]> gitweb @ CieloNegro.org - youtube-dl.git/blob - youtube_dl/extractor/megaphone.py
[cbs] extract smpte and vtt subtitles
[youtube-dl.git] / youtube_dl / extractor / megaphone.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..utils import js_to_json
8
9
10 class MegaphoneIE(InfoExtractor):
11     IE_NAME = 'megaphone.fm'
12     IE_DESC = 'megaphone.fm embedded players'
13     _VALID_URL = r'https://player\.megaphone\.fm/(?P<id>[A-Z0-9]+)'
14     _TEST = {
15         'url': 'https://player.megaphone.fm/GLT9749789991?"',
16         'md5': '4816a0de523eb3e972dc0dda2c191f96',
17         'info_dict': {
18             'id': 'GLT9749789991',
19             'ext': 'mp3',
20             'title': '#97 What Kind Of Idiot Gets Phished?',
21             'thumbnail': r're:^https://.*\.png.*$',
22             'duration': 1776.26375,
23             'author': 'Reply All',
24         },
25     }
26
27     def _real_extract(self, url):
28         video_id = self._match_id(url)
29         webpage = self._download_webpage(url, video_id)
30
31         title = self._og_search_property('audio:title', webpage)
32         author = self._og_search_property('audio:artist', webpage)
33         thumbnail = self._og_search_thumbnail(webpage)
34
35         episode_json = self._search_regex(r'(?s)var\s+episode\s*=\s*(\{.+?\});', webpage, 'episode JSON')
36         episode_data = self._parse_json(episode_json, video_id, js_to_json)
37         video_url = self._proto_relative_url(episode_data['mediaUrl'], 'https:')
38
39         formats = [{
40             'url': video_url,
41         }]
42
43         return {
44             'id': video_id,
45             'thumbnail': thumbnail,
46             'title': title,
47             'author': author,
48             'duration': episode_data['duration'],
49             'formats': formats,
50         }
51
52     @classmethod
53     def _extract_urls(cls, webpage):
54         return [m[0] for m in re.findall(
55             r'<iframe[^>]*?\ssrc=["\'](%s)' % cls._VALID_URL, webpage)]