yt_dlp/extractor/democracynow.py

   1 import os.path
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..compat import compat_urlparse
   6 from ..utils import (
   7     remove_start,
   8     url_basename,
   9 )
  10
  11
  12 class DemocracynowIE(InfoExtractor):
  13     _VALID_URL = r'https?://(?:www\.)?democracynow\.org/(?P<id>[^\?]*)'
  14     IE_NAME = 'democracynow'
  15     _TESTS = [{
  16         'url': 'http://www.democracynow.org/shows/2015/7/3',
  17         'md5': '3757c182d3d84da68f5c8f506c18c196',
  18         'info_dict': {
  19             'id': '2015-0703-001',
  20             'ext': 'mp4',
  21             'title': 'Daily Show for July 03, 2015',
  22             'description': 'md5:80eb927244d6749900de6072c7cc2c86',
  23         },
  24     }, {
  25         'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
  26         'info_dict': {
  27             'id': '2015-0703-001',
  28             'ext': 'mp4',
  29             'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
  30             'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
  31         },
  32         'params': {
  33             'skip_download': True,
  34         },
  35     }]
  36
  37     def _real_extract(self, url):
  38         display_id = self._match_id(url)
  39
  40         webpage = self._download_webpage(url, display_id)
  41
  42         json_data = self._parse_json(self._search_regex(
  43             r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'),
  44             display_id)
  45
  46         title = json_data['title']
  47         formats = []
  48
  49         video_id = None
  50
  51         for key in ('file', 'audio', 'video', 'high_res_video'):
  52             media_url = json_data.get(key, '')
  53             if not media_url:
  54                 continue
  55             media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
  56             video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
  57             formats.append({
  58                 'url': media_url,
  59                 'vcodec': 'none' if key == 'audio' else None,
  60             })
  61
  62         default_lang = 'en'
  63         subtitles = {}
  64
  65         def add_subtitle_item(lang, info_dict):
  66             if lang not in subtitles:
  67                 subtitles[lang] = []
  68             subtitles[lang].append(info_dict)
  69
  70         # chapter_file are not subtitles
  71         if 'caption_file' in json_data:
  72             add_subtitle_item(default_lang, {
  73                 'url': compat_urlparse.urljoin(url, json_data['caption_file']),
  74             })
  75
  76         for subtitle_item in json_data.get('captions', []):
  77             lang = subtitle_item.get('language', '').lower() or default_lang
  78             add_subtitle_item(lang, {
  79                 'url': compat_urlparse.urljoin(url, subtitle_item['url']),
  80             })
  81
  82         description = self._og_search_description(webpage, default=None)
  83
  84         return {
  85             'id': video_id or display_id,
  86             'title': title,
  87             'description': description,
  88             'thumbnail': json_data.get('image'),
  89             'subtitles': subtitles,
  90             'formats': formats,
  91         }