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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import os
   5 import urlparse
   6
   7 from youtube_dl import utils
   8 from .common import InfoExtractor
   9
  10
  11 class CanvasIE(InfoExtractor):
  12     _VALID_URL = r'(?:https?://)?(?:www\.)?canvas\.be/video/(?P<id>.+)'
  13     _TEST = {
  14         'url': 'http://www.canvas.be/video/de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
  15         'md5': 'ea838375a547ac787d4064d8c7860a6c',
  16         'info_dict': {
  17             'id': 'de-afspraak/najaar-2015/de-afspraak-veilt-voor-de-warmste-week',
  18             'title': 'De afspraak veilt voor de Warmste Week',
  19             'ext': 'mp4',
  20             'duration': 49,
  21         }
  22     }
  23
  24     def _real_extract(self, url):
  25         video_id = self._match_id(url)
  26
  27         webpage = self._download_webpage(url, video_id)
  28
  29         title = self._search_regex(
  30             r'<h1 class="video__body__header__title">(.+?)</h1>', webpage,
  31             'title')
  32         data_video = self._html_search_regex(
  33             r'data-video=(["\'])(?P<id>.+?)\1', webpage, 'data-video', group='id')
  34         json_url = 'https://mediazone.vrt.be/api/v1/canvas/assets/' + data_video
  35         data = self._download_json(json_url, video_id)
  36
  37         formats = []
  38         for target in data['targetUrls']:
  39             if 'type' and 'url' in target:
  40                 extension = utils.determine_ext(target['url'])
  41                 if target['type'] == 'PROGRESSIVE_DOWNLOAD':
  42                     formats.append({
  43                         'format_id': extension,
  44                         'url': target['url'],
  45                         'protocol': 'http',
  46                     })
  47                 elif target['type'] == 'HLS':
  48                     formats.extend(self._extract_m3u8_formats(
  49                         target['url'], video_id, entry_protocol='m3u8_native',
  50                         ext='mp4',
  51                         preference=0,
  52                         fatal=False,
  53                         m3u8_id='hls'))
  54                 elif target['type'] == 'HDS':
  55                     formats.append({
  56                         'format_id': extension,
  57                         'url': target['url'],
  58                         'protocol': 'HDS',
  59                     })
  60                 elif target['type'] == 'RTMP':
  61                     formats.append({
  62                         'format_id': extension,
  63                         'url': target['url'],
  64                         'protocol': 'rtmp',
  65                     })
  66                 elif target['type'] == 'RTSP':
  67                     formats.append({
  68                         'format_id': extension,
  69                         'url': target['url'],
  70                         'protocol': 'rtsp',
  71                     })
  72
  73         self._sort_formats(formats)
  74         duration = utils.int_or_none(data.get('duration')) / 1000
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'formats': formats,
  79             'duration': duration,
  80         }