]>
Commit | Line | Data |
---|---|---|
e1fb2456 | 1 | # coding: utf-8 |
6e249060 | 2 | from __future__ import unicode_literals |
e1fb2456 JMF |
3 | |
4 | import re | |
5 | import json | |
6 | ||
7 | from .common import InfoExtractor | |
1cc79574 | 8 | from ..compat import ( |
e1fb2456 | 9 | compat_urllib_parse, |
e1fb2456 JMF |
10 | compat_urllib_parse_urlparse, |
11 | compat_urlparse, | |
12 | ) | |
1cc79574 PH |
13 | from ..utils import ( |
14 | orderedSet, | |
15 | ) | |
e1fb2456 JMF |
16 | |
17 | ||
18 | class CondeNastIE(InfoExtractor): | |
19 | """ | |
20 | Condé Nast is a media group, some of its sites use a custom HTML5 player | |
21 | that works the same in all of them. | |
22 | """ | |
23 | ||
24 | # The keys are the supported sites and the values are the name to be shown | |
25 | # to the user and in the extractor description. | |
6e249060 PH |
26 | _SITES = { |
27 | 'wired': 'WIRED', | |
28 | 'gq': 'GQ', | |
29 | 'vogue': 'Vogue', | |
30 | 'glamour': 'Glamour', | |
31 | 'wmagazine': 'W Magazine', | |
32 | 'vanityfair': 'Vanity Fair', | |
fa35cdad | 33 | 'cnevids': 'Condé Nast', |
6e249060 | 34 | } |
e1fb2456 | 35 | |
fa35cdad | 36 | _VALID_URL = r'http://(video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys()) |
6e249060 | 37 | IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) |
e1fb2456 | 38 | |
1419fafd S |
39 | EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed)/.+?' % '|'.join(_SITES.keys()) |
40 | ||
e1fb2456 | 41 | _TEST = { |
6e249060 | 42 | 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', |
6e249060 PH |
43 | 'md5': '1921f713ed48aabd715691f774c451f7', |
44 | 'info_dict': { | |
fa35cdad PH |
45 | 'id': '5171b343c2b4c00dd0c1ccb3', |
46 | 'ext': 'mp4', | |
6e249060 PH |
47 | 'title': '3D Printed Speakers Lit With LED', |
48 | 'description': 'Check out these beautiful 3D printed LED speakers. You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.', | |
e1fb2456 JMF |
49 | } |
50 | } | |
51 | ||
52 | def _extract_series(self, url, webpage): | |
53 | title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>', | |
6e249060 | 54 | webpage, 'series title', flags=re.DOTALL) |
e1fb2456 JMF |
55 | url_object = compat_urllib_parse_urlparse(url) |
56 | base_url = '%s://%s' % (url_object.scheme, url_object.netloc) | |
57 | m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', | |
58 | webpage, flags=re.DOTALL) | |
59 | paths = orderedSet(m.group(1) for m in m_paths) | |
60 | build_url = lambda path: compat_urlparse.urljoin(base_url, path) | |
61 | entries = [self.url_result(build_url(path), 'CondeNast') for path in paths] | |
62 | return self.playlist_result(entries, playlist_title=title) | |
63 | ||
fa35cdad PH |
64 | def _extract_video(self, webpage, url_type): |
65 | if url_type != 'embed': | |
66 | description = self._html_search_regex( | |
67 | [ | |
68 | r'<div class="cne-video-description">(.+?)</div>', | |
69 | r'<div class="video-post-content">(.+?)</div>', | |
70 | ], | |
71 | webpage, 'description', fatal=False, flags=re.DOTALL) | |
72 | else: | |
73 | description = None | |
e1fb2456 | 74 | params = self._search_regex(r'var params = {(.+?)}[;,]', webpage, |
6e249060 PH |
75 | 'player params', flags=re.DOTALL) |
76 | video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id') | |
77 | player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id') | |
78 | target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target') | |
e1fb2456 JMF |
79 | data = compat_urllib_parse.urlencode({'videoId': video_id, |
80 | 'playerId': player_id, | |
81 | 'target': target, | |
82 | }) | |
83 | base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]', | |
6e249060 | 84 | webpage, 'base info url', |
e1fb2456 JMF |
85 | default='http://player.cnevids.com/player/loader.js?') |
86 | info_url = base_info_url + data | |
87 | info_page = self._download_webpage(info_url, video_id, | |
6e249060 PH |
88 | 'Downloading video info') |
89 | video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info') | |
e1fb2456 JMF |
90 | video_info = json.loads(video_info) |
91 | ||
d0da491e PH |
92 | formats = [{ |
93 | 'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']), | |
94 | 'url': fdata['src'], | |
95 | 'ext': fdata['type'].split('/')[-1], | |
96 | 'quality': 1 if fdata['quality'] == 'high' else 0, | |
97 | } for fdata in video_info['sources'][0]] | |
98 | self._sort_formats(formats) | |
e1fb2456 | 99 | |
d0da491e PH |
100 | return { |
101 | 'id': video_id, | |
102 | 'formats': formats, | |
103 | 'title': video_info['title'], | |
104 | 'thumbnail': video_info['poster_frame'], | |
105 | 'description': description, | |
106 | } | |
e1fb2456 JMF |
107 | |
108 | def _real_extract(self, url): | |
109 | mobj = re.match(self._VALID_URL, url) | |
110 | site = mobj.group('site') | |
111 | url_type = mobj.group('type') | |
fa35cdad | 112 | item_id = mobj.group('id') |
e1fb2456 | 113 | |
fa35cdad PH |
114 | self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site]) |
115 | webpage = self._download_webpage(url, item_id) | |
e1fb2456 JMF |
116 | |
117 | if url_type == 'series': | |
118 | return self._extract_series(url, webpage) | |
119 | else: | |
fa35cdad | 120 | return self._extract_video(webpage, url_type) |