[telegraaf] Fix extractor (closes #9318)
[youtube-dl] / youtube_dl / extractor / telegraaf.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 from .common import InfoExtractor
5 from ..utils import (
6     determine_ext,
7     remove_end,
8 )
9
10
11 class TelegraafIE(InfoExtractor):
12     _VALID_URL = r'https?://(?:www\.)?telegraaf\.nl/tv/(?:[^/]+/)+(?P<id>\d+)/[^/]+\.html'
13     _TEST = {
14         'url': 'http://www.telegraaf.nl/tv/nieuws/binnenland/24353229/__Tikibad_ontruimd_wegens_brand__.html',
15         'info_dict': {
16             'id': '24353229',
17             'ext': 'mp4',
18             'title': 'Tikibad ontruimd wegens brand',
19             'description': 'md5:05ca046ff47b931f9b04855015e163a4',
20             'thumbnail': 're:^https?://.*\.jpg$',
21             'duration': 33,
22         },
23         'params': {
24             # m3u8 download
25             'skip_download': True,
26         },
27     }
28
29     def _real_extract(self, url):
30         video_id = self._match_id(url)
31
32         webpage = self._download_webpage(url, video_id)
33
34         player_url = self._html_search_regex(
35             r'<iframe[^>]+src="([^"]+")', webpage, 'player URL')
36         player_page = self._download_webpage(
37             player_url, video_id, note='Download player webpage')
38         playlist_url = self._search_regex(
39             r'playlist\s*:\s*"([^"]+)"', player_page, 'playlist URL')
40         playlist_data = self._download_json(playlist_url, video_id)
41
42         item = playlist_data['items'][0]
43         formats = []
44         locations = item['locations']
45         for location in locations.get('adaptive', []):
46             manifest_url = location['src']
47             ext = determine_ext(manifest_url)
48             if ext == 'm3u8':
49                 formats.extend(self._extract_m3u8_formats(
50                     manifest_url, video_id, ext='mp4', m3u8_id='hls'))
51             elif ext == 'mpd':
52                 # TODO: Current DASH formats are broken - $Time$ pattern in
53                 # <SegmentTemplate> not implemented yet
54                 continue
55             else:
56                 self.report_warning('Unknown adaptive format %s' % ext)
57         for location in locations.get('progressive', []):
58             formats.append({
59                 'url': location['sources'][0]['src'],
60                 'width': location.get('width'),
61                 'height': location.get('height'),
62                 'format_id': 'http-%s' % location['label'],
63             })
64
65         self._sort_formats(formats)
66
67         title = remove_end(self._og_search_title(webpage), ' - VIDEO')
68         description = self._og_search_description(webpage)
69         duration = item.get('duration')
70         thumbnail = item.get('poster')
71
72         return {
73             'id': video_id,
74             'title': title,
75             'description': description,
76             'formats': formats,
77             'duration': duration,
78             'thumbnail': thumbnail,
79         }