youtube-dl/youtube_dl/extractor/youku.py

281 lines
9.0 KiB
Python
Raw Normal View History

2013-06-27 11:32:02 -07:00
# coding: utf-8
2014-09-25 00:58:09 -07:00
from __future__ import unicode_literals
import json
2015-05-28 02:00:09 -07:00
import base64
2013-06-23 13:01:02 -07:00
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse,
compat_cookiejar,
compat_cookies,
compat_urllib_request,
compat_ord,
Switch codebase to use sanitized_Request instead of compat_urllib_request.Request [downloader/dash] Use sanitized_Request [downloader/http] Use sanitized_Request [atresplayer] Use sanitized_Request [bambuser] Use sanitized_Request [bliptv] Use sanitized_Request [brightcove] Use sanitized_Request [cbs] Use sanitized_Request [ceskatelevize] Use sanitized_Request [collegerama] Use sanitized_Request [extractor/common] Use sanitized_Request [crunchyroll] Use sanitized_Request [dailymotion] Use sanitized_Request [dcn] Use sanitized_Request [dramafever] Use sanitized_Request [dumpert] Use sanitized_Request [eitb] Use sanitized_Request [escapist] Use sanitized_Request [everyonesmixtape] Use sanitized_Request [extremetube] Use sanitized_Request [facebook] Use sanitized_Request [fc2] Use sanitized_Request [flickr] Use sanitized_Request [4tube] Use sanitized_Request [gdcvault] Use sanitized_Request [extractor/generic] Use sanitized_Request [hearthisat] Use sanitized_Request [hotnewhiphop] Use sanitized_Request [hypem] Use sanitized_Request [iprima] Use sanitized_Request [ivi] Use sanitized_Request [keezmovies] Use sanitized_Request [letv] Use sanitized_Request [lynda] Use sanitized_Request [metacafe] Use sanitized_Request [minhateca] Use sanitized_Request [miomio] Use sanitized_Request [meovideo] Use sanitized_Request [mofosex] Use sanitized_Request [moniker] Use sanitized_Request [mooshare] Use sanitized_Request [movieclips] Use sanitized_Request [mtv] Use sanitized_Request [myvideo] Use sanitized_Request [neteasemusic] Use sanitized_Request [nfb] Use sanitized_Request [niconico] Use sanitized_Request [noco] Use sanitized_Request [nosvideo] Use sanitized_Request [novamov] Use sanitized_Request [nowness] Use sanitized_Request [nuvid] Use sanitized_Request [played] Use sanitized_Request [pluralsight] Use sanitized_Request [pornhub] Use sanitized_Request [pornotube] Use sanitized_Request [primesharetv] Use sanitized_Request [promptfile] Use sanitized_Request [qqmusic] Use sanitized_Request [rtve] Use sanitized_Request [safari] Use sanitized_Request [sandia] Use sanitized_Request [shared] Use sanitized_Request [sharesix] Use sanitized_Request [sina] Use sanitized_Request [smotri] Use sanitized_Request [sohu] Use sanitized_Request [spankwire] Use sanitized_Request [sportdeutschland] Use sanitized_Request [streamcloud] Use sanitized_Request [streamcz] Use sanitized_Request [tapely] Use sanitized_Request [tube8] Use sanitized_Request [tubitv] Use sanitized_Request [twitch] Use sanitized_Request [twitter] Use sanitized_Request [udemy] Use sanitized_Request [vbox7] Use sanitized_Request [veoh] Use sanitized_Request [vessel] Use sanitized_Request [vevo] Use sanitized_Request [viddler] Use sanitized_Request [videomega] Use sanitized_Request [viewvster] Use sanitized_Request [viki] Use sanitized_Request [vk] Use sanitized_Request [vodlocker] Use sanitized_Request [voicerepublic] Use sanitized_Request [wistia] Use sanitized_Request [xfileshare] Use sanitized_Request [xtube] Use sanitized_Request [xvideos] Use sanitized_Request [yandexmusic] Use sanitized_Request [youku] Use sanitized_Request [youporn] Use sanitized_Request [youtube] Use sanitized_Request [patreon] Use sanitized_Request [extractor/common] Remove unused import [nfb] PEP 8
2015-11-21 08:18:17 -08:00
)
from ..utils import (
ExtractorError,
sanitized_Request,
)
2015-06-15 07:41:24 -07:00
2013-06-23 13:01:02 -07:00
class YoukuIE(InfoExtractor):
2015-05-28 02:00:09 -07:00
IE_NAME = 'youku'
2015-07-10 09:58:13 -07:00
IE_DESC = '优酷'
2014-09-25 00:58:09 -07:00
_VALID_URL = r'''(?x)
(?:
http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
youku:)
(?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
'''
2015-05-28 02:00:09 -07:00
2015-06-15 08:36:28 -07:00
_TESTS = [{
2015-06-15 07:41:24 -07:00
'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
'md5': '5f3af4192eabacc4501508d54a8cabd7',
'info_dict': {
'id': 'XMTc1ODE5Njcy_part1',
2015-06-15 07:41:24 -07:00
'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
'ext': 'flv'
}
2015-06-15 08:36:28 -07:00
}, {
'url': 'http://player.youku.com/player.php/sid/XNDgyMDQ2NTQw/v.swf',
'only_matching': True,
}, {
'url': 'http://v.youku.com/v_show/id_XODgxNjg1Mzk2_ev_1.html',
'info_dict': {
'id': 'XODgxNjg1Mzk2',
'title': '武媚娘传奇 85',
},
'playlist_count': 11,
}, {
'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
'info_dict': {
'id': 'XMTI1OTczNDM5Mg',
'title': '花千骨 04',
},
'playlist_count': 13,
'skip': 'Available in China only',
2015-08-31 09:19:58 -07:00
}, {
'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
'note': 'Video protected with password',
'info_dict': {
'id': 'XNjA1NzA2Njgw',
2015-09-01 09:26:17 -07:00
'title': '邢義田复旦讲座之想象中的胡人—从“左衽孔子”说起',
2015-08-31 09:19:58 -07:00
},
2015-09-01 09:28:03 -07:00
'playlist_count': 19,
2015-08-31 09:19:58 -07:00
'params': {
'videopassword': '100600',
},
2015-06-15 08:36:28 -07:00
}]
2013-06-27 11:25:46 -07:00
2015-05-28 02:00:09 -07:00
def construct_video_urls(self, data1, data2):
# get sid, token
def yk_t(s1, s2):
ls = list(range(256))
t = 0
for i in range(256):
t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
2015-05-28 02:00:09 -07:00
ls[i], ls[t] = ls[t], ls[i]
s = bytearray()
x, y = 0, 0
2015-05-28 02:00:09 -07:00
for i in range(len(s2)):
y = (y + 1) % 256
x = (x + ls[y]) % 256
ls[x], ls[y] = ls[y], ls[x]
s.append(compat_ord(s2[i]) ^ ls[(ls[x] + ls[y]) % 256])
return bytes(s)
2015-05-28 02:00:09 -07:00
sid, token = yk_t(
b'becaf9be', base64.b64decode(data2['security']['encrypt_string'].encode('ascii'))
).decode('ascii').split('_')
2015-05-28 02:00:09 -07:00
# get oip
oip = data2['security']['ip']
2015-05-28 02:00:09 -07:00
# get fileid
string_ls = list(
'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
shuffled_string_ls = []
seed = data1.get('seed')
seed = 0
2015-05-28 02:00:09 -07:00
N = len(string_ls)
for ii in range(N):
seed = (seed * 0xd3 + 0x754f) % 0x10000
idx = seed * len(string_ls) // 0x10000
shuffled_string_ls.append(string_ls[idx])
del string_ls[idx]
fileid_dict = {}
for stream in data1['stream']:
format = stream.get('stream_type')
#streamfileid = [
# int(i) for i in data1['stream']['streamfileids']#[format].strip('*').split('*')]
#fileid = ''.join(
# [shuffled_string_ls[i] for i in streamfileid])
fileid = stream['stream_fileid']
fileid_dict[format] = fileid
#fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
2015-05-28 02:00:09 -07:00
def get_fileid(format, n):
#fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
fileid = fileid_dict[format]
2015-05-28 02:00:09 -07:00
return fileid
# get ep
def generate_ep(format, n):
fileid = get_fileid(format, n)
ep_t = yk_t(
b'bf7e5f01',
('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
)
ep = base64.b64encode(ep_t).decode('ascii')
2015-05-28 02:00:09 -07:00
return ep
# generate video_urls
video_urls_dict = {}
for stream in data1['stream']:
format = stream.get('stream_type')
2015-05-28 02:00:09 -07:00
video_urls = []
for dt in stream['segs']:
n = str(int(dt['size']))
param = {
'K': dt['key'],
'hd': self.get_hd(format),
'myp': 0,
'ts': dt['total_milliseconds_video'],
'ypp': 0,
'ctype': 12,
'ev': 1,
'token': token,
'oip': oip,
'ep': generate_ep(format, n)
}
2015-05-28 02:00:09 -07:00
video_url = \
'http://k.youku.com/player/getFlvPath/' + \
'sid/' + sid + \
2015-06-15 07:41:24 -07:00
'_' + str(int(n) + 1).zfill(2) + \
2015-05-28 02:00:09 -07:00
'/st/' + self.parse_ext_l(format) + \
2015-06-15 07:41:24 -07:00
'/fileid/' + get_fileid(format, n) + '?' + \
compat_urllib_parse.urlencode(param)
2015-05-28 02:00:09 -07:00
video_urls.append(video_url)
video_urls_dict[format] = video_urls
return video_urls_dict
def get_hd(self, fm):
hd_id_dict = {
2015-06-15 07:41:24 -07:00
'flv': '0',
'mp4': '1',
'hd2': '2',
'hd3': '3',
'3gp': '0',
'3gphd': '1',
'flvhd': '0'
2015-05-28 02:00:09 -07:00
}
return hd_id_dict[fm]
def parse_ext_l(self, fm):
ext_dict = {
2015-06-15 07:41:24 -07:00
'flv': 'flv',
'mp4': 'mp4',
'hd2': 'flv',
'hd3': 'flv',
'3gp': 'flv',
'3gphd': 'mp4',
'flvhd': 'flv'
2015-05-28 02:00:09 -07:00
}
return ext_dict[fm]
2013-06-23 13:01:02 -07:00
2015-05-29 19:03:32 -07:00
def get_format_name(self, fm):
_dict = {
2015-06-15 07:41:24 -07:00
'3gp': 'h6',
'3gphd': 'h5',
'flvhd': 'h4',
2015-06-15 07:41:24 -07:00
'flv': 'h4',
'mp4': 'h3',
'hd2': 'h2',
'hd3': 'h1'
2015-05-29 19:03:32 -07:00
}
return _dict[fm]
2013-06-23 13:01:02 -07:00
def _real_extract(self, url):
2015-06-15 08:31:30 -07:00
video_id = self._match_id(url)
2013-06-23 13:01:02 -07:00
def retrieve_data(req_url, note):
headers = {
'Referer': req_url,
}
self._set_cookie('youku.com','xreferrer','http://www.youku.com')
req = sanitized_Request(req_url,headers=headers)
2013-06-23 13:01:02 -07:00
cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
if cn_verification_proxy:
req.add_header('Ytdl-request-proxy', cn_verification_proxy)
raw_data = self._download_json(req, video_id, note=note)
jsonDumpIn = json.dumps(raw_data,indent = 1)
return raw_data['data']
2015-08-31 09:19:58 -07:00
video_password = self._downloader.params.get('videopassword', None)
# request basic data
#basic_data_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
basic_data_url = "http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id
2015-08-31 09:19:58 -07:00
if video_password:
2015-09-01 09:26:17 -07:00
basic_data_url += '?password=%s' % video_password
2015-08-31 20:07:47 -07:00
data1 = retrieve_data(
basic_data_url,
'Downloading JSON metadata 1')
data2 = retrieve_data(
#'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
"http://play.youku.com/play/get.json?vid=%s&ct=12" % video_id,
'Downloading JSON metadata 2')
2014-09-25 00:58:09 -07:00
2015-05-28 02:00:09 -07:00
error_code = data1.get('error_code')
2014-09-25 00:58:09 -07:00
if error_code:
2015-05-28 02:00:09 -07:00
error = data1.get('error')
2015-06-15 08:54:55 -07:00
if error is not None and '因版权原因无法观看此视频' in error:
raise ExtractorError(
'Youku said: Sorry, this video is available in China only', expected=True)
else:
msg = 'Youku server reported error %i' % error_code
if error is not None:
msg += ': ' + error
raise ExtractorError(msg)
2015-05-28 02:00:09 -07:00
#title = data1['title']
title = data1['video']['title']
2015-05-28 02:00:09 -07:00
# generate video_urls_dict
video_urls_dict = self.construct_video_urls(data1, data2)
# construct info
entries = [{
'id': '%s_part%d' % (video_id, i + 1),
'title': title,
'formats': [],
# some formats are not available for all parts, we have to detect
# which one has all
} for i in range(max(len(v) for v in data1['stream']))]
for stream in data1['stream']:
fm = stream.get('stream_type')
2015-05-28 02:00:09 -07:00
video_urls = video_urls_dict[fm]
for video_url, seg, entry in zip(video_urls, stream['segs'], entries):
entry['formats'].append({
'url': video_url,
2015-06-15 09:15:09 -07:00
'format_id': self.get_format_name(fm),
'ext': self.parse_ext_l(fm),
'filesize': int(seg['size']),
2015-06-15 09:15:09 -07:00
})
2015-05-28 02:00:09 -07:00
return {
'_type': 'multi_video',
'id': video_id,
'title': title,
'entries': entries,
}