youtube-dl/youtube_dl/extractor/vk.py

602 lines
22 KiB
Python
Raw Normal View History

2016-10-02 11:39:18 +00:00
# coding: utf-8
2014-01-21 16:32:03 +00:00
from __future__ import unicode_literals
2016-08-17 23:14:05 +00:00
import collections
import re
import sys
from .common import InfoExtractor
2016-07-13 14:51:44 +00:00
from ..compat import (
compat_str,
compat_urlparse,
)
from ..utils import (
2016-07-13 14:51:44 +00:00
clean_html,
2014-02-16 19:05:15 +00:00
ExtractorError,
2016-07-13 14:51:44 +00:00
get_element_by_class,
int_or_none,
orderedSet,
2016-07-13 14:51:44 +00:00
remove_start,
2015-06-15 14:55:25 +00:00
str_to_int,
unescapeHTML,
2016-09-29 16:48:21 +00:00
unified_timestamp,
urlencode_postdata,
)
from .dailymotion import DailymotionIE
from .pladform import PladformIE
from .vimeo import VimeoIE
2016-07-13 14:51:44 +00:00
class VKBaseIE(InfoExtractor):
_NETRC_MACHINE = 'vk'
def _login(self):
(username, password) = self._get_login_info()
if username is None:
return
login_page, url_handle = self._download_webpage_handle(
'https://vk.com', None, 'Downloading login page')
login_form = self._hidden_inputs(login_page)
login_form.update({
'email': username.encode('cp1251'),
'pass': password.encode('cp1251'),
})
# https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
# and expects the first one to be set rather than second (see
# https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
# As of RFC6265 the newer one cookie should be set into cookie store
# what actually happens.
# We will workaround this VK issue by resetting the remixlhk cookie to
# the first one manually.
2016-08-17 22:22:23 +00:00
for header, cookies in url_handle.headers.items():
if header.lower() != 'set-cookie':
continue
2016-07-13 14:52:52 +00:00
if sys.version_info[0] >= 3:
cookies = cookies.encode('iso-8859-1')
cookies = cookies.decode('utf-8')
remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
if remixlhk:
value, domain = remixlhk.groups()
self._set_cookie(domain, 'remixlhk', value)
2016-08-17 22:22:23 +00:00
break
2016-07-13 14:51:44 +00:00
login_page = self._download_webpage(
'https://login.vk.com/?act=login', None,
note='Logging in as %s' % username,
data=urlencode_postdata(login_form))
if re.search(r'onLoginFailed', login_page):
raise ExtractorError(
'Unable to login, incorrect username and/or password', expected=True)
def _real_initialize(self):
self._login()
class VKIE(VKBaseIE):
2015-07-18 11:23:33 +00:00
IE_NAME = 'vk'
IE_DESC = 'VK'
_VALID_URL = r'''(?x)
https?://
(?:
2016-05-06 14:02:40 +00:00
(?:
(?:(?:m|new)\.)?vk\.com/video_|
2016-05-06 14:02:40 +00:00
(?:www\.)?daxab.com/
)
ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
(?:
(?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
2016-05-06 14:02:40 +00:00
(?:www\.)?daxab.com/embed/
)
2016-05-06 14:02:40 +00:00
(?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
)
'''
2014-02-16 19:05:15 +00:00
_TESTS = [
{
'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
'md5': '0deae91935c54e00003c2a00646315f0',
'info_dict': {
'id': '162222515',
'ext': 'flv',
'title': 'ProtivoGunz - Хуёвая песня',
2015-03-01 21:30:18 +00:00
'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
2014-02-16 19:05:15 +00:00
'duration': 195,
2016-09-29 16:48:21 +00:00
'timestamp': 1329060660,
'upload_date': '20120212',
2015-06-15 14:55:25 +00:00
'view_count': int,
2014-02-16 19:05:15 +00:00
},
},
2014-02-16 19:05:15 +00:00
{
'url': 'http://vk.com/video205387401_165548505',
'md5': '6c0aeb2e90396ba97035b9cbde548700',
2014-02-16 19:05:15 +00:00
'info_dict': {
'id': '165548505',
2014-02-16 19:05:15 +00:00
'ext': 'mp4',
'uploader': 'Tom Cruise',
'title': 'No name',
'duration': 9,
2016-09-29 16:48:21 +00:00
'timestamp': 1374374880,
2015-06-15 14:55:25 +00:00
'upload_date': '20130721',
'view_count': int,
2014-02-16 19:05:15 +00:00
}
},
{
'note': 'Embedded video',
'url': 'http://vk.com/video_ext.php?oid=32194266&id=162925554&hash=7d8c2e0d5e05aeaa&hd=1',
'md5': 'c7ce8f1f87bec05b3de07fdeafe21a0a',
'info_dict': {
'id': '162925554',
'ext': 'mp4',
'uploader': 'Vladimir Gavrin',
'title': 'Lin Dan',
'duration': 101,
'upload_date': '20120730',
2015-06-15 14:55:25 +00:00
'view_count': int,
2016-05-06 14:02:40 +00:00
},
'skip': 'This video has been removed from public access.',
},
2014-02-16 19:05:15 +00:00
{
# VIDEO NOW REMOVED
# please update if you find a video whose URL follows the same pattern
2014-02-16 19:05:15 +00:00
'url': 'http://vk.com/video-8871596_164049491',
'md5': 'a590bcaf3d543576c9bd162812387666',
'note': 'Only available for registered users',
'info_dict': {
'id': '164049491',
'ext': 'mp4',
'uploader': 'Триллеры',
'title': '► Бойцовский клуб / Fight Club 1999 [HD 720]',
2014-02-16 19:05:15 +00:00
'duration': 8352,
2015-06-15 14:55:25 +00:00
'upload_date': '20121218',
'view_count': int,
2014-02-16 19:05:15 +00:00
},
'skip': 'Requires vk account credentials',
},
{
'url': 'http://vk.com/hd_kino_mania?z=video-43215063_168067957%2F15c66b9b533119788d',
'md5': '4d7a5ef8cf114dfa09577e57b2993202',
'info_dict': {
'id': '168067957',
'ext': 'mp4',
'uploader': 'Киномания - лучшее из мира кино',
'title': ' ',
'duration': 7291,
'upload_date': '20140328',
},
'skip': 'Requires vk account credentials',
},
2014-06-29 13:07:59 +00:00
{
'url': 'http://m.vk.com/video-43215063_169084319?list=125c627d1aa1cebb83&from=wall-43215063_2566540',
'md5': '0c45586baa71b7cb1d0784ee3f4e00a6',
'note': 'ivi.ru embed',
'info_dict': {
'id': '60690',
'ext': 'mp4',
'title': 'Книга Илая',
'duration': 6771,
'upload_date': '20140626',
2015-06-15 14:55:25 +00:00
'view_count': int,
2014-06-29 13:07:59 +00:00
},
2014-06-29 20:16:31 +00:00
'skip': 'Only works from Russia',
2014-06-29 13:07:59 +00:00
},
2015-07-11 15:23:49 +00:00
{
# video (removed?) only available with list id
'url': 'https://vk.com/video30481095_171201961?list=8764ae2d21f14088d4',
'md5': '091287af5402239a1051c37ec7b92913',
'info_dict': {
'id': '171201961',
'ext': 'mp4',
'title': 'ТюменцевВВ_09.07.2015',
'uploader': 'Anton Ivanov',
'duration': 109,
'upload_date': '20150709',
'view_count': int,
},
2016-09-29 16:48:21 +00:00
'skip': 'Removed',
2015-07-11 15:23:49 +00:00
},
2015-07-08 14:41:08 +00:00
{
# youtube embed
'url': 'https://vk.com/video276849682_170681728',
'info_dict': {
'id': 'V3K4mi0SYkc',
'ext': 'webm',
2015-07-08 14:41:08 +00:00
'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
2016-05-06 14:02:40 +00:00
'description': 'md5:d9903938abdc74c738af77f527ca0596',
'duration': 178,
2015-07-08 14:41:08 +00:00
'upload_date': '20130116',
'uploader': "Children's Joy Foundation",
'uploader_id': 'thecjf',
'view_count': int,
},
},
{
# dailymotion embed
'url': 'https://vk.com/video-37468416_456239855',
'info_dict': {
'id': 'k3lz2cmXyRuJQSjGHUv',
'ext': 'mp4',
'title': 'md5:d52606645c20b0ddbb21655adaa4f56f',
'description': 'md5:c651358f03c56f1150b555c26d90a0fd',
'uploader': 'AniLibria.Tv',
'upload_date': '20160914',
'uploader_id': 'x1p5vl5',
'timestamp': 1473877246,
},
'params': {
'skip_download': True,
},
},
{
# video key is extra_data not url\d+
'url': 'http://vk.com/video-110305615_171782105',
'md5': 'e13fcda136f99764872e739d13fac1d1',
'info_dict': {
'id': '171782105',
'ext': 'mp4',
'title': 'S-Dance, репетиции к The way show',
'uploader': 'THE WAY SHOW | 17 апреля',
2016-09-29 16:48:21 +00:00
'timestamp': 1454870100,
'upload_date': '20160207',
'view_count': int,
},
},
{
# finished live stream, live_mp4
'url': 'https://vk.com/videos-387766?z=video-387766_456242764%2Fpl_-387766_-2',
'md5': '90d22d051fccbbe9becfccc615be6791',
'info_dict': {
'id': '456242764',
'ext': 'mp4',
'title': 'ИгроМир 2016 — день 1',
'uploader': 'Игромания',
'duration': 5239,
2016-09-29 16:51:52 +00:00
'view_count': int,
},
},
{
# live stream, hls and rtmp links,most likely already finished live
# stream by the time you are reading this comment
'url': 'https://vk.com/video-140332_456239111',
'only_matching': True,
},
2014-11-23 09:11:04 +00:00
{
# removed video, just testing that we match the pattern
'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
'only_matching': True,
},
{
# age restricted video, requires vk account credentials
'url': 'https://vk.com/video205387401_164765225',
'only_matching': True,
},
2015-12-07 16:05:54 +00:00
{
# pladform embed
'url': 'https://vk.com/video-76116461_171554880',
'only_matching': True,
},
{
'url': 'http://new.vk.com/video205387401_165548505',
'only_matching': True,
}
2014-02-16 19:05:15 +00:00
]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('videoid')
2016-05-06 14:02:40 +00:00
if video_id:
info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
# Some videos (removed?) can only be downloaded with list id specified
list_id = mobj.group('list_id')
if list_id:
info_url += '&list=%s' % list_id
else:
info_url = 'http://vk.com/video_ext.php?' + mobj.group('embed_query')
video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
2014-02-16 19:05:15 +00:00
info_page = self._download_webpage(info_url, video_id)
2014-02-16 19:05:15 +00:00
2015-07-18 13:15:20 +00:00
error_message = self._html_search_regex(
2016-05-06 14:02:40 +00:00
[r'(?s)<!><div[^>]+class="video_layer_message"[^>]*>(.+?)</div>',
r'(?s)<div[^>]+id="video_ext_msg"[^>]*>(.+?)</div>'],
2015-07-18 13:15:20 +00:00
info_page, 'error message', default=None)
if error_message:
raise ExtractorError(error_message, expected=True)
if re.search(r'<!>/login\.php\?.*\bact=security_check', info_page):
raise ExtractorError(
'You are trying to log in from an unusual location. You should confirm ownership at vk.com to log in with this IP.',
expected=True)
2014-10-28 14:35:25 +00:00
ERRORS = {
r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
2014-11-23 21:14:27 +00:00
'Video %s has been removed from public access due to rightholder complaint.',
2014-10-28 14:35:25 +00:00
r'<!>Please log in or <':
2014-11-23 21:14:27 +00:00
'Video %s is only available for registered users, '
'use --username and --password options to provide account credentials.',
r'<!>Unknown error':
'Video %s does not exist.',
r'<!>Видео временно недоступно':
'Video %s is temporarily unavailable.',
2015-07-11 15:26:03 +00:00
r'<!>Access denied':
'Access denied to video %s.',
2014-10-28 14:35:25 +00:00
}
2014-02-16 19:05:15 +00:00
2014-10-28 14:35:25 +00:00
for error_re, error_msg in ERRORS.items():
if re.search(error_re, info_page):
raise ExtractorError(error_msg % video_id, expected=True)
2014-10-28 14:06:07 +00:00
2015-07-08 14:34:50 +00:00
youtube_url = self._search_regex(
r'<iframe[^>]+src="((?:https?:)?//www.youtube.com/embed/[^"]+)"',
info_page, 'youtube iframe', default=None)
if youtube_url:
2015-07-08 14:41:08 +00:00
return self.url_result(youtube_url, 'Youtube')
2014-06-29 13:07:59 +00:00
vimeo_url = VimeoIE._extract_url(url, info_page)
if vimeo_url is not None:
return self.url_result(vimeo_url)
pladform_url = PladformIE._extract_url(info_page)
if pladform_url:
return self.url_result(pladform_url)
m_rutube = re.search(
r'\ssrc="((?:https?:)?//rutube\.ru\\?/(?:video|play)\\?/embed(?:.*?))\\?"', info_page)
if m_rutube is not None:
rutube_url = self._proto_relative_url(
m_rutube.group(1).replace('\\', ''))
return self.url_result(rutube_url)
dailymotion_urls = DailymotionIE._extract_urls(info_page)
if dailymotion_urls:
return self.url_result(dailymotion_urls[0], DailymotionIE.ie_key())
2015-06-15 14:46:10 +00:00
m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
2014-06-29 13:07:59 +00:00
if m_opts:
2015-06-15 14:46:10 +00:00
m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
2014-06-29 13:07:59 +00:00
if m_opts_url:
opts_url = m_opts_url.group(1)
if opts_url.startswith('//'):
opts_url = 'http:' + opts_url
return self.url_result(opts_url)
2016-10-25 14:05:29 +00:00
# vars does not look to be served anymore since 24.10.2016
data = self._parse_json(
self._search_regex(
r'var\s+vars\s*=\s*({.+?});', info_page, 'vars', default='{}'),
video_id, fatal=False)
# <!json> is served instead
if not data:
data = self._parse_json(
self._search_regex(
r'<!json>\s*({.+?})\s*<!>', info_page, 'json'),
video_id)['player']['params'][0]
title = unescapeHTML(data['md_title'])
if data.get('live') == 2:
title = self._live_title(title)
2016-09-29 16:48:21 +00:00
timestamp = unified_timestamp(self._html_search_regex(
2016-09-29 16:51:52 +00:00
r'class=["\']mv_info_date[^>]+>([^<]+)(?:<|from)', info_page,
2016-09-29 16:48:21 +00:00
'upload date', fatal=False))
2016-09-29 16:51:52 +00:00
view_count = str_to_int(self._search_regex(
r'class=["\']mv_views_count[^>]+>\s*([\d,.]+)',
info_page, 'view count', fatal=False))
2015-06-15 14:55:25 +00:00
formats = []
for format_id, format_url in data.items():
if not isinstance(format_url, compat_str) or not format_url.startswith(('http', '//', 'rtmp')):
continue
if format_id.startswith(('url', 'cache')) or format_id in ('extra_data', 'live_mp4'):
height = int_or_none(self._search_regex(
r'^(?:url|cache)(\d+)', format_id, 'height', default=None))
formats.append({
'format_id': format_id,
'url': format_url,
'height': height,
})
elif format_id == 'hls':
formats.extend(self._extract_m3u8_formats(
format_url, video_id, 'mp4', m3u8_id=format_id,
fatal=False, live=True))
elif format_id == 'rtmp':
formats.append({
'format_id': format_id,
'url': format_url,
'ext': 'flv',
})
self._sort_formats(formats)
return {
'id': compat_str(data.get('vid') or video_id),
'formats': formats,
'title': title,
'thumbnail': data.get('jpg'),
'uploader': data.get('md_author'),
'duration': data.get('duration'),
2016-09-29 16:48:21 +00:00
'timestamp': timestamp,
2015-06-15 14:55:25 +00:00
'view_count': view_count,
}
2016-07-13 14:51:44 +00:00
class VKUserVideosIE(VKBaseIE):
2015-07-18 11:23:33 +00:00
IE_NAME = 'vk:uservideos'
IE_DESC = "VK - User's Videos"
_VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
_TEMPLATE_URL = 'https://vk.com/videos'
2015-07-18 11:22:25 +00:00
_TESTS = [{
'url': 'http://vk.com/videos205387401',
2015-02-17 23:33:41 +00:00
'info_dict': {
'id': '205387401',
2015-07-18 11:22:25 +00:00
'title': "Tom Cruise's Videos",
2015-02-17 23:33:41 +00:00
},
'playlist_mincount': 4,
2015-07-18 11:22:25 +00:00
}, {
'url': 'http://vk.com/videos-77521',
'only_matching': True,
}, {
'url': 'http://vk.com/videos-97664626?section=all',
'only_matching': True,
}, {
'url': 'http://m.vk.com/videos205387401',
'only_matching': True,
}, {
'url': 'http://new.vk.com/videos205387401',
'only_matching': True,
2015-07-18 11:22:25 +00:00
}]
def _real_extract(self, url):
page_id = self._match_id(url)
2015-07-18 11:22:25 +00:00
webpage = self._download_webpage(url, page_id)
entries = [
2014-11-23 09:11:52 +00:00
self.url_result(
'http://vk.com/video' + video_id, 'VK', video_id=video_id)
2015-07-18 12:05:08 +00:00
for video_id in orderedSet(re.findall(r'href="/video(-?[0-9_]+)"', webpage))]
2015-07-18 11:22:25 +00:00
title = unescapeHTML(self._search_regex(
r'<title>\s*([^<]+?)\s+\|\s+\d+\s+videos',
webpage, 'title', default=page_id))
return self.playlist_result(entries, page_id, title)
2016-07-13 14:51:44 +00:00
class VKWallPostIE(VKBaseIE):
IE_NAME = 'vk:wallpost'
_VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
_TESTS = [{
# public page URL, audio playlist
'url': 'https://vk.com/bs.official?w=wall-23538238_35',
'info_dict': {
'id': '23538238_35',
'title': 'Black Shadow - Wall post 23538238_35',
'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
},
'playlist': [{
'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
'info_dict': {
'id': '135220665_111806521',
'ext': 'mp3',
'title': 'Black Shadow - Слепое Верование',
'duration': 370,
'uploader': 'Black Shadow',
'artist': 'Black Shadow',
'track': 'Слепое Верование',
},
}, {
'md5': '4cc7e804579122b17ea95af7834c9233',
'info_dict': {
'id': '135220665_111802303',
'ext': 'mp3',
'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
'duration': 423,
'uploader': 'Black Shadow',
'artist': 'Black Shadow',
'track': 'Война - Негасимое Бездны Пламя!',
},
'params': {
'skip_download': True,
},
}],
2016-08-17 23:14:05 +00:00
'params': {
'usenetrc': True,
},
2016-07-13 14:51:44 +00:00
'skip': 'Requires vk account credentials',
}, {
# single YouTube embed, no leading -
'url': 'https://vk.com/wall85155021_6319',
'info_dict': {
'id': '85155021_6319',
'title': 'Sergey Gorbunov - Wall post 85155021_6319',
},
'playlist_count': 1,
2016-08-17 23:14:05 +00:00
'params': {
'usenetrc': True,
},
2016-07-13 14:51:44 +00:00
'skip': 'Requires vk account credentials',
}, {
# wall page URL
'url': 'https://vk.com/wall-23538238_35',
'only_matching': True,
}, {
# mobile wall page URL
'url': 'https://m.vk.com/wall-23538238_35',
'only_matching': True,
}]
def _real_extract(self, url):
post_id = self._match_id(url)
wall_url = 'https://vk.com/wall%s' % post_id
post_id = remove_start(post_id, '-')
webpage = self._download_webpage(wall_url, post_id)
error = self._html_search_regex(
r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
webpage, 'error', default=None)
if error:
raise ExtractorError('VK said: %s' % error, expected=True)
description = clean_html(get_element_by_class('wall_post_text', webpage))
2016-08-17 23:14:05 +00:00
uploader = clean_html(get_element_by_class('author', webpage))
2016-07-13 14:51:44 +00:00
thumbnail = self._og_search_thumbnail(webpage)
entries = []
2016-08-17 23:14:05 +00:00
audio_ids = re.findall(r'data-full-id=["\'](\d+_\d+)', webpage)
if audio_ids:
al_audio = self._download_webpage(
'https://vk.com/al_audio.php', post_id,
note='Downloading audio info', fatal=False,
data=urlencode_postdata({
'act': 'reload_audio',
'al': '1',
'ids': ','.join(audio_ids)
}))
if al_audio:
Audio = collections.namedtuple(
'Audio', ['id', 'user_id', 'url', 'track', 'artist', 'duration'])
audios = self._parse_json(
self._search_regex(
r'<!json>(.+?)<!>', al_audio, 'audios', default='[]'),
post_id, fatal=False, transform_source=unescapeHTML)
if isinstance(audios, list):
for audio in audios:
a = Audio._make(audio[:6])
entries.append({
'id': '%s_%s' % (a.user_id, a.id),
'url': a.url,
'title': '%s - %s' % (a.artist, a.track) if a.artist and a.track else a.id,
'thumbnail': thumbnail,
'duration': a.duration,
'uploader': uploader,
'artist': a.artist,
'track': a.track,
})
2016-07-13 14:51:44 +00:00
for video in re.finditer(
r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
entries.append(self.url_result(
compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
title = 'Wall post %s' % post_id
return self.playlist_result(
orderedSet(entries), post_id,
'%s - %s' % (uploader, title) if uploader else title,
description)