]> Raphaƫl G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nfb.py
ea077254b4320fe18e59eb9b67461b13c146b873
[youtubedl] / youtube_dl / extractor / nfb.py
1 from __future__ import unicode_literals
2
3 from .common import InfoExtractor
4 from ..compat import (
5 compat_urllib_request,
6 compat_urllib_parse,
7 )
8
9
10 class NFBIE(InfoExtractor):
11 IE_NAME = 'nfb'
12 IE_DESC = 'National Film Board of Canada'
13 _VALID_URL = r'https?://(?:www\.)?(?:nfb|onf)\.ca/film/(?P<id>[\da-z_-]+)'
14
15 _TEST = {
16 'url': 'https://www.nfb.ca/film/qallunaat_why_white_people_are_funny',
17 'info_dict': {
18 'id': 'qallunaat_why_white_people_are_funny',
19 'ext': 'mp4',
20 'title': 'Qallunaat! Why White People Are Funny ',
21 'description': 'md5:836d8aff55e087d04d9f6df554d4e038',
22 'duration': 3128,
23 'uploader': 'Mark Sandiford',
24 'uploader_id': 'mark-sandiford',
25 },
26 'params': {
27 # rtmp download
28 'skip_download': True,
29 }
30 }
31
32 def _real_extract(self, url):
33 video_id = self._match_id(url)
34 page = self._download_webpage(
35 'https://www.nfb.ca/film/%s' % video_id, video_id,
36 'Downloading film page')
37
38 uploader_id = self._html_search_regex(r'<a class="director-link" href="/explore-all-directors/([^/]+)/"',
39 page, 'director id', fatal=False)
40 uploader = self._html_search_regex(r'<em class="director-name" itemprop="name">([^<]+)</em>',
41 page, 'director name', fatal=False)
42
43 request = compat_urllib_request.Request('https://www.nfb.ca/film/%s/player_config' % video_id,
44 compat_urllib_parse.urlencode({'getConfig': 'true'}).encode('ascii'))
45 request.add_header('Content-Type', 'application/x-www-form-urlencoded')
46 request.add_header('X-NFB-Referer', 'http://www.nfb.ca/medias/flash/NFBVideoPlayer.swf')
47
48 config = self._download_xml(request, video_id, 'Downloading player config XML')
49
50 title = None
51 description = None
52 thumbnail = None
53 duration = None
54 formats = []
55
56 def extract_thumbnail(media):
57 thumbnails = {}
58 for asset in media.findall('assets/asset'):
59 thumbnails[asset.get('quality')] = asset.find('default/url').text
60 if not thumbnails:
61 return None
62 if 'high' in thumbnails:
63 return thumbnails['high']
64 return list(thumbnails.values())[0]
65
66 for media in config.findall('./player/stream/media'):
67 if media.get('type') == 'posterImage':
68 thumbnail = extract_thumbnail(media)
69 elif media.get('type') == 'video':
70 duration = int(media.get('duration'))
71 title = media.find('title').text
72 description = media.find('description').text
73 # It seems assets always go from lower to better quality, so no need to sort
74 for asset in media.findall('assets/asset'):
75 for x in asset:
76 formats.append({
77 'url': x.find('streamerURI').text,
78 'app': x.find('streamerURI').text.split('/', 3)[3],
79 'play_path': x.find('url').text,
80 'rtmp_live': False,
81 'ext': 'mp4',
82 'format_id': '%s-%s' % (x.tag, asset.get('quality')),
83 })
84
85 return {
86 'id': video_id,
87 'title': title,
88 'description': description,
89 'thumbnail': thumbnail,
90 'duration': duration,
91 'uploader': uploader,
92 'uploader_id': uploader_id,
93 'formats': formats,
94 }