yt-dlp/youtube_dl/extractor/vk.py

291 lines
10 KiB
Python
Raw Normal View History

# encoding: utf-8
2014-01-21 11:32:03 -05:00
from __future__ import unicode_literals
import re
import json
from .common import InfoExtractor
from ..compat import (
compat_str,
compat_urllib_parse,
compat_urllib_request,
)
from ..utils import (
2014-02-16 14:05:15 -05:00
ExtractorError,
orderedSet,
2015-06-15 10:55:25 -04:00
str_to_int,
unescapeHTML,
unified_strdate,
)
class VKIE(InfoExtractor):
2014-01-21 11:32:03 -05:00
IE_NAME = 'vk.com'
_VALID_URL = r'''(?x)
https?://
(?:
(?:m\.)?vk\.com/video_ext\.php\?.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+)|
(?:
(?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
(?:www\.)?biqle\.ru/watch/
)
(?P<videoid>[^s].*?)(?:\?|%2F|$)
)
'''
_NETRC_MACHINE = 'vk'
2014-02-16 14:05:15 -05:00
_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 16:30:18 -05:00
'uploader': 're:(?:Noize MC|Alexander Ilyashenko).*',
2014-02-16 14:05:15 -05:00
'duration': 195,
'upload_date': '20120212',
2015-06-15 10:55:25 -04:00
'view_count': int,
2014-02-16 14:05:15 -05:00
},
},
2014-02-16 14:05:15 -05:00
{
'url': 'http://vk.com/video205387401_165548505',
'md5': '6c0aeb2e90396ba97035b9cbde548700',
2014-02-16 14:05:15 -05:00
'info_dict': {
'id': '165548505',
2014-02-16 14:05:15 -05:00
'ext': 'mp4',
'uploader': 'Tom Cruise',
'title': 'No name',
'duration': 9,
2015-06-15 10:55:25 -04:00
'upload_date': '20130721',
'view_count': int,
2014-02-16 14:05:15 -05:00
}
},
{
'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 10:55:25 -04:00
'view_count': int,
}
},
2014-02-16 14:05:15 -05:00
{
# VIDEO NOW REMOVED
# please update if you find a video whose URL follows the same pattern
2014-02-16 14:05:15 -05:00
'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 14:05:15 -05:00
'duration': 8352,
2015-06-15 10:55:25 -04:00
'upload_date': '20121218',
'view_count': int,
2014-02-16 14:05:15 -05:00
},
'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 09:07:59 -04:00
{
'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 10:55:25 -04:00
'view_count': int,
2014-06-29 09:07:59 -04:00
},
2014-06-29 16:16:31 -04:00
'skip': 'Only works from Russia',
2014-06-29 09:07:59 -04:00
},
2015-07-08 10:41:08 -04:00
{
# youtube embed
'url': 'https://vk.com/video276849682_170681728',
'info_dict': {
'id': 'V3K4mi0SYkc',
'ext': 'mp4',
'title': "DSWD Awards 'Children's Joy Foundation, Inc.' Certificate of Registration and License to Operate",
'description': 'md5:bf9c26cfa4acdfb146362682edd3827a',
'duration': 179,
'upload_date': '20130116',
'uploader': "Children's Joy Foundation",
'uploader_id': 'thecjf',
'view_count': int,
},
},
2014-11-23 04:11:04 -05:00
{
# removed video, just testing that we match the pattern
'url': 'http://vk.com/feed?z=video-43215063_166094326%2Fbb50cacd3177146d7a',
'only_matching': True,
},
{
# vk wrapper
'url': 'http://www.biqle.ru/watch/847655_160197695',
'only_matching': True,
}
2014-02-16 14:05:15 -05:00
]
def _login(self):
(username, password) = self._get_login_info()
if username is None:
return
2015-06-27 04:04:55 -04:00
login_page = self._download_webpage(
'https://vk.com', None, 'Downloading login page')
login_form = self._form_hidden_inputs(login_page)
2015-06-27 04:04:55 -04:00
login_form.update({
'email': username.encode('cp1251'),
'pass': password.encode('cp1251'),
2015-06-27 04:04:55 -04:00
})
2014-02-16 14:05:15 -05:00
2015-06-27 04:04:55 -04:00
request = compat_urllib_request.Request(
'https://login.vk.com/?act=login',
compat_urllib_parse.urlencode(login_form).encode('utf-8'))
login_page = self._download_webpage(
request, None, note='Logging in as %s' % username)
2014-02-16 14:05:15 -05:00
if re.search(r'onLoginFailed', login_page):
2015-06-27 04:04:55 -04:00
raise ExtractorError(
'Unable to login, incorrect username and/or password', expected=True)
2014-02-16 14:05:15 -05:00
def _real_initialize(self):
self._login()
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('videoid')
if not video_id:
video_id = '%s_%s' % (mobj.group('oid'), mobj.group('id'))
2014-02-16 14:05:15 -05:00
2015-07-06 14:02:34 -04:00
info_url = 'https://vk.com/al_video.php?act=show&al=1&module=video&video=%s' % video_id
info_page = self._download_webpage(info_url, video_id)
2014-02-16 14:05:15 -05:00
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 10:35:25 -04:00
ERRORS = {
r'>Видеозапись .*? была изъята из публичного доступа в связи с обращением правообладателя.<':
2014-11-23 16:14:27 -05:00
'Video %s has been removed from public access due to rightholder complaint.',
2014-10-28 10:35:25 -04:00
r'<!>Please log in or <':
2014-11-23 16:14:27 -05:00
'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.',
2014-10-28 10:35:25 -04:00
}
2014-02-16 14:05:15 -05:00
2014-10-28 10:35:25 -04:00
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 10:06:07 -04:00
2015-07-08 10:34:50 -04:00
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 10:41:08 -04:00
return self.url_result(youtube_url, 'Youtube')
2014-06-29 09:07:59 -04:00
m_rutube = re.search(
r'\ssrc="((?:https?:)?//rutube\.ru\\?/video\\?/embed(?:.*?))\\?"', info_page)
if m_rutube is not None:
self.to_screen('rutube video detected')
rutube_url = self._proto_relative_url(
m_rutube.group(1).replace('\\', ''))
return self.url_result(rutube_url)
2015-06-15 10:46:10 -04:00
m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page)
2014-06-29 09:07:59 -04:00
if m_opts:
2015-06-15 10:46:10 -04:00
m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1))
2014-06-29 09:07:59 -04:00
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)
2015-06-15 10:46:10 -04:00
data_json = self._search_regex(r'var\s+vars\s*=\s*({.+?});', info_page, 'vars')
2014-01-21 11:29:04 -05:00
data = json.loads(data_json)
# Extract upload date
upload_date = None
2015-06-15 10:47:01 -04:00
mobj = re.search(r'id="mv_date(?:_views)?_wrap"[^>]*>([a-zA-Z]+ [0-9]+), ([0-9]+) at', info_page)
if mobj is not None:
2014-11-23 15:39:15 -05:00
mobj.group(1) + ' ' + mobj.group(2)
upload_date = unified_strdate(mobj.group(1) + ' ' + mobj.group(2))
2015-06-15 10:55:25 -04:00
view_count = str_to_int(self._search_regex(
r'"mv_views_count_number"[^>]*>([\d,.]+) views<',
info_page, 'view count', fatal=False))
formats = [{
'format_id': k,
'url': v,
'width': int(k[len('url'):]),
} for k, v in data.items()
if k.startswith('url')]
self._sort_formats(formats)
return {
2014-01-21 11:29:04 -05:00
'id': compat_str(data['vid']),
'formats': formats,
2014-01-21 11:29:04 -05:00
'title': unescapeHTML(data['md_title']),
'thumbnail': data.get('jpg'),
'uploader': data.get('md_author'),
'duration': data.get('duration'),
'upload_date': upload_date,
2015-06-15 10:55:25 -04:00
'view_count': view_count,
}
class VKUserVideosIE(InfoExtractor):
IE_NAME = 'vk.com:user-videos'
IE_DESC = 'vk.com:All of a user\'s videos'
_VALID_URL = r'https?://vk\.com/videos(?P<id>[0-9]+)(?:m\?.*)?'
_TEMPLATE_URL = 'https://vk.com/videos'
_TEST = {
'url': 'http://vk.com/videos205387401',
2015-02-17 18:33:41 -05:00
'info_dict': {
'id': '205387401',
},
'playlist_mincount': 4,
}
def _real_extract(self, url):
page_id = self._match_id(url)
page = self._download_webpage(url, page_id)
2014-11-23 04:11:52 -05:00
video_ids = orderedSet(
m.group(1) for m in re.finditer(r'href="/video([0-9_]+)"', page))
url_entries = [
self.url_result(
'http://vk.com/video' + video_id, 'VK', video_id=video_id)
for video_id in video_ids]
return self.playlist_result(url_entries, page_id)