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