]> Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vk.py
Merge tag 'upstream/2017.02.07'
[youtubedl] / youtube_dl / extractor / vk.py
1 # coding: utf-8
2 from __future__ import unicode_literals
3
4 import collections
5 import re
6 import sys
7
8 from .common import InfoExtractor
9 from ..compat import (
10 compat_str,
11 compat_urlparse,
12 )
13 from ..utils import (
14 clean_html,
15 ExtractorError,
16 get_element_by_class,
17 int_or_none,
18 orderedSet,
19 remove_start,
20 str_to_int,
21 unescapeHTML,
22 unified_timestamp,
23 urlencode_postdata,
24 )
25 from .dailymotion import DailymotionIE
26 from .pladform import PladformIE
27 from .vimeo import VimeoIE
28
29
30 class VKBaseIE(InfoExtractor):
31 _NETRC_MACHINE = 'vk'
32
33 def _login(self):
34 (username, password) = self._get_login_info()
35 if username is None:
36 return
37
38 login_page, url_handle = self._download_webpage_handle(
39 'https://vk.com', None, 'Downloading login page')
40
41 login_form = self._hidden_inputs(login_page)
42
43 login_form.update({
44 'email': username.encode('cp1251'),
45 'pass': password.encode('cp1251'),
46 })
47
48 # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
49 # and expects the first one to be set rather than second (see
50 # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
51 # As of RFC6265 the newer one cookie should be set into cookie store
52 # what actually happens.
53 # We will workaround this VK issue by resetting the remixlhk cookie to
54 # the first one manually.
55 for header, cookies in url_handle.headers.items():
56 if header.lower() != 'set-cookie':
57 continue
58 if sys.version_info[0] >= 3:
59 cookies = cookies.encode('iso-8859-1')
60 cookies = cookies.decode('utf-8')
61 remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
62 if remixlhk:
63 value, domain = remixlhk.groups()
64 self._set_cookie(domain, 'remixlhk', value)
65 break
66
67 login_page = self._download_webpage(
68 'https://login.vk.com/?act=login', None,
69 note='Logging in as %s' % username,
70 data=urlencode_postdata(login_form))
71
72 if re.search(r'onLoginFailed', login_page):
73 raise ExtractorError(
74 'Unable to login, incorrect username and/or password', expected=True)
75
76 def _real_initialize(self):
77 self._login()
78
79
80 class VKIE(VKBaseIE):
81 IE_NAME = 'vk'
82 IE_DESC = 'VK'
83 _VALID_URL = r'''(?x)
84 https?://
85 (?:
86 (?:
87 (?:(?:m|new)\.)?vk\.com/video_|
88 (?:www\.)?daxab.com/
89 )
90 ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
91 (?:
92 (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
93 (?:www\.)?daxab.com/embed/
94 )
95 (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
96 )
97 '''
98 _TESTS = [
99 {
100 'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
101 'md5': '0deae91935c54e00003c2a00646315f0',
102 'info_dict': {
103 'id': '162222515',
104 'ext': 'flv',
105 'title': 'ProtivoGunz - Хуёвая песня',
106 'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
107 'duration': 195,
108 'timestamp': 1329060660,
109 'upload_date': '20120212',
110 'view_count': int,
111 },
112 },
113 {
114 'url': 'http://vk.com/video205387401_165548505',
115 'md5': '6c0aeb2e90396ba97035b9cbde548700',
116 'info_dict': {
117 'id': '165548505',
118 'ext': 'mp4',
119 'uploader': 'Tom Cruise',
120 'title': 'No name',
121 'duration': 9,
122 'timestamp': 1374374880,
123 'upload_date': '20130721',
124 'view_count': int,
125 }
126 },
127 {
128 'note': 'Embedded video',
129 'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
130 'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
131 'info_dict': {
132 'id': '162925554',
133 'ext': 'mp4',
134 'uploader': 'Vladimir Gavrin',
135 'title': 'Lin Dan',
136 'duration': 101,
137 'upload_date': '20120730',
138 'view_count': int,
139 },
140 'skip': 'This video has been removed from public access.',
141 },
142 {
143 # VIDEO NOW REMOVED
144 # please update if you find a video whose URL follows the same pattern
145 'url': 'http://vk.com/video-8871596_164049491',
146 'md5': 'a590bcaf3d543576c9bd162812387666',
147 'note': 'Only available for registered users',
148 'info_dict': {
149 'id': '164049491',
150 'ext': 'mp4',
151 'uploader': 'Триллеры',
152 'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
153 'duration': 8352,
154 'upload_date': '20121218',
155 'view_count': int,
156 },
157 'skip': 'Requires vk account credentials',
158 },
159 {
160 'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
161 'md5': '4d7a5ef8cf114dfa09577e57b2993202',
162 'info_dict': {
163 'id': '168067957',
164 'ext': 'mp4',
165 'uploader': 'Киномания - лучшее из мира кино',
166 'title': ' ',
167 'duration': 7291,
168 'upload_date': '20140328',
169 },
170 'skip': 'Requires vk account credentials',
171 },
172 {
173 'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
174 'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
175 'note': 'ivi.ru embed',
176 'info_dict': {
177 'id': '60690',
178 'ext': 'mp4',
179 'title': 'Книга Илая',
180 'duration': 6771,
181 'upload_date': '20140626',
182 'view_count': int,
183 },
184 'skip': 'Only works from Russia',
185 },
186 {
187 # video (removed?) only available with list id
188 'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
189 'md5': '091287af5402239a1051c37ec7b92913',
190 'info_dict': {
191 'id': '171201961',
192 'ext': 'mp4',
193 'title': 'ТюменцевВВ_09.07.2015',
194 'uploader': 'Anton Ivanov',
195 'duration': 109,
196 'upload_date': '20150709',
197 'view_count': int,
198 },
199 'skip': 'Removed',
200 },
201 {
202 # youtube embed
203 'url': 'https://vk.com/video276849682_170681728',
204 'info_dict': {
205 'id': 'V3K4mi0SYkc',
206 'ext': 'webm',
207 'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
208 'description': 'md5:d9903938abdc74c738af77f527ca0596',
209 'duration': 178,
210 'upload_date': '20130116',
211 'uploader': "Children's Joy Foundation",
212 'uploader_id': 'thecjf',
213 'view_count': int,
214 },
215 },
216 {
217 # dailymotion embed
218 'url': 'https://vk.com/video-37468416_456239855',
219 'info_dict': {
220 'id': 'k3lz2cmXyRuJQSjGHUv',
221 'ext': 'mp4',
222 'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
223 'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
224 'uploader': 'AniLibria.Tv',
225 'upload_date': '20160914',
226 'uploader_id': 'x1p5vl5',
227 'timestamp': 1473877246,
228 },
229 'params': {
230 'skip_download': True,
231 },
232 },
233 {
234 # video key is extra_data not url\d+
235 'url': 'http://vk.com/video-110305615_171782105',
236 'md5': 'e13fcda136f99764872e739d13fac1d1',
237 'info_dict': {
238 'id': '171782105',
239 'ext': 'mp4',
240 'title': 'S-Dance, репетиции к The way show',
241 'uploader': 'THE WAY SHOW | 17 апреля',
242 'timestamp': 1454870100,
243 'upload_date': '20160207',
244 'view_count': int,
245 },
246 },
247 {
248 # finished live stream, postlive_mp4
249 'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
250 'md5': '90d22d051fccbbe9becfccc615be6791',
251 'info_dict': {
252 'id': '456242764',
253 'ext': 'mp4',
254 'title': 'ИгроМир 2016 — день 1',
255 'uploader': 'Игромания',
256 'duration': 5239,
257 'view_count': int,
258 },
259 },
260 {
261 # live stream, hls and rtmp links, most likely already finished live
262 # stream by the time you are reading this comment
263 'url': 'https://vk.com/video-140332_456239111',
264 'only_matching': True,
265 },
266 {
267 # removed video, just testing that we match the pattern
268 'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
269 'only_matching': True,
270 },
271 {
272 # age restricted video, requires vk account credentials
273 'url': 'https://vk.com/video205387401_164765225',
274 'only_matching': True,
275 },
276 {
277 # pladform embed
278 'url': 'https://vk.com/video-76116461_171554880',
279 'only_matching': True,
280 },
281 {
282 'url': 'http://new.vk.com/video205387401_165548505',
283 'only_matching': True,
284 },
285 {
286 # This video is no longer available, because its author has been blocked.
287 'url': 'https://vk.com/video-10639516_456240611',
288 'only_matching': True,
289 }
290 ]
291
292 def _real_extract(self, url):
293 mobj = re.match(self._VALID_URL, url)
294 video_id = mobj.group('videoid')
295
296 if video_id:
297 info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
298 # Some videos (removed?) can only be downloaded with list id specified
299 list_id = mobj.group('list_id')
300 if list_id:
301 info_url += '&list=%s' % list_id
302 else:
303 info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
304 video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
305
306 info_page = self._download_webpage(info_url, video_id)
307
308 error_message = self._html_search_regex(
309 [r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
310 r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
311 info_page, 'error message', default=None)
312 if error_message:
313 raise ExtractorError(error_message, expected=True)
314
315 if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
316 raise ExtractorError(
317 'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
318 expected=True)
319
320 ERRORS = {
321 r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
322 'Video %s has been removed from public access due to rightholder complaint.',
323
324 r'<!>Please log in or <':
325 'Video %s is only available for registered users, '
326 'use --username and --password options to provide account credentials.',
327
328 r'<!>Unknown error':
329 'Video %s does not exist.',
330
331 r'<!>Видео временно недоступно':
332 'Video %s is temporarily unavailable.',
333
334 r'<!>Access denied':
335 'Access denied to video %s.',
336
337 r'<!>Видеозапись недоступна, так как её автор был заблокирован.':
338 'Video %s is no longer available, because its author has been blocked.',
339
340 r'<!>This video is no longer available, because its author has been blocked.':
341 'Video %s is no longer available, because its author has been blocked.',
342 }
343
344 for error_re, error_msg in ERRORS.items():
345 if re.search(error_re, info_page):
346 raise ExtractorError(error_msg % video_id, expected=True)
347
348 youtube_url = self._search_regex(
349 r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
350 info_page, 'youtube iframe', default=None)
351 if youtube_url:
352 return self.url_result(youtube_url, 'Youtube')
353
354 vimeo_url = VimeoIE._extract_url(url, info_page)
355 if vimeo_url is not None:
356 return self.url_result(vimeo_url)
357
358 pladform_url = PladformIE._extract_url(info_page)
359 if pladform_url:
360 return self.url_result(pladform_url)
361
362 m_rutube = re.search(
363 r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
364 if m_rutube is not None:
365 rutube_url = self._proto_relative_url(
366 m_rutube.group(1).replace('\\', ''))
367 return self.url_result(rutube_url)
368
369 dailymotion_urls = DailymotionIE._extract_urls(info_page)
370 if dailymotion_urls:
371 return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
372
373 m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
374 if m_opts:
375 m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
376 if m_opts_url:
377 opts_url = m_opts_url.group(1)
378 if opts_url.startswith('//'):
379 opts_url = 'http:' + opts_url
380 return self.url_result(opts_url)
381
382 # vars does not look to be served anymore since 24.10.2016
383 data = self._parse_json(
384 self._search_regex(
385 r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
386 video_id, fatal=False)
387
388 # <!json> is served instead
389 if not data:
390 data = self._parse_json(
391 self._search_regex(
392 r'<!json>\s*({.+?})\s*<!>', info_page, 'json', default='{}'),
393 video_id)
394 if data:
395 data = data['player']['params'][0]
396
397 if not data:
398 data = self._parse_json(
399 self._search_regex(
400 r'var\s+playerParams\s*=\s*({.+?})\s*;\s*\n', info_page,
401 'player params'),
402 video_id)['params'][0]
403
404 title = unescapeHTML(data['md_title'])
405
406 # 2 = live
407 # 3 = post live (finished live)
408 is_live = data.get('live') == 2
409 if is_live:
410 title = self._live_title(title)
411
412 timestamp = unified_timestamp(self._html_search_regex(
413 r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
414 'upload date', fatal=False))
415
416 view_count = str_to_int(self._search_regex(
417 r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
418 info_page, 'view count', fatal=False))
419
420 formats = []
421 for format_id, format_url in data.items():
422 if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
423 continue
424 if (format_id.startswith(('url', 'cache')) or
425 format_id in ('extra_data', 'live_mp4', 'postlive_mp4')):
426 height = int_or_none(self._search_regex(
427 r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
428 formats.append({
429 'format_id': format_id,
430 'url': format_url,
431 'height': height,
432 })
433 elif format_id == 'hls':
434 formats.extend(self._extract_m3u8_formats(
435 format_url, video_id, 'mp4',
436 entry_protocol='m3u8' if is_live else 'm3u8_native',
437 m3u8_id=format_id, fatal=False, live=is_live))
438 elif format_id == 'rtmp':
439 formats.append({
440 'format_id': format_id,
441 'url': format_url,
442 'ext': 'flv',
443 })
444 self._sort_formats(formats)
445
446 return {
447 'id': compat_str(data.get('vid') or video_id),
448 'formats': formats,
449 'title': title,
450 'thumbnail': data.get('jpg'),
451 'uploader': data.get('md_author'),
452 'duration': data.get('duration'),
453 'timestamp': timestamp,
454 'view_count': view_count,
455 'is_live': is_live,
456 }
457
458
459 class VKUserVideosIE(VKBaseIE):
460 IE_NAME = 'vk:uservideos'
461 IE_DESC = "VK - User's Videos"
462 _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
463 _TEMPLATE_URL = 'https://vk.com/videos'
464 _TESTS = [{
465 'url': 'http://vk.com/videos205387401',
466 'info_dict': {
467 'id': '205387401',
468 'title': "Tom Cruise's Videos",
469 },
470 'playlist_mincount': 4,
471 }, {
472 'url': 'http://vk.com/videos-77521',
473 'only_matching': True,
474 }, {
475 'url': 'http://vk.com/videos-97664626?section=all',
476 'only_matching': True,
477 }, {
478 'url': 'http://m.vk.com/videos205387401',
479 'only_matching': True,
480 }, {
481 'url': 'http://new.vk.com/videos205387401',
482 'only_matching': True,
483 }]
484
485 def _real_extract(self, url):
486 page_id = self._match_id(url)
487
488 webpage = self._download_webpage(url, page_id)
489
490 entries = [
491 self.url_result(
492 'http://vk.com/video' + video_id, 'VK', video_id=video_id)
493 for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
494
495 title = unescapeHTML(self._search_regex(
496 r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
497 webpage, 'title', default=page_id))
498
499 return self.playlist_result(entries, page_id, title)
500
501
502 class VKWallPostIE(VKBaseIE):
503 IE_NAME = 'vk:wallpost'
504 _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
505 _TESTS = [{
506 # public page URL, audio playlist
507 'url': 'https://vk.com/bs.official?w=wall-23538238_35',
508 'info_dict': {
509 'id': '23538238_35',
510 'title': 'Black Shadow - Wall post 23538238_35',
511 'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
512 },
513 'playlist': [{
514 'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
515 'info_dict': {
516 'id': '135220665_111806521',
517 'ext': 'mp3',
518 'title': 'Black Shadow - Слепое Верование',
519 'duration': 370,
520 'uploader': 'Black Shadow',
521 'artist': 'Black Shadow',
522 'track': 'Слепое Верование',
523 },
524 }, {
525 'md5': '4cc7e804579122b17ea95af7834c9233',
526 'info_dict': {
527 'id': '135220665_111802303',
528 'ext': 'mp3',
529 'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
530 'duration': 423,
531 'uploader': 'Black Shadow',
532 'artist': 'Black Shadow',
533 'track': 'Война - Негасимое Бездны Пламя!',
534 },
535 'params': {
536 'skip_download': True,
537 },
538 }],
539 'params': {
540 'usenetrc': True,
541 },
542 'skip': 'Requires vk account credentials',
543 }, {
544 # single YouTube embed, no leading -
545 'url': 'https://vk.com/wall85155021_6319',
546 'info_dict': {
547 'id': '85155021_6319',
548 'title': 'Sergey Gorbunov - Wall post 85155021_6319',
549 },
550 'playlist_count': 1,
551 'params': {
552 'usenetrc': True,
553 },
554 'skip': 'Requires vk account credentials',
555 }, {
556 # wall page URL
557 'url': 'https://vk.com/wall-23538238_35',
558 'only_matching': True,
559 }, {
560 # mobile wall page URL
561 'url': 'https://m.vk.com/wall-23538238_35',
562 'only_matching': True,
563 }]
564
565 def _real_extract(self, url):
566 post_id = self._match_id(url)
567
568 wall_url = 'https://vk.com/wall%s' % post_id
569
570 post_id = remove_start(post_id, '-')
571
572 webpage = self._download_webpage(wall_url, post_id)
573
574 error = self._html_search_regex(
575 r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
576 webpage, 'error', default=None)
577 if error:
578 raise ExtractorError('VK said: %s' % error, expected=True)
579
580 description = clean_html(get_element_by_class('wall_post_text', webpage))
581 uploader = clean_html(get_element_by_class('author', webpage))
582 thumbnail = self._og_search_thumbnail(webpage)
583
584 entries = []
585
586 audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
587 if audio_ids:
588 al_audio = self._download_webpage(
589 'https://vk.com/al_audio.php', post_id,
590 note='Downloading audio info', fatal=False,
591 data=urlencode_postdata({
592 'act': 'reload_audio',
593 'al': '1',
594 'ids': ','.join(audio_ids)
595 }))
596 if al_audio:
597 Audio = collections.namedtuple(
598 'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
599 audios = self._parse_json(
600 self._search_regex(
601 r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
602 post_id, fatal=False, transform_source=unescapeHTML)
603 if isinstance(audios, list):
604 for audio in audios:
605 a = Audio._make(audio[:6])
606 entries.append({
607 'id': '%s_%s' % (a.user_id, a.id),
608 'url': a.url,
609 'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
610 'thumbnail': thumbnail,
611 'duration': a.duration,
612 'uploader': uploader,
613 'artist': a.artist,
614 'track': a.track,
615 })
616
617 for video in re.finditer(
618 r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
619 entries.append(self.url_result(
620 compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
621
622 title = 'Wall post %s' % post_id
623
624 return self.playlist_result(
625 orderedSet(entries), post_id,
626 '%s - %s' % (uploader, title) if uploader else title,
627 description)