Roxedus/ytdl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Merge branch 'Celthi-youku_bugfix'

Browse Source
This commit is contained in:
Yen Chi Hsuan 2015-12-12 15:26:43 +08:00
commit 9c568178fb
1 changed files with 59 additions and 56 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

115
youtube_dl/extractor/youku.py
View File

@ -25,8 +25,8 @@ class YoukuIE(InfoExtractor):
''' '''
_TESTS = [{ _TESTS = [{
# MD5 is unstable
'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html', 'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
'md5': '5f3af4192eabacc4501508d54a8cabd7',
'info_dict': { 'info_dict': {
'id': 'XMTc1ODE5Njcy_part1', 'id': 'XMTc1ODE5Njcy_part1',
'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.', 'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
@ -42,6 +42,7 @@ class YoukuIE(InfoExtractor):
'title': '武媚娘传奇 85', 'title': '武媚娘传奇 85',
}, },
'playlist_count': 11, 'playlist_count': 11,
'skip': 'Available in China only',
}, { }, {
'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html', 'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
'info_dict': { 'info_dict': {
@ -49,7 +50,6 @@ class YoukuIE(InfoExtractor):
'title': '花千骨 04', 'title': '花千骨 04',
}, },
'playlist_count': 13, 'playlist_count': 13,
'skip': 'Available in China only',
}, { }, {
'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html', 'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
'note': 'Video protected with password', 'note': 'Video protected with password',
@ -63,7 +63,7 @@ class YoukuIE(InfoExtractor):
}, },
}] }]
def construct_video_urls(self, data1, data2): def construct_video_urls(self, data):
# get sid, token # get sid, token
def yk_t(s1, s2): def yk_t(s1, s2):
ls = list(range(256)) ls = list(range(256))
@ -81,34 +81,24 @@ class YoukuIE(InfoExtractor):
return bytes(s) return bytes(s)
sid, token = yk_t( sid, token = yk_t(
b'becaf9be', base64.b64decode(data2['ep'].encode('ascii')) b'becaf9be', base64.b64decode(data['security']['encrypt_string'].encode('ascii'))
).decode('ascii').split('_') ).decode('ascii').split('_')
# get oip # get oip
oip = data2['ip'] oip = data['security']['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 = {} fileid_dict = {}
for format in data1['streamtypes']: for stream in data['stream']:
streamfileid = [ format = stream.get('stream_type')
int(i) for i in data1['streamfileids'][format].strip('*').split('*')] fileid = stream['stream_fileid']
fileid = ''.join( fileid_dict[format] = fileid
[shuffled_string_ls[i] for i in streamfileid])
fileid_dict[format] = fileid[:8] + '%s' + fileid[10:]
def get_fileid(format, n): def get_fileid(format, n):
fileid = fileid_dict[format] % hex(int(n))[2:].upper().zfill(2) number = hex(int(str(n), 10))[2:].upper()
if len(number) == 1:
number = '0' + number
streamfileids = fileid_dict[format]
fileid = streamfileids[0:8] + number + streamfileids[10:]
return fileid return fileid
# get ep # get ep
@ -123,15 +113,15 @@ class YoukuIE(InfoExtractor):
# generate video_urls # generate video_urls
video_urls_dict = {} video_urls_dict = {}
for format in data1['streamtypes']: for stream in data['stream']:
format = stream.get('stream_type')
video_urls = [] video_urls = []
for dt in data1['segs'][format]: for dt in stream['segs']:
n = str(int(dt['no'])) n = str(stream['segs'].index(dt))
param = { param = {
'K': dt['k'], 'K': dt['key'],
'hd': self.get_hd(format), 'hd': self.get_hd(format),
'myp': 0, 'myp': 0,
'ts': dt['seconds'],
'ypp': 0, 'ypp': 0,
'ctype': 12, 'ctype': 12,
'ev': 1, 'ev': 1,
@ -142,7 +132,7 @@ class YoukuIE(InfoExtractor):
video_url = \ video_url = \
'http://k.youku.com/player/getFlvPath/' + \ 'http://k.youku.com/player/getFlvPath/' + \
'sid/' + sid + \ 'sid/' + sid + \
'_' + str(int(n) + 1).zfill(2) + \ '_00' + \
'/st/' + self.parse_ext_l(format) + \ '/st/' + self.parse_ext_l(format) + \
'/fileid/' + get_fileid(format, n) + '?' + \ '/fileid/' + get_fileid(format, n) + '?' + \
compat_urllib_parse.urlencode(param) compat_urllib_parse.urlencode(param)
@ -153,23 +143,30 @@ class YoukuIE(InfoExtractor):
def get_hd(self, fm): def get_hd(self, fm):
hd_id_dict = { hd_id_dict = {
'3gp': '0',
'3gphd': '1',
'flv': '0', 'flv': '0',
'flvhd': '0',
'mp4': '1', 'mp4': '1',
'mp4hd': '1',
'mp4hd2': '1',
'hd2': '2', 'hd2': '2',
'hd3': '3', 'hd3': '3',
'3gp': '0',
'3gphd': '1'
} }
return hd_id_dict[fm] return hd_id_dict[fm]
def parse_ext_l(self, fm): def parse_ext_l(self, fm):
ext_dict = { ext_dict = {
'3gp': 'flv',
'3gphd': 'mp4',
'flv': 'flv', 'flv': 'flv',
'flvhd': 'flv',
'mp4': 'mp4', 'mp4': 'mp4',
'mp4hd': 'mp4',
'mp4hd2': 'flv',
'mp4hd3': 'flv',
'hd2': 'flv', 'hd2': 'flv',
'hd3': 'flv', 'hd3': 'flv',
'3gp': 'flv',
'3gphd': 'mp4'
} }
return ext_dict[fm] return ext_dict[fm]
@ -178,9 +175,13 @@ class YoukuIE(InfoExtractor):
'3gp': 'h6', '3gp': 'h6',
'3gphd': 'h5', '3gphd': 'h5',
'flv': 'h4', 'flv': 'h4',
'flvhd': 'h4',
'mp4': 'h3', 'mp4': 'h3',
'mp4hd': 'h3',
'mp4hd2': 'h4',
'mp4hd3': 'h4',
'hd2': 'h2', 'hd2': 'h2',
'hd3': 'h1' 'hd3': 'h1',
} }
return _dict[fm] return _dict[fm]
@ -188,45 +189,46 @@ class YoukuIE(InfoExtractor):
video_id = self._match_id(url) video_id = self._match_id(url)
def retrieve_data(req_url, note): def retrieve_data(req_url, note):
req = sanitized_Request(req_url) headers = {
'Referer': req_url,
}
self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
req = sanitized_Request(req_url, headers=headers)
cn_verification_proxy = self._downloader.params.get('cn_verification_proxy') cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
if cn_verification_proxy: if cn_verification_proxy:
req.add_header('Ytdl-request-proxy', cn_verification_proxy) req.add_header('Ytdl-request-proxy', cn_verification_proxy)
raw_data = self._download_json(req, video_id, note=note) raw_data = self._download_json(req, video_id, note=note)
return raw_data['data'][0]
return raw_data['data']
video_password = self._downloader.params.get('videopassword', None) video_password = self._downloader.params.get('videopassword', None)
# request basic data # 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
if video_password: if video_password:
basic_data_url += '?password=%s' % video_password basic_data_url += '&pwd=%s' % video_password
data1 = retrieve_data( data = retrieve_data(basic_data_url, 'Downloading JSON metadata')
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,
'Downloading JSON metadata 2')
error_code = data1.get('error_code') error = data.get('error')
if error_code: if error:
error = data1.get('error') error_note = error.get('note')
if error is not None and '因版权原因无法观看此视频' in error: if error_note is not None and '因版权原因无法观看此视频' in error_note:
raise ExtractorError( raise ExtractorError(
'Youku said: Sorry, this video is available in China only', expected=True) 'Youku said: Sorry, this video is available in China only', expected=True)
else: else:
msg = 'Youku server reported error %i' % error_code msg = 'Youku server reported error %i' % error.get('code')
if error is not None: if error is not None:
msg += ': ' + error msg += ': ' + error_note
raise ExtractorError(msg) raise ExtractorError(msg)
title = data1['title'] # get video title
title = data['video']['title']
# generate video_urls_dict # generate video_urls_dict
video_urls_dict = self.construct_video_urls(data1, data2) video_urls_dict = self.construct_video_urls(data)
# construct info # construct info
entries = [{ entries = [{
@ -235,10 +237,11 @@ class YoukuIE(InfoExtractor):
'formats': [], 'formats': [],
# some formats are not available for all parts, we have to detect # some formats are not available for all parts, we have to detect
# which one has all # which one has all
} for i in range(max(len(v) for v in data1['segs'].values()))] } for i in range(max(len(v.get('segs')) for v in data['stream']))]
for fm in data1['streamtypes']: for stream in data['stream']:
fm = stream.get('stream_type')
video_urls = video_urls_dict[fm] video_urls = video_urls_dict[fm]
for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries): for video_url, seg, entry in zip(video_urls, stream['segs'], entries):
entry['formats'].append({ entry['formats'].append({
'url': video_url, 'url': video_url,
'format_id': self.get_format_name(fm), 'format_id': self.get_format_name(fm),