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': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d',
  21         'info_dict': {
  22             'id': '2015-0703-001',
  23             'ext': 'mp4',
  24             'title': 'July 03, 2015 - Democracy Now!',
  25             'description': 'A daily independent global news hour with Amy Goodman & Juan González "What to the Slave is 4th of July?": James Earl Jones Reads Frederick Douglass\u2019 Historic Speech : "This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag : "We Shall Overcome": Remembering Folk Icon, Activist Pete Seeger in His Own Words & Songs',
  26         },
  27     }, {
  28         'url': 'http://www.democracynow.org/2015/7/3/this_flag_comes_down_today_bree',
  29         'md5': 'fbb8fe3d7a56a5e12431ce2f9b2fab0d',
  30         'info_dict': {
  31             'id': '2015-0703-001',
  32             'ext': 'mp4',
  33             'title': '"This Flag Comes Down Today": Bree Newsome Scales SC Capitol Flagpole, Takes Down Confederate Flag',
  34             'description': 'md5:4d2bc4f0d29f5553c2210a4bc7761a21',
  35         },
  36     }]
  37
  38     def _real_extract(self, url):
  39         display_id = self._match_id(url)
  40         webpage = self._download_webpage(url, display_id)
  41         description = self._og_search_description(webpage)
  42
  43         json_data = self._parse_json(self._search_regex(
  44             r'<script[^>]+type="text/json"[^>]*>\s*({[^>]+})', webpage, 'json'),
  45             display_id)
  46         video_id = None
  47         formats = []
  48
  49         default_lang = 'en'
  50
  51         subtitles = {}
  52
  53         def add_subtitle_item(lang, info_dict):
  54             if lang not in subtitles:
  55                 subtitles[lang] = []
  56             subtitles[lang].append(info_dict)
  57
  58         # chapter_file are not subtitles
  59         if 'caption_file' in json_data:
  60             add_subtitle_item(default_lang, {
  61                 'url': compat_urlparse.urljoin(url, json_data['caption_file']),
  62             })
  63
  64         for subtitle_item in json_data.get('captions', []):
  65             lang = subtitle_item.get('language', '').lower() or default_lang
  66             add_subtitle_item(lang, {
  67                 'url': compat_urlparse.urljoin(url, subtitle_item['url']),
  68             })
  69
  70         for key in ('file', 'audio', 'video'):
  71             media_url = json_data.get(key, '')
  72             if not media_url:
  73                 continue
  74             media_url = re.sub(r'\?.*', '', compat_urlparse.urljoin(url, media_url))
  75             video_id = video_id or remove_start(os.path.splitext(url_basename(media_url))[0], 'dn')
  76             formats.append({
  77                 'url': media_url,
  78             })
  79
  80         self._sort_formats(formats)
  81
  82         return {
  83             'id': video_id or display_id,
  84             'title': json_data['title'],
  85             'description': description,
  86             'subtitles': subtitles,
  87             'formats': formats,
  88         }