[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / medici.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     unified_strdate,
7     update_url_query,
8     urlencode_postdata,
9 )
10
11
12 class MediciIE(InfoExtractor):
13     _VALID_URL = r'https?://(?:www\.)?medici\.tv/#!/(?P<id>[^?#&]+)'
14     _TEST = {
15         'url': 'http://www.medici.tv/#!/daniel-harding-frans-helmerson-verbier-festival-music-camp',
16         'md5': '004c21bb0a57248085b6ff3fec72719d',
17         'info_dict': {
18             'id': '3059',
19             'ext': 'flv',
20             'title': 'Daniel Harding conducts the Verbier Festival Music Camp \u2013 With Frans Helmerson',
21             'description': 'md5:322a1e952bafb725174fd8c1a8212f58',
22             'thumbnail': r're:^https?://.*\.jpg$',
23             'upload_date': '20170408',
24         },
25     }
26
27     def _real_extract(self, url):
28         video_id = self._match_id(url)
29
30         # Sets csrftoken cookie
31         self._download_webpage(url, video_id)
32
33         MEDICI_URL = 'http://www.medici.tv/'
34
35         data = self._download_json(
36             MEDICI_URL, video_id,
37             data=urlencode_postdata({
38                 'json': 'true',
39                 'page': '/%s' % video_id,
40                 'timezone_offset': -420,
41             }), headers={
42                 'X-CSRFToken': self._get_cookies(url)['csrftoken'].value,
43                 'X-Requested-With': 'XMLHttpRequest',
44                 'Referer': MEDICI_URL,
45                 'Content-Type': 'application/x-www-form-urlencoded',
46             })
47
48         video = data['video']['videos']['video1']
49
50         title = video.get('nom') or data['title']
51
52         video_id = video.get('id') or video_id
53         formats = self._extract_f4m_formats(
54             update_url_query(video['url_akamai'], {
55                 'hdcore': '3.1.0',
56                 'plugin=aasp': '3.1.0.43.124',
57             }), video_id, f4m_id='hds')
58
59         description = data.get('meta_description')
60         thumbnail = video.get('url_thumbnail') or data.get('main_image')
61         upload_date = unified_strdate(data['video'].get('date'))
62
63         return {
64             'id': video_id,
65             'title': title,
66             'description': description,
67             'thumbnail': thumbnail,
68             'upload_date': upload_date,
69             'formats': formats,
70         }