Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/democracynow.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import os.path
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_urlparse
   9 from ..utils import (
  10     url_basename,
  11     remove_start,
  12 )
  13
  14
  15 class DemocracynowIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:www\.)?democracynow\.org/(?P<id>[^\?]*)'
  17     IE_NAME = 'democracynow'
  18     _TESTS = [{
  19         'url': 'http://www.democracynow.org/shows/2015/7/3',
  20         'md5': '3757c182d3d84da68f5c8f506c18c196',
  21         'info_dict': {
  22             'id': '2015-0703-001',
  23             'ext': 'mp4',
  24             'title': 'Daily Show for July 03, 2015',
  25             'description': 'md5:80eb927244d6749900de6072c7cc2c86',
  26         },
  27     }, {
  28         'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
  29         'info_dict': {
  30             'id': '2015-0703-001',
  31             'ext': 'mp4',
  32             'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
  33             'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
  34         },
  35         'params': {
  36             'skip_download': True,
  37         },
  38     }]
  39
  40     def _real_extract(self, url):
  41         display_id = self._match_id(url)
  42
  43         webpage = self._download_webpage(url, display_id)
  44
  45         json_data = self._parse_json(self._search_regex(
  46             r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'),
  47             display_id)
  48
  49         title = json_data['title']
  50         formats = []
  51
  52         video_id = None
  53
  54         for key in ('file', 'audio', 'video', 'high_res_video'):
  55             media_url = json_data.get(key, '')
  56             if not media_url:
  57                 continue
  58             media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
  59             video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
  60             formats.append({
  61                 'url': media_url,
  62                 'vcodec': 'none' if key == 'audio' else None,
  63             })
  64
  65         self._sort_formats(formats)
  66
  67         default_lang = 'en'
  68         subtitles = {}
  69
  70         def add_subtitle_item(lang, info_dict):
  71             if lang not in subtitles:
  72                 subtitles[lang] = []
  73             subtitles[lang].append(info_dict)
  74
  75         # chapter_file are not subtitles
  76         if 'caption_file' in json_data:
  77             add_subtitle_item(default_lang, {
  78                 'url': compat_urlparse.urljoin(url, json_data['caption_file']),
  79             })
  80
  81         for subtitle_item in json_data.get('captions', []):
  82             lang = subtitle_item.get('language', '').lower() or default_lang
  83             add_subtitle_item(lang, {
  84                 'url': compat_urlparse.urljoin(url, subtitle_item['url']),
  85             })
  86
  87         description = self._og_search_description(webpage, default=None)
  88
  89         return {
  90             'id': video_id or display_id,
  91             'title': title,
  92             'description': description,
  93             'thumbnail': json_data.get('image'),
  94             'subtitles': subtitles,
  95             'formats': formats,
  96         }