_formats(
  53                 http_base_url + '/manifest.f4m',
  54                 video_id, f4m_id='hds', fatal=False))
  55         self._sort_formats(formats)
  56
  57         return {
  58             'id': video_id,
  59             'title': meta['title'].strip(),
  60             'thumbnail': meta.get('image'),
  61             'formats': formats
  62         }