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

   1 # encoding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7
   8 from ..utils import (
   9     ExtractorError,
  10 )
  11
  12 class TheInterceptIE(InfoExtractor):
  13     _VALID_URL = r'https://theintercept.com/fieldofvision/(?P<id>.+?)/'
  14     _TESTS = [{
  15         'url': 'https://theintercept.com/fieldofvision/thisisacoup-episode-four-surrender-or-die/',
  16         'info_dict': {
  17             'id': 'thisisacoup-episode-four-surrender-or-die',
  18             'ext': 'mp4',
  19             'title': '#ThisIsACoup – Episode Four: Surrender or Die',
  20             'upload_date': '20151218',
  21             'description': 'md5:74dd27f0e2fbd50817829f97eaa33140',
  22         }
  23     }]
  24
  25     def _real_extract(self, url):
  26         display_id = self._match_id(url)
  27         webpage = self._download_webpage(url, display_id)
  28
  29         mobj = re.search(r'initialStoreTree =(?P<json_data>.+})', webpage)
  30         if mobj is None:
  31             raise ExtractorError('Unable to extract initialStoreTree')
  32         json_data = self._parse_json(mobj.group('json_data'), display_id)
  33
  34         info = None
  35         for post in json_data['resources']['posts'].values():
  36             if post['slug'] == display_id:
  37                 info = post
  38                 break
  39         if info is None:
  40             raise ExtractorError('Unable to find info for %s'%display_id)
  41
  42         title = info['title']
  43         description = info['excerpt']
  44         upload_date = info['date'][:10].replace('-', '')
  45         video_id = info['fov_videoid']
  46         creator = ','.join([a['display_name'] for a in info['authors']])
  47         thumbnail = self._og_search_property('image', webpage)
  48         content_id = thumbnail.split('/')[-1].split('.')[0]
  49         content_url = 'https://content.jwplatform.com/jw6/{content_id}.xml'.format(content_id=content_id)
  50         content = self._download_xml(content_url, video_id)
  51
  52         formats = []
  53         for source in content.findall('.//{http://rss.jwpcdn.com/}source'):
  54             if source.attrib['file'].endswith('.m3u8'):
  55                 formats.extend(self._extract_m3u8_formats(
  56                     source.attrib['file'], video_id, 'mp4', preference=1, m3u8_id='hls'))
  57
  58         return {
  59             'creator': creator,
  60             'description': description,
  61             'display_id': display_id,
  62             'formats': formats,
  63             'id': video_id,
  64             'id': video_id,
  65             'thumbnail': thumbnail,
  66             'title': title,
  67             'upload_date': upload_date,
  68         }