2 from __future__ import unicode_literals
6 from .common import InfoExtractor
8 compat_urllib_parse_urlparse,
22 class CondeNastIE(InfoExtractor):
24 Condé Nast is a media group, some of its sites use a custom HTML5 player
25 that works the same in all of them.
28 # The keys are the supported sites and the values are the name to be shown
29 # to the user and in the extractor description.
32 'architecturaldigest': 'Architectural Digest',
33 'arstechnica': 'Ars Technica',
34 'bonappetit': 'Bon Appétit',
36 'cnevids': 'Condé Nast',
37 'cntraveler': 'Condé Nast Traveler',
39 'epicurious': 'Epicurious',
41 'golfdigest': 'Golf Digest',
43 'newyorker': 'The New Yorker',
45 'teenvogue': 'Teen Vogue',
46 'vanityfair': 'Vanity Fair',
49 'wmagazine': 'W Magazine',
52 _VALID_URL = r'https?://(?:video|www|player)\.(?P<site>%s)\.com/(?P<type>watch|series|video|embed(?:js)?)/(?P<id>[^/?#]+)' % '|'.join(_SITES.keys())
53 IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
55 EMBED_URL = r'(?:https?:)?//player\.(?P<site>%s)\.com/(?P<type>embed(?:js)?)/.+?' % '|'.join(_SITES.keys())
58 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
59 'md5': '1921f713ed48aabd715691f774c451f7',
61 'id': '5171b343c2b4c00dd0c1ccb3',
63 'title': '3D Printed Speakers Lit With LED',
64 '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.',
66 'upload_date': '20130314',
67 'timestamp': 1363219200,
71 'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
72 'md5': 'f1a6f9cafb7083bab74a710f65d08999',
74 'id': '55f9cf8b61646d1acf00000c',
76 'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
77 'uploader': 'arstechnica',
78 'upload_date': '20150916',
79 'timestamp': 1442434955,
83 def _extract_series(self, url, webpage):
84 title = self._html_search_regex(
85 r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>',
86 webpage, 'series title')
87 url_object = compat_urllib_parse_urlparse(url)
88 base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
89 m_paths = re.finditer(
90 r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage)
91 paths = orderedSet(m.group(1) for m in m_paths)
92 build_url = lambda path: compat_urlparse.urljoin(base_url, path)
93 entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
94 return self.playlist_result(entries, playlist_title=title)
96 def _extract_video(self, webpage, url_type):
98 params = self._search_regex(
99 r'(?s)var params = {(.+?)}[;,]', webpage, 'player params', default=None)
102 'videoId': self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id'),
103 'playerId': self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id'),
104 'target': self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target'),
107 params = extract_attributes(self._search_regex(
108 r'(<[^>]+data-js="video-player"[^>]+>)',
109 webpage, 'player params element'))
111 'videoId': params['data-video'],
112 'playerId': params['data-player'],
113 'target': params['id'],
115 video_id = query['videoId']
117 info_page = self._download_webpage(
118 'http://player.cnevids.com/player/video.js',
119 video_id, 'Downloading video info', query=query, fatal=False)
121 video_info = self._parse_json(self._search_regex(
122 r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
124 info_page = self._download_webpage(
125 'http://player.cnevids.com/player/loader.js',
126 video_id, 'Downloading loader info', query=query)
127 video_info = self._parse_json(self._search_regex(
128 r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
129 title = video_info['title']
132 for fdata in video_info.get('sources', [{}])[0]:
133 src = fdata.get('src')
136 ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
137 quality = fdata.get('quality')
139 'format_id': ext + ('-%s' % quality if quality else ''),
142 'quality': 1 if quality == 'high' else 0,
144 self._sort_formats(formats)
146 info = self._search_json_ld(
147 webpage, video_id, fatal=False) if url_type != 'embed' else {}
152 'thumbnail': video_info.get('poster_frame'),
153 'uploader': video_info.get('brand'),
154 'duration': int_or_none(video_info.get('duration')),
155 'tags': video_info.get('tags'),
156 'series': video_info.get('series_title'),
157 'season': video_info.get('season_title'),
158 'timestamp': parse_iso8601(video_info.get('premiere_date')),
162 def _real_extract(self, url):
163 site, url_type, item_id = re.match(self._VALID_URL, url).groups()
165 # Convert JS embed to regular embed
166 if url_type == 'embedjs':
167 parsed_url = compat_urlparse.urlparse(url)
168 url = compat_urlparse.urlunparse(parsed_url._replace(
169 path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
172 self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
173 webpage = self._download_webpage(url, item_id)
175 if url_type == 'series':
176 return self._extract_series(url, webpage)
178 return self._extract_video(webpage, url_type)