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

   1 from __future__ import unicode_literals
   2
   3 from .common import InfoExtractor
   4 from ..utils import float_or_none
   5
   6
   7 class CanvasIE(InfoExtractor):
   8     _VALID_URL = r'https?://(?:www\.)?canvas\.be/video/(?:[^/]+/)*(?P<id>[^/?#&]+)'
   9     _TEST = {
  10         'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
  11         'md5': 'ea838375a547ac787d4064d8c7860a6c',
  12         'info_dict': {
  13             'id': 'mz-ast-5e5f90b6-2d72-4c40-82c2-e134f884e93e',
  14             'display_id': 'de-afspraak-veilt-voor-de-warmste-week',
  15             'ext': 'mp4',
  16             'title': 'De afspraak veilt voor de Warmste Week',
  17             'description': 'md5:24cb860c320dc2be7358e0e5aa317ba6',
  18             'thumbnail': 're:^https?://.*\.jpg$',
  19             'duration': 49.02,
  20         }
  21     }
  22
  23     def _real_extract(self, url):
  24         display_id = self._match_id(url)
  25
  26         webpage = self._download_webpage(url, display_id)
  27
  28         title = self._search_regex(
  29             r'<h1[^>]+class="video__body__header__title"[^>]*>(.+?)</h1>',
  30             webpage, 'title', default=None) or self._og_search_title(webpage)
  31
  32         video_id = self._html_search_regex(
  33             r'data-video=(["\'])(?P<id>.+?)\1', webpage, 'video id', group='id')
  34
  35         data = self._download_json(
  36             'https://mediazone.vrt.be/api/v1/canvas/assets/%s' % video_id, display_id)
  37
  38         formats = []
  39         for target in data['targetUrls']:
  40             format_url, format_type = target.get('url'), target.get('type')
  41             if not format_url or not format_type:
  42                 continue
  43             if format_type == 'HLS':
  44                 formats.extend(self._extract_m3u8_formats(
  45                     format_url, display_id, entry_protocol='m3u8_native',
  46                     ext='mp4', preference=0, fatal=False, m3u8_id=format_type))
  47             elif format_type == 'HDS':
  48                 formats.extend(self._extract_f4m_formats(
  49                     format_url, display_id, f4m_id=format_type, fatal=False))
  50             else:
  51                 formats.append({
  52                     'format_id': format_type,
  53                     'url': format_url,
  54                 })
  55         self._sort_formats(formats)
  56
  57         return {
  58             'id': video_id,
  59             'display_id': display_id,
  60             'title': title,
  61             'description': self._og_search_description(webpage),
  62             'formats': formats,
  63             'duration': float_or_none(data.get('duration'), 1000),
  64             'thumbnail': data.get('posterImageUrl'),
  65         }