yt-dlp/youtube_dl/extractor/youku.py

231 lines
7.1 KiB
Python
Raw Normal View History

2013-06-27 18:32:02 +00:00
# coding: utf-8
2014-09-25 07:58:09 +00:00
from __future__ import unicode_literals
2013-06-23 20:01:02 +00:00
import re
2015-05-28 09:00:09 +00:00
import base64
2013-06-23 20:01:02 +00:00
from .common import InfoExtractor
2015-05-28 09:00:09 +00:00
from ..utils import ExtractorError
2013-06-23 20:01:02 +00:00
from ..compat import compat_urllib_parse
bytes_is_str = (bytes == str) # for compatible
2013-06-23 20:01:02 +00:00
class YoukuIE(InfoExtractor):
2015-05-28 09:00:09 +00:00
IE_NAME = 'youku'
2014-09-25 07:58:09 +00:00
_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 09:00:09 +00:00
2014-09-25 07:58:09 +00:00
_TEST = {
2015-05-28 09:00:09 +00:00
'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
'md5': '5f3af4192eabacc4501508d54a8cabd7',
'info_dict': {
'id': 'XMTc1ODE5Njcy',
'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
'ext': 'flv'
}
2013-06-27 18:25:46 +00:00
}
2015-05-28 09:00:09 +00:00
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] + ord(s1[i%len(s1)])) % 256
ls[i], ls[t] = ls[t], ls[i]
s = '' if not bytes_is_str else b''
x, y = 0, 0
2015-05-28 09:00:09 +00:00
for i in range(len(s2)):
y = (y + 1) % 256
x = (x + ls[y]) % 256
ls[x], ls[y] = ls[y], ls[x]
if isinstance(s2[i], int):
s += chr(s2[i] ^ ls[(ls[x]+ls[y]) % 256])
else:
s += chr(ord(s2[i]) ^ ls[(ls[x]+ls[y]) % 256])
2015-05-28 09:00:09 +00:00
return s
sid, token = yk_t(
'becaf9be',
base64.b64decode(bytes(data2['ep'], 'ascii')) \
if not bytes_is_str \
else base64.b64decode(data2['ep'])
2015-05-28 09:00:09 +00:00
).split('_')
# get oip
oip = data2['ip']
# get fileid
string_ls = list(
'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
shuffled_string_ls = []
seed = data1['seed']
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 format in data1['streamtypes']:
streamfileid = [
int(i) for i in data1['streamfileids'][format].strip('*').split('*')]
fileid = ''.join(
[shuffled_string_ls[i] for i in streamfileid])
fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
def get_fileid(format, n):
fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2)
return fileid
# get ep
def generate_ep(format, n):
fileid = get_fileid(format, n)
ep_t = yk_t(
'bf7e5f01',
bytes('%s_%s_%s' % (sid, fileid, token), 'ascii') \
if not bytes_is_str \
else ('%s_%s_%s' % (sid, fileid, token))
)
ep = base64.b64encode(
bytes(ep_t, 'latin') \
if not bytes_is_str \
else ep_t
).decode()
2015-05-28 09:00:09 +00:00
return ep
# generate video_urls
video_urls_dict = {}
for format in data1['streamtypes']:
video_urls = []
for dt in data1['segs'][format]:
n = str(int(dt['no']))
param = {
'K': dt['k'],
'hd': self.get_hd(format),
'myp': 0,
'ts': dt['seconds'],
'ypp': 0,
'ctype': 12,
'ev': 1,
'token': token,
'oip': oip,
'ep': generate_ep(format, n)
}
2015-05-28 09:00:09 +00:00
video_url = \
'http://k.youku.com/player/getFlvPath/' + \
'sid/' + sid + \
'_' + str(int(n)+1).zfill(2) + \
'/st/' + self.parse_ext_l(format) + \
'/fileid/' + get_fileid(format, n) + '?' + \
compat_urllib_parse.urlencode(param)
2015-05-28 09:00:09 +00:00
video_urls.append(video_url)
video_urls_dict[format] = video_urls
return video_urls_dict
def get_hd(self, fm):
hd_id_dict = {
2015-05-30 02:03:32 +00:00
'flv' : '0',
'mp4' : '1',
'hd2' : '2',
'hd3' : '3',
'3gp' : '0',
'3gphd' : '1'
2015-05-28 09:00:09 +00:00
}
return hd_id_dict[fm]
def parse_ext_l(self, fm):
ext_dict = {
2015-05-30 02:03:32 +00:00
'flv' : 'flv',
'mp4' : 'mp4',
'hd2' : 'flv',
'hd3' : 'flv',
'3gp' : 'flv',
'3gphd' : 'mp4'
2015-05-28 09:00:09 +00:00
}
return ext_dict[fm]
2013-06-23 20:01:02 +00:00
2015-05-30 02:03:32 +00:00
def get_format_name(self, fm):
_dict = {
'3gp' : 'h6',
'3gphd' : 'h5',
'flv' : 'h4',
'mp4' : 'h3',
'hd2' : 'h2',
'hd3' : 'h1'
}
return _dict[fm]
2013-06-23 20:01:02 +00:00
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
2014-09-25 07:58:09 +00:00
video_id = mobj.group('id')
2013-06-23 20:01:02 +00:00
2015-05-28 09:00:09 +00:00
# request basic data
data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id
2013-06-23 20:01:02 +00:00
2015-05-28 09:00:09 +00:00
raw_data1 = self._download_json(data1_url, video_id)
raw_data2 = self._download_json(data2_url, video_id)
data1 = raw_data1['data'][0]
data2 = raw_data2['data'][0]
2014-09-25 07:58:09 +00:00
2015-05-28 09:00:09 +00:00
error_code = data1.get('error_code')
2014-09-25 07:58:09 +00:00
if error_code:
# -8 means blocked outside China.
2015-05-28 09:00:09 +00:00
# Chinese and English, separated by newline.
error = data1.get('error')
raise ExtractorError(
error or 'Server reported error %i' %
error_code,
expected=True)
title = data1['title']
# generate video_urls_dict
video_urls_dict = self.construct_video_urls(data1, data2)
# construct info
entries = []
for fm in data1['streamtypes']:
#formats = []
video_urls = video_urls_dict[fm]
for i in range(len(video_urls)):
if len(entries) < i+1:
entries.append({'formats': []})
entries[i]['formats'].append(
{
'url': video_urls[i],
2015-05-30 02:03:32 +00:00
'format_id': self.get_format_name(fm),
2015-05-28 09:00:09 +00:00
'ext': self.parse_ext_l(fm),
'filesize': int(data1['segs'][fm][i]['size'])
}
)
for i in range(len(entries)):
entries[i].update(
{
'id': '_part%d' % (i+1),
'title': title,
}
)
if len(entries) > 1:
2013-06-23 20:01:02 +00:00
info = {
2015-05-28 09:00:09 +00:00
'_type': 'multi_video',
'id': video_id,
'title': title,
'entries': entries,
2013-06-23 20:01:02 +00:00
}
2015-05-28 09:00:09 +00:00
else:
info = entries[0]
info['id'] = video_id
2013-06-23 20:01:02 +00:00
2015-05-28 09:00:09 +00:00
return info