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

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8
   9 class CTVNewsIE(InfoExtractor):
  10     _VALID_URL = r'https?://(?:www\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
  11     _TESTS = [{
  12         'url': 'http://www.ctvnews.ca/video?clipId=901995',
  13         'md5': '10deb320dc0ccb8d01d34d12fc2ea672',
  14         'info_dict': {
  15             'id': '901995',
  16             'ext': 'mp4',
  17             'title': 'Extended: \'That person cannot be me\' Johnson says',
  18             'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
  19             'timestamp': 1467286284,
  20             'upload_date': '20160630',
  21         }
  22     }, {
  23         'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
  24         'info_dict':
  25         {
  26             'id': '1.2966224',
  27         },
  28         'playlist_mincount': 19,
  29     }, {
  30         'url': 'http://www.ctvnews.ca/video?binId=1.810401',
  31         'info_dict':
  32         {
  33             'id': '1.810401',
  34         },
  35         'playlist_mincount': 91,
  36     }, {
  37         'url': 'http://www.ctvnews.ca/1.810401',
  38         'only_matching': True,
  39     }, {
  40         'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231',
  41         'only_matching': True,
  42     }]
  43
  44     def _real_extract(self, url):
  45         page_id = self._match_id(url)
  46
  47         def ninecninemedia_url_result(clip_id):
  48             return {
  49                 '_type': 'url_transparent',
  50                 'id': clip_id,
  51                 'url': '9c9media:ctvnews_web:%s' % clip_id,
  52                 'ie_key': 'NineCNineMedia',
  53             }
  54
  55         if page_id.isdigit():
  56             return ninecninemedia_url_result(page_id)
  57         else:
  58             webpage = self._download_webpage('http://www.ctvnews.ca/%s' % page_id, page_id, query={
  59                 'ot': 'example.AjaxPageLayout.ot',
  60                 'maxItemsPerPage': 20,
  61             })
  62             entries = [ninecninemedia_url_result(clip_id) for clip_id in set(
  63                 re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
  64             return self.playlist_result(entries, page_id)