]>
Commit | Line | Data |
---|---|---|
1 | import re | |
2 | ||
3 | from .common import InfoExtractor | |
4 | from ..compat import ( | |
5 | compat_urllib_parse_urlparse, | |
6 | compat_urlparse, | |
7 | ) | |
8 | from ..utils import ( | |
9 | determine_ext, | |
10 | extract_attributes, | |
11 | int_or_none, | |
12 | js_to_json, | |
13 | mimetype2ext, | |
14 | orderedSet, | |
15 | parse_iso8601, | |
16 | strip_or_none, | |
17 | try_get, | |
18 | ) | |
19 | ||
20 | ||
21 | class CondeNastIE(InfoExtractor): | |
22 | """ | |
23 | Condé Nast is a media group, some of its sites use a custom HTML5 player | |
24 | that works the same in all of them. | |
25 | """ | |
26 | ||
27 | # The keys are the supported sites and the values are the name to be shown | |
28 | # to the user and in the extractor description. | |
29 | _SITES = { | |
30 | 'allure': 'Allure', | |
31 | 'architecturaldigest': 'Architectural Digest', | |
32 | 'arstechnica': 'Ars Technica', | |
33 | 'bonappetit': 'Bon Appétit', | |
34 | 'brides': 'Brides', | |
35 | 'cnevids': 'Condé Nast', | |
36 | 'cntraveler': 'Condé Nast Traveler', | |
37 | 'details': 'Details', | |
38 | 'epicurious': 'Epicurious', | |
39 | 'glamour': 'Glamour', | |
40 | 'golfdigest': 'Golf Digest', | |
41 | 'gq': 'GQ', | |
42 | 'newyorker': 'The New Yorker', | |
43 | 'self': 'SELF', | |
44 | 'teenvogue': 'Teen Vogue', | |
45 | 'vanityfair': 'Vanity Fair', | |
46 | 'vogue': 'Vogue', | |
47 | 'wired': 'WIRED', | |
48 | 'wmagazine': 'W Magazine', | |
49 | } | |
50 | ||
51 | _VALID_URL = r'''(?x)https?://(?:video|www|player(?:-backend)?)\.(?:%s)\.com/ | |
52 | (?: | |
53 | (?: | |
54 | embed(?:js)?| | |
55 | (?:script|inline)/video | |
56 | )/(?P<id>[0-9a-f]{24})(?:/(?P<player_id>[0-9a-f]{24}))?(?:.+?\btarget=(?P<target>[^&]+))?| | |
57 | (?P<type>watch|series|video)/(?P<display_id>[^/?#]+) | |
58 | )''' % '|'.join(_SITES.keys()) | |
59 | IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values())) | |
60 | ||
61 | _EMBED_REGEX = [r'''(?x) | |
62 | <(?:iframe|script)[^>]+?src=(["\'])(?P<url> | |
63 | (?:https?:)?//player(?:-backend)?\.(?:%s)\.com/(?:embed(?:js)?|(?:script|inline)/video)/.+? | |
64 | )\1''' % '|'.join(_SITES.keys())] | |
65 | ||
66 | _TESTS = [{ | |
67 | 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led', | |
68 | 'md5': '1921f713ed48aabd715691f774c451f7', | |
69 | 'info_dict': { | |
70 | 'id': '5171b343c2b4c00dd0c1ccb3', | |
71 | 'ext': 'mp4', | |
72 | 'title': '3D Printed Speakers Lit With LED', | |
73 | '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.', | |
74 | 'uploader': 'wired', | |
75 | 'upload_date': '20130314', | |
76 | 'timestamp': 1363219200, | |
77 | } | |
78 | }, { | |
79 | 'url': 'http://video.gq.com/watch/the-closer-with-keith-olbermann-the-only-true-surprise-trump-s-an-idiot?c=series', | |
80 | 'info_dict': { | |
81 | 'id': '58d1865bfd2e6126e2000015', | |
82 | 'ext': 'mp4', | |
83 | 'title': 'The Only True Surprise? Trump’s an Idiot', | |
84 | 'uploader': 'gq', | |
85 | 'upload_date': '20170321', | |
86 | 'timestamp': 1490126427, | |
87 | 'description': 'How much grimmer would things be if these people were competent?', | |
88 | }, | |
89 | }, { | |
90 | # JS embed | |
91 | 'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js', | |
92 | 'md5': 'f1a6f9cafb7083bab74a710f65d08999', | |
93 | 'info_dict': { | |
94 | 'id': '55f9cf8b61646d1acf00000c', | |
95 | 'ext': 'mp4', | |
96 | 'title': '3D printed TSA Travel Sentry keys really do open TSA locks', | |
97 | 'uploader': 'arstechnica', | |
98 | 'upload_date': '20150916', | |
99 | 'timestamp': 1442434920, | |
100 | } | |
101 | }, { | |
102 | 'url': 'https://player.cnevids.com/inline/video/59138decb57ac36b83000005.js?target=js-cne-player', | |
103 | 'only_matching': True, | |
104 | }, { | |
105 | 'url': 'http://player-backend.cnevids.com/script/video/59138decb57ac36b83000005.js', | |
106 | 'only_matching': True, | |
107 | }] | |
108 | ||
109 | def _extract_series(self, url, webpage): | |
110 | title = self._html_search_regex( | |
111 | r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>', | |
112 | webpage, 'series title') | |
113 | url_object = compat_urllib_parse_urlparse(url) | |
114 | base_url = '%s://%s' % (url_object.scheme, url_object.netloc) | |
115 | m_paths = re.finditer( | |
116 | r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage) | |
117 | paths = orderedSet(m.group(1) for m in m_paths) | |
118 | build_url = lambda path: compat_urlparse.urljoin(base_url, path) | |
119 | entries = [self.url_result(build_url(path), 'CondeNast') for path in paths] | |
120 | return self.playlist_result(entries, playlist_title=title) | |
121 | ||
122 | def _extract_video_params(self, webpage, display_id): | |
123 | query = self._parse_json( | |
124 | self._search_regex( | |
125 | r'(?s)var\s+params\s*=\s*({.+?})[;,]', webpage, 'player params', | |
126 | default='{}'), | |
127 | display_id, transform_source=js_to_json, fatal=False) | |
128 | if query: | |
129 | query['videoId'] = self._search_regex( | |
130 | r'(?:data-video-id=|currentVideoId\s*=\s*)["\']([\da-f]+)', | |
131 | webpage, 'video id', default=None) | |
132 | else: | |
133 | params = extract_attributes(self._search_regex( | |
134 | r'(<[^>]+data-js="video-player"[^>]+>)', | |
135 | webpage, 'player params element')) | |
136 | query.update({ | |
137 | 'videoId': params['data-video'], | |
138 | 'playerId': params['data-player'], | |
139 | 'target': params['id'], | |
140 | }) | |
141 | return query | |
142 | ||
143 | def _extract_video(self, params): | |
144 | video_id = params['videoId'] | |
145 | ||
146 | video_info = None | |
147 | ||
148 | # New API path | |
149 | query = params.copy() | |
150 | query['embedType'] = 'inline' | |
151 | info_page = self._download_json( | |
152 | 'http://player.cnevids.com/embed-api.json', video_id, | |
153 | 'Downloading embed info', fatal=False, query=query) | |
154 | ||
155 | # Old fallbacks | |
156 | if not info_page: | |
157 | if params.get('playerId'): | |
158 | info_page = self._download_json( | |
159 | 'http://player.cnevids.com/player/video.js', video_id, | |
160 | 'Downloading video info', fatal=False, query=params) | |
161 | if info_page: | |
162 | video_info = info_page.get('video') | |
163 | if not video_info: | |
164 | info_page = self._download_webpage( | |
165 | 'http://player.cnevids.com/player/loader.js', | |
166 | video_id, 'Downloading loader info', query=params) | |
167 | if not video_info: | |
168 | info_page = self._download_webpage( | |
169 | 'https://player.cnevids.com/inline/video/%s.js' % video_id, | |
170 | video_id, 'Downloading inline info', query={ | |
171 | 'target': params.get('target', 'embedplayer') | |
172 | }) | |
173 | ||
174 | if not video_info: | |
175 | video_info = self._parse_json( | |
176 | self._search_regex( | |
177 | r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'), | |
178 | video_id, transform_source=js_to_json)['video'] | |
179 | ||
180 | title = video_info['title'] | |
181 | ||
182 | formats = [] | |
183 | for fdata in video_info['sources']: | |
184 | src = fdata.get('src') | |
185 | if not src: | |
186 | continue | |
187 | ext = mimetype2ext(fdata.get('type')) or determine_ext(src) | |
188 | if ext == 'm3u8': | |
189 | formats.extend(self._extract_m3u8_formats( | |
190 | src, video_id, 'mp4', entry_protocol='m3u8_native', | |
191 | m3u8_id='hls', fatal=False)) | |
192 | continue | |
193 | quality = fdata.get('quality') | |
194 | formats.append({ | |
195 | 'format_id': ext + ('-%s' % quality if quality else ''), | |
196 | 'url': src, | |
197 | 'ext': ext, | |
198 | 'quality': 1 if quality == 'high' else 0, | |
199 | }) | |
200 | ||
201 | subtitles = {} | |
202 | for t, caption in video_info.get('captions', {}).items(): | |
203 | caption_url = caption.get('src') | |
204 | if not (t in ('vtt', 'srt', 'tml') and caption_url): | |
205 | continue | |
206 | subtitles.setdefault('en', []).append({'url': caption_url}) | |
207 | ||
208 | return { | |
209 | 'id': video_id, | |
210 | 'formats': formats, | |
211 | 'title': title, | |
212 | 'thumbnail': video_info.get('poster_frame'), | |
213 | 'uploader': video_info.get('brand'), | |
214 | 'duration': int_or_none(video_info.get('duration')), | |
215 | 'tags': video_info.get('tags'), | |
216 | 'series': video_info.get('series_title'), | |
217 | 'season': video_info.get('season_title'), | |
218 | 'timestamp': parse_iso8601(video_info.get('premiere_date')), | |
219 | 'categories': video_info.get('categories'), | |
220 | 'subtitles': subtitles, | |
221 | } | |
222 | ||
223 | def _real_extract(self, url): | |
224 | video_id, player_id, target, url_type, display_id = self._match_valid_url(url).groups() | |
225 | ||
226 | if video_id: | |
227 | return self._extract_video({ | |
228 | 'videoId': video_id, | |
229 | 'playerId': player_id, | |
230 | 'target': target, | |
231 | }) | |
232 | ||
233 | webpage = self._download_webpage(url, display_id) | |
234 | ||
235 | if url_type == 'series': | |
236 | return self._extract_series(url, webpage) | |
237 | else: | |
238 | video = try_get(self._parse_json(self._search_regex( | |
239 | r'__PRELOADED_STATE__\s*=\s*({.+?});', webpage, | |
240 | 'preload state', '{}'), display_id), | |
241 | lambda x: x['transformed']['video']) | |
242 | if video: | |
243 | params = {'videoId': video['id']} | |
244 | info = {'description': strip_or_none(video.get('description'))} | |
245 | else: | |
246 | params = self._extract_video_params(webpage, display_id) | |
247 | info = self._search_json_ld( | |
248 | webpage, display_id, fatal=False) | |
249 | info.update(self._extract_video(params)) | |
250 | return info |