]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py
d99a42a9f1dd982b50cc2b324714d6a75cb15d9b
[youtubedl] / youtube_dl / extractor / vk.py
1 # encoding: utf-8
2 from __future__ import unicode_literals
3
4 import re
5 import json
6
7 from .common import InfoExtractor
8 from ..compat import (
9 compat_str,
10 compat_urllib_parse,
11 )
12 from ..utils import (
13 ExtractorError,
14 orderedSet,
15 sanitized_Request,
16 str_to_int,
17 unescapeHTML,
18 unified_strdate,
19 )
20 from .vimeo import VimeoIE
21
22
23 class VKIE(InfoExtractor):
24 IE_NAME = 'vk'
25 IE_DESC = 'VK'
26 _VALID_URL = r'''(?x)
27 https?://
28 (?:
29 (?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|
30 (?:
31 (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
32 (?:www\.)?biqle\.ru/watch/
33 )
34 (?P<videoid>[^s].*?)(?:\?(?:.*\blist=(?P<list_id>[\da-f]+))?|%2F|$)
35 )
36 '''
37 _NETRC_MACHINE = 'vk'
38
39 _TESTS = [
40 {
41 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
42 'md5': '0deae91935c54e00003c2a00646315f0',
43 'info_dict': {
44 'id': '162222515',
45 'ext': 'flv',
46 'title': 'ProtivoGunz - Хуёвая песня',
47 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
48 'duration': 195,
49 'upload_date': '20120212',
50 'view_count': int,
51 },
52 },
53 {
54 'url': 'http://vk.com/video205387401_165548505',
55 'md5': '6c0aeb2e90396ba97035b9cbde548700',
56 'info_dict': {
57 'id': '165548505',
58 'ext': 'mp4',
59 'uploader': 'Tom Cruise',
60 'title': 'No name',
61 'duration': 9,
62 'upload_date': '20130721',
63 'view_count': int,
64 }
65 },
66 {
67 'note': 'Embedded video',
68 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
69 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
70 'info_dict': {
71 'id': '162925554',
72 'ext': 'mp4',
73 'uploader': 'Vladimir Gavrin',
74 'title': 'Lin Dan',
75 'duration': 101,
76 'upload_date': '20120730',
77 'view_count': int,
78 }
79 },
80 {
81 # VIDEO NOW REMOVED
82 # please update if you find a video whose URL follows the same pattern
83 'url': 'http://vk.com/video-8871596_164049491',
84 'md5': 'a590bcaf3d543576c9bd162812387666',
85 'note': 'Only available for registered users',
86 'info_dict': {
87 'id': '164049491',
88 'ext': 'mp4',
89 'uploader': 'Триллеры',
90 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
91 'duration': 8352,
92 'upload_date': '20121218',
93 'view_count': int,
94 },
95 'skip': 'Requires vk account credentials',
96 },
97 {
98 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
99 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
100 'info_dict': {
101 'id': '168067957',
102 'ext': 'mp4',
103 'uploader': 'Киномания - лучшее из мира кино',
104 'title': ' ',
105 'duration': 7291,
106 'upload_date': '20140328',
107 },
108 'skip': 'Requires vk account credentials',
109 },
110 {
111 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
112 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
113 'note': 'ivi.ru embed',
114 'info_dict': {
115 'id': '60690',
116 'ext': 'mp4',
117 'title': 'Книга Илая',
118 'duration': 6771,
119 'upload_date': '20140626',
120 'view_count': int,
121 },
122 'skip': 'Only works from Russia',
123 },
124 {
125 # video (removed?) only available with list id
126 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
127 'md5': '091287af5402239a1051c37ec7b92913',
128 'info_dict': {
129 'id': '171201961',
130 'ext': 'mp4',
131 'title': 'ТюменцевВВ_09.07.2015',
132 'uploader': 'Anton Ivanov',
133 'duration': 109,
134 'upload_date': '20150709',
135 'view_count': int,
136 },
137 },
138 {
139 # youtube embed
140 'url': 'https://vk.com/video276849682_170681728',
141 'info_dict': {
142 'id': 'V3K4mi0SYkc',
143 'ext': 'mp4',
144 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
145 'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
146 'duration': 179,
147 'upload_date': '20130116',
148 'uploader': "Children's Joy Foundation",
149 'uploader_id': 'thecjf',
150 'view_count': int,
151 },
152 },
153 {
154 # removed video, just testing that we match the pattern
155 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
156 'only_matching': True,
157 },
158 {
159 # age restricted video, requires vk account credentials
160 'url': 'https://vk.com/video205387401_164765225',
161 'only_matching': True,
162 },
163 {
164 # vk wrapper
165 'url': 'http://www.biqle.ru/watch/847655_160197695',
166 'only_matching': True,
167 }
168 ]
169
170 def _login(self):
171 (username, password) = self._get_login_info()
172 if username is None:
173 return
174
175 login_page = self._download_webpage(
176 'https://vk.com', None, 'Downloading login page')
177
178 login_form = self._hidden_inputs(login_page)
179
180 login_form.update({
181 'email': username.encode('cp1251'),
182 'pass': password.encode('cp1251'),
183 })
184
185 request = sanitized_Request(
186 'https://login.vk.com/?act=login',
187 compat_urllib_parse.urlencode(login_form).encode('utf-8'))
188 login_page = self._download_webpage(
189 request, None, note='Logging in as %s' % username)
190
191 if re.search(r'onLoginFailed', login_page):
192 raise ExtractorError(
193 'Unable to login, incorrect username and/or password', expected=True)
194
195 def _real_initialize(self):
196 self._login()
197
198 def _real_extract(self, url):
199 mobj = re.match(self._VALID_URL, url)
200 video_id = mobj.group('videoid')
201
202 if not video_id:
203 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
204
205 info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
206
207 # Some videos (removed?) can only be downloaded with list id specified
208 list_id = mobj.group('list_id')
209 if list_id:
210 info_url += '&list=%s' % list_id
211
212 info_page = self._download_webpage(info_url, video_id)
213
214 error_message = self._html_search_regex(
215 r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
216 info_page, 'error message', default=None)
217 if error_message:
218 raise ExtractorError(error_message, expected=True)
219
220 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
221 raise ExtractorError(
222 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
223 expected=True)
224
225 ERRORS = {
226 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
227 'Video %s has been removed from public access due to rightholder complaint.',
228
229 r'<!>Please log in or <':
230 'Video %s is only available for registered users, '
231 'use --username and --password options to provide account credentials.',
232
233 r'<!>Unknown error':
234 'Video %s does not exist.',
235
236 r'<!>Видео временно недоступно':
237 'Video %s is temporarily unavailable.',
238
239 r'<!>Access denied':
240 'Access denied to video %s.',
241 }
242
243 for error_re, error_msg in ERRORS.items():
244 if re.search(error_re, info_page):
245 raise ExtractorError(error_msg % video_id, expected=True)
246
247 youtube_url = self._search_regex(
248 r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
249 info_page, 'youtube iframe', default=None)
250 if youtube_url:
251 return self.url_result(youtube_url, 'Youtube')
252
253 vimeo_url = VimeoIE._extract_vimeo_url(url, info_page)
254 if vimeo_url is not None:
255 return self.url_result(vimeo_url)
256
257 m_rutube = re.search(
258 r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
259 if m_rutube is not None:
260 self.to_screen('rutube video detected')
261 rutube_url = self._proto_relative_url(
262 m_rutube.group(1).replace('\\', ''))
263 return self.url_result(rutube_url)
264
265 m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
266 if m_opts:
267 m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
268 if m_opts_url:
269 opts_url = m_opts_url.group(1)
270 if opts_url.startswith('//'):
271 opts_url = 'http:' + opts_url
272 return self.url_result(opts_url)
273
274 data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
275 data = json.loads(data_json)
276
277 # Extract upload date
278 upload_date = None
279 mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
280 if mobj is not None:
281 mobj.group(1) + ' ' + mobj.group(2)
282 upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
283
284 view_count = None
285 views = self._html_search_regex(
286 r'"mv_views_count_number"[^>]*>(.+?\bviews?)<',
287 info_page, 'view count', fatal=False)
288 if views:
289 view_count = str_to_int(self._search_regex(
290 r'([\d,.]+)', views, 'view count', fatal=False))
291
292 formats = [{
293 'format_id': k,
294 'url': v,
295 'width': int(k[len('url'):]),
296 } for k, v in data.items()
297 if k.startswith('url')]
298 self._sort_formats(formats)
299
300 return {
301 'id': compat_str(data['vid']),
302 'formats': formats,
303 'title': unescapeHTML(data['md_title']),
304 'thumbnail': data.get('jpg'),
305 'uploader': data.get('md_author'),
306 'duration': data.get('duration'),
307 'upload_date': upload_date,
308 'view_count': view_count,
309 }
310
311
312 class VKUserVideosIE(InfoExtractor):
313 IE_NAME = 'vk:uservideos'
314 IE_DESC = "VK - User's Videos"
315 _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)$'
316 _TEMPLATE_URL = 'https://vk.com/videos'
317 _TESTS = [{
318 'url': 'http://vk.com/videos205387401',
319 'info_dict': {
320 'id': '205387401',
321 'title': "Tom Cruise's Videos",
322 },
323 'playlist_mincount': 4,
324 }, {
325 'url': 'http://vk.com/videos-77521',
326 'only_matching': True,
327 }]
328
329 def _real_extract(self, url):
330 page_id = self._match_id(url)
331
332 webpage = self._download_webpage(url, page_id)
333
334 entries = [
335 self.url_result(
336 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
337 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
338
339 title = unescapeHTML(self._search_regex(
340 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
341 webpage, 'title', default=page_id))
342
343 return self.playlist_result(entries, page_id, title)