]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/condenast.py
Fix extraction from youtube.
[youtubedl] / youtube_dl / extractor / condenast.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5
6 from .common import InfoExtractor
7 from ..compat import (
8 compat_urllib_parse_urlparse,
9 compat_urlparse,
10 )
11 from ..utils import (
12 determine_ext,
13 extract_attributes,
14 int_or_none,
15 js_to_json,
16 mimetype2ext,
17 orderedSet,
18 parse_iso8601,
19 )
20
21
22 class CondeNastIE(InfoExtractor):
23 """
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.
26 """
27
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.
30 _SITES = {
31 'allure': 'Allure',
32 'architecturaldigest': 'Architectural Digest',
33 'arstechnica': 'Ars Technica',
34 'bonappetit': 'Bon Appétit',
35 'brides': 'Brides',
36 'cnevids': 'Condé Nast',
37 'cntraveler': 'Condé Nast Traveler',
38 'details': 'Details',
39 'epicurious': 'Epicurious',
40 'glamour': 'Glamour',
41 'golfdigest': 'Golf Digest',
42 'gq': 'GQ',
43 'newyorker': 'The New Yorker',
44 'self': 'SELF',
45 'teenvogue': 'Teen Vogue',
46 'vanityfair': 'Vanity Fair',
47 'vogue': 'Vogue',
48 'wired': 'WIRED',
49 'wmagazine': 'W Magazine',
50 }
51
52 _VALID_URL = r'''(?x)https?://(?:video|www|player(?:-backend)?)\.(?:%s)\.com/
53 (?:
54 (?:
55 embed(?:js)?|
56 (?:script|inline)/video
57 )/(?P<id>[0-9a-f]{24})(?:/(?P<player_id>[0-9a-f]{24}))?(?:.+?\btarget=(?P<target>[^&]+))?|
58 (?P<type>watch|series|video)/(?P<display_id>[^/?#]+)
59 )''' % '|'.join(_SITES.keys())
60 IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
61
62 EMBED_URL = r'(?:https?:)?//player(?:-backend)?\.(?:%s)\.com/(?:embed(?:js)?|(?:script|inline)/video)/.+?' % '|'.join(_SITES.keys())
63
64 _TESTS = [{
65 'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
66 'md5': '1921f713ed48aabd715691f774c451f7',
67 'info_dict': {
68 'id': '5171b343c2b4c00dd0c1ccb3',
69 'ext': 'mp4',
70 'title': '3D Printed Speakers Lit With LED',
71 '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.',
72 'uploader': 'wired',
73 'upload_date': '20130314',
74 'timestamp': 1363219200,
75 }
76 }, {
77 'url': 'http://video.gq.com/watch/the-closer-with-keith-olbermann-the-only-true-surprise-trump-s-an-idiot?c=series',
78 'info_dict': {
79 'id': '58d1865bfd2e6126e2000015',
80 'ext': 'mp4',
81 'title': 'The Only True Surprise? Trump’s an Idiot',
82 'uploader': 'gq',
83 'upload_date': '20170321',
84 'timestamp': 1490126427,
85 },
86 }, {
87 # JS embed
88 'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
89 'md5': 'f1a6f9cafb7083bab74a710f65d08999',
90 'info_dict': {
91 'id': '55f9cf8b61646d1acf00000c',
92 'ext': 'mp4',
93 'title': '3D printed TSA Travel Sentry keys really do open TSA locks',
94 'uploader': 'arstechnica',
95 'upload_date': '20150916',
96 'timestamp': 1442434955,
97 }
98 }, {
99 'url': 'https://player.cnevids.com/inline/video/59138decb57ac36b83000005.js?target=js-cne-player',
100 'only_matching': True,
101 }, {
102 'url': 'http://player-backend.cnevids.com/script/video/59138decb57ac36b83000005.js',
103 'only_matching': True,
104 }]
105
106 def _extract_series(self, url, webpage):
107 title = self._html_search_regex(
108 r'(?s)<div class="cne-series-info">.*?<h1>(.+?)</h1>',
109 webpage, 'series title')
110 url_object = compat_urllib_parse_urlparse(url)
111 base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
112 m_paths = re.finditer(
113 r'(?s)<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]', webpage)
114 paths = orderedSet(m.group(1) for m in m_paths)
115 build_url = lambda path: compat_urlparse.urljoin(base_url, path)
116 entries = [self.url_result(build_url(path), 'CondeNast') for path in paths]
117 return self.playlist_result(entries, playlist_title=title)
118
119 def _extract_video_params(self, webpage, display_id):
120 query = self._parse_json(
121 self._search_regex(
122 r'(?s)var\s+params\s*=\s*({.+?})[;,]', webpage, 'player params',
123 default='{}'),
124 display_id, transform_source=js_to_json, fatal=False)
125 if query:
126 query['videoId'] = self._search_regex(
127 r'(?:data-video-id=|currentVideoId\s*=\s*)["\']([\da-f]+)',
128 webpage, 'video id', default=None)
129 else:
130 params = extract_attributes(self._search_regex(
131 r'(<[^>]+data-js="video-player"[^>]+>)',
132 webpage, 'player params element'))
133 query.update({
134 'videoId': params['data-video'],
135 'playerId': params['data-player'],
136 'target': params['id'],
137 })
138 return query
139
140 def _extract_video(self, params):
141 video_id = params['videoId']
142
143 video_info = None
144
145 # New API path
146 query = params.copy()
147 query['embedType'] = 'inline'
148 info_page = self._download_json(
149 'http://player.cnevids.com/embed-api.json', video_id,
150 'Downloading embed info', fatal=False, query=query)
151
152 # Old fallbacks
153 if not info_page:
154 if params.get('playerId'):
155 info_page = self._download_json(
156 'http://player.cnevids.com/player/video.js', video_id,
157 'Downloading video info', fatal=False, query=params)
158 if info_page:
159 video_info = info_page.get('video')
160 if not video_info:
161 info_page = self._download_webpage(
162 'http://player.cnevids.com/player/loader.js',
163 video_id, 'Downloading loader info', query=params)
164 if not video_info:
165 info_page = self._download_webpage(
166 'https://player.cnevids.com/inline/video/%s.js' % video_id,
167 video_id, 'Downloading inline info', query={
168 'target': params.get('target', 'embedplayer')
169 })
170
171 if not video_info:
172 video_info = self._parse_json(
173 self._search_regex(
174 r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'),
175 video_id, transform_source=js_to_json)['video']
176
177 title = video_info['title']
178
179 formats = []
180 for fdata in video_info['sources']:
181 src = fdata.get('src')
182 if not src:
183 continue
184 ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
185 if ext == 'm3u8':
186 formats.extend(self._extract_m3u8_formats(
187 src, video_id, 'mp4', entry_protocol='m3u8_native',
188 m3u8_id='hls', fatal=False))
189 continue
190 quality = fdata.get('quality')
191 formats.append({
192 'format_id': ext + ('-%s' % quality if quality else ''),
193 'url': src,
194 'ext': ext,
195 'quality': 1 if quality == 'high' else 0,
196 })
197 self._sort_formats(formats)
198
199 return {
200 'id': video_id,
201 'formats': formats,
202 'title': title,
203 'thumbnail': video_info.get('poster_frame'),
204 'uploader': video_info.get('brand'),
205 'duration': int_or_none(video_info.get('duration')),
206 'tags': video_info.get('tags'),
207 'series': video_info.get('series_title'),
208 'season': video_info.get('season_title'),
209 'timestamp': parse_iso8601(video_info.get('premiere_date')),
210 'categories': video_info.get('categories'),
211 }
212
213 def _real_extract(self, url):
214 video_id, player_id, target, url_type, display_id = re.match(self._VALID_URL, url).groups()
215
216 if video_id:
217 return self._extract_video({
218 'videoId': video_id,
219 'playerId': player_id,
220 'target': target,
221 })
222
223 webpage = self._download_webpage(url, display_id)
224
225 if url_type == 'series':
226 return self._extract_series(url, webpage)
227 else:
228 params = self._extract_video_params(webpage, display_id)
229 info = self._search_json_ld(
230 webpage, display_id, fatal=False)
231 info.update(self._extract_video(params))
232 return info