Merge branch 'PeterDing-youku'

This commit is contained in:
Yen Chi Hsuan 2015-06-16 00:20:52 +08:00
commit 79fe954d79
2 changed files with 204 additions and 96 deletions

View File

@ -1033,12 +1033,6 @@ def process_video_result(self, info_dict, download=True):
info_dict['id'], info_dict.get('subtitles'), info_dict['id'], info_dict.get('subtitles'),
info_dict.get('automatic_captions')) info_dict.get('automatic_captions'))
# This extractors handle format selection themselves
if info_dict['extractor'] in ['Youku']:
if download:
self.process_info(info_dict)
return info_dict
# We now pick which formats have to be downloaded # We now pick which formats have to be downloaded
if info_dict.get('formats') is None: if info_dict.get('formats') is None:
# There's only one format available # There's only one format available

View File

@ -1,123 +1,237 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import math import base64
import random
import re
import time
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import ExtractorError
ExtractorError,
from ..compat import (
compat_urllib_parse,
compat_ord,
compat_urllib_request,
) )
class YoukuIE(InfoExtractor): class YoukuIE(InfoExtractor):
IE_NAME = 'youku'
_VALID_URL = r'''(?x) _VALID_URL = r'''(?x)
(?: (?:
http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)| http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
youku:) youku:)
(?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|) (?P<id>[A-Za-z0-9]+)(?:\.html|/v\.swf|)
''' '''
_TEST = {
'url': 'http://v.youku.com/v_show/id_XNDgyMDQ2NTQw.html', _TESTS = [{
'md5': 'ffe3f2e435663dc2d1eea34faeff5b5b', 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
'params': { 'md5': '5f3af4192eabacc4501508d54a8cabd7',
'test': False
},
'info_dict': { 'info_dict': {
'id': 'XNDgyMDQ2NTQw_part00', 'id': 'XMTc1ODE5Njcy_part1',
'ext': 'flv', 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
'title': 'youtube-dl test video "\'/\\ä↭𝕐' 'ext': 'flv'
} }
} }, {
'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',
}]
def _gen_sid(self): def construct_video_urls(self, data1, data2):
nowTime = int(time.time() * 1000) # get sid, token
random1 = random.randint(1000, 1998) def yk_t(s1, s2):
random2 = random.randint(1000, 9999) ls = list(range(256))
t = 0
for i in range(256):
t = (t + ls[i] + compat_ord(s1[i % len(s1)])) % 256
ls[i], ls[t] = ls[t], ls[i]
s = bytearray()
x, y = 0, 0
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)
return "%d%d%d" % (nowTime, random1, random2) sid, token = yk_t(
b'becaf9be', base64.b64decode(data2['ep'].encode('ascii'))
).decode('ascii').split('_')
def _get_file_ID_mix_string(self, seed): # get oip
mixed = [] oip = data2['ip']
source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890")
seed = float(seed)
for i in range(len(source)):
seed = (seed * 211 + 30031) % 65536
index = math.floor(seed / 65536 * len(source))
mixed.append(source[int(index)])
source.remove(source[int(index)])
# return ''.join(mixed)
return mixed
def _get_file_id(self, fileId, seed): # get fileid
mixed = self._get_file_ID_mix_string(seed) string_ls = list(
ids = fileId.split('*') 'abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\:._-1234567890')
realId = [] shuffled_string_ls = []
for ch in ids: seed = data1['seed']
if ch: N = len(string_ls)
realId.append(mixed[int(ch)]) for ii in range(N):
return ''.join(realId) 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(
b'bf7e5f01',
('%s_%s_%s' % (sid, fileid, token)).encode('ascii')
)
ep = base64.b64encode(ep_t).decode('ascii')
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)
}
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)
video_urls.append(video_url)
video_urls_dict[format] = video_urls
return video_urls_dict
def get_hd(self, fm):
hd_id_dict = {
'flv': '0',
'mp4': '1',
'hd2': '2',
'hd3': '3',
'3gp': '0',
'3gphd': '1'
}
return hd_id_dict[fm]
def parse_ext_l(self, fm):
ext_dict = {
'flv': 'flv',
'mp4': 'mp4',
'hd2': 'flv',
'hd3': 'flv',
'3gp': 'flv',
'3gphd': 'mp4'
}
return ext_dict[fm]
def get_format_name(self, fm):
_dict = {
'3gp': 'h6',
'3gphd': 'h5',
'flv': 'h4',
'mp4': 'h3',
'hd2': 'h2',
'hd3': 'h1'
}
return _dict[fm]
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) video_id = self._match_id(url)
video_id = mobj.group('id')
info_url = 'http://v.youku.com/player/getPlayList/VideoIDS/' + video_id def retrieve_data(req_url, note):
req = compat_urllib_request.Request(req_url)
config = self._download_json(info_url, video_id) cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
if cn_verification_proxy:
req.add_header('Ytdl-request-proxy', cn_verification_proxy)
error_code = config['data'][0].get('error_code') raw_data = self._download_json(req, video_id, note=note)
return raw_data['data'][0]
# request basic data
data1 = retrieve_data(
'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id,
'Downloading JSON metadata 1')
data2 = retrieve_data(
'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
'Downloading JSON metadata 2')
error_code = data1.get('error_code')
if error_code: if error_code:
# -8 means blocked outside China. error = data1.get('error')
error = config['data'][0].get('error') # Chinese and English, separated by newline. if error is not None and '因版权原因无法观看此视频' in error:
raise ExtractorError(error or 'Server reported error %i' % error_code, raise ExtractorError(
expected=True) 'Youku said: Sorry, this video is available in China only', expected=True)
video_title = config['data'][0]['title']
seed = config['data'][0]['seed']
format = self._downloader.params.get('format', None)
supported_format = list(config['data'][0]['streamfileids'].keys())
# TODO proper format selection
if format is None or format == 'best':
if 'hd2' in supported_format:
format = 'hd2'
else: else:
format = 'flv' msg = 'Youku server reported error %i' % error_code
ext = 'flv' if error is not None:
elif format == 'worst': msg += ': ' + error
format = 'mp4' raise ExtractorError(msg)
ext = 'mp4'
else:
format = 'flv'
ext = 'flv'
fileid = config['data'][0]['streamfileids'][format] title = data1['title']
keys = [s['k'] for s in config['data'][0]['segs'][format]]
# segs is usually a dictionary, but an empty *list* if an error occured.
files_info = [] # generate video_urls_dict
sid = self._gen_sid() video_urls_dict = self.construct_video_urls(data1, data2)
fileid = self._get_file_id(fileid, seed)
# column 8,9 of fileid represent the segment number # construct info
# fileid[7:9] should be changed entries = []
for index, key in enumerate(keys): for fm in data1['streamtypes']:
temp_fileid = '%s%02X%s' % (fileid[0:8], index, fileid[10:]) video_urls = video_urls_dict[fm]
download_url = 'http://k.youku.com/player/getFlvPath/sid/%s_%02X/st/flv/fileid/%s?k=%s' % (sid, index, temp_fileid, key) for i in range(len(video_urls)):
if len(entries) < i + 1:
entries.append({'formats': []})
entries[i]['formats'].append({
'url': video_urls[i],
'format_id': self.get_format_name(fm),
'ext': self.parse_ext_l(fm),
'filesize': int(data1['segs'][fm][i]['size'])
})
info = { for i in range(len(entries)):
'id': '%s_part%02d' % (video_id, index), entries[i].update({
'url': download_url, 'id': '%s_part%d' % (video_id, i + 1),
'uploader': None, 'title': title,
'upload_date': None, })
'title': video_title,
'ext': ext,
}
files_info.append(info)
return files_info return {
'_type': 'multi_video',
'id': video_id,
'title': title,
'entries': entries,
}