yt-dlp/youtube_dl/extractor/youku.py

335 lines
11 KiB
Python
Raw Normal View History

2013-06-27 14:32:02 -04:00
# coding: utf-8
2014-09-25 03:58:09 -04:00
from __future__ import unicode_literals
2015-05-28 05:00:09 -04:00
import base64
2016-06-08 11:45:46 -04:00
import itertools
2015-12-31 00:05:46 -05:00
import random
2016-06-08 11:45:46 -04:00
import re
2015-12-31 00:05:46 -05:00
import string
import time
2013-06-23 16:01:02 -04:00
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlencode,
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 11:18:17 -05:00
)
from ..utils import (
ExtractorError,
2016-06-08 11:45:46 -04:00
get_element_by_attribute,
)
2015-06-15 10:41:24 -04:00
2013-06-23 16:01:02 -04:00
class YoukuIE(InfoExtractor):
2015-05-28 05:00:09 -04:00
IE_NAME = 'youku'
2015-07-10 12:58:13 -04:00
IE_DESC = '优酷'
2014-09-25 03:58:09 -04:00
_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 05:00:09 -04:00
2015-06-15 11:36:28 -04:00
_TESTS = [{
2015-12-12 01:48:46 -05:00
# MD5 is unstable
2015-06-15 10:41:24 -04:00
'url': 'http://v.youku.com/v_show/id_XMTc1ODE5Njcy.html',
'info_dict': {
'id': 'XMTc1ODE5Njcy_part1',
2015-06-15 10:41:24 -04:00
'title': '★Smile﹗♡ Git Fresh -Booty Music舞蹈.',
'ext': 'flv'
}
2015-06-15 11:36:28 -04:00
}, {
'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,
2015-12-12 01:57:14 -05:00
'skip': 'Available in China only',
}, {
'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
'info_dict': {
'id': 'XMTI1OTczNDM5Mg',
'title': '花千骨 04',
},
'playlist_count': 13,
2015-08-31 12:19:58 -04:00
}, {
'url': 'http://v.youku.com/v_show/id_XNjA1NzA2Njgw.html',
'note': 'Video protected with password',
'info_dict': {
'id': 'XNjA1NzA2Njgw',
2015-09-01 12:26:17 -04:00
'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 12:19:58 -04:00
},
2015-09-01 12:28:03 -04:00
'playlist_count': 19,
2015-08-31 12:19:58 -04:00
'params': {
'videopassword': '100600',
},
}, {
# /play/get.json contains streams with "channel_type":"tail"
'url': 'http://v.youku.com/v_show/id_XOTUxMzg4NDMy.html',
'info_dict': {
'id': 'XOTUxMzg4NDMy',
'title': '我的世界☆明月庄主☆车震猎杀☆杀人艺术Minecraft',
},
'playlist_count': 6,
2015-06-15 11:36:28 -04:00
}]
2013-06-27 14:25:46 -04:00
2015-12-11 22:26:15 -05:00
def construct_video_urls(self, data):
2015-05-28 05:00:09 -04:00
# 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 05:00:09 -04:00
ls[i], ls[t] = ls[t], ls[i]
s = bytearray()
x, y = 0, 0
2015-05-28 05:00:09 -04:00
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 05:00:09 -04:00
sid, token = yk_t(
2015-12-11 22:26:15 -05:00
b'becaf9be', base64.b64decode(data['security']['encrypt_string'].encode('ascii'))
).decode('ascii').split('_')
2015-05-28 05:00:09 -04:00
# get oip
2015-12-11 22:26:15 -05:00
oip = data['security']['ip']
2015-05-28 05:00:09 -04:00
fileid_dict = {}
2015-12-11 22:26:15 -05:00
for stream in data['stream']:
if stream.get('channel_type') == 'tail':
continue
format = stream.get('stream_type')
fileid = stream['stream_fileid']
fileid_dict[format] = fileid
2015-05-28 05:00:09 -04:00
def get_fileid(format, n):
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:]
2015-05-28 05:00:09 -04:00
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 05:00:09 -04:00
return ep
# generate video_urls
video_urls_dict = {}
2015-12-11 22:26:15 -05:00
for stream in data['stream']:
if stream.get('channel_type') == 'tail':
continue
format = stream.get('stream_type')
2015-05-28 05:00:09 -04:00
video_urls = []
for dt in stream['segs']:
2015-12-11 06:18:14 -05:00
n = str(stream['segs'].index(dt))
param = {
'K': dt['key'],
'hd': self.get_hd(format),
'myp': 0,
'ypp': 0,
'ctype': 12,
'ev': 1,
'token': token,
'oip': oip,
'ep': generate_ep(format, n)
}
2015-05-28 05:00:09 -04:00
video_url = \
'http://k.youku.com/player/getFlvPath/' + \
'sid/' + sid + \
2015-12-12 01:41:53 -05:00
'_00' + \
2015-05-28 05:00:09 -04:00
'/st/' + self.parse_ext_l(format) + \
2015-06-15 10:41:24 -04:00
'/fileid/' + get_fileid(format, n) + '?' + \
compat_urllib_parse_urlencode(param)
2015-05-28 05:00:09 -04:00
video_urls.append(video_url)
video_urls_dict[format] = video_urls
return video_urls_dict
2015-12-31 00:05:46 -05:00
@staticmethod
def get_ysuid():
return '%d%s' % (int(time.time()), ''.join([
random.choice(string.ascii_letters) for i in range(3)]))
2015-05-28 05:00:09 -04:00
def get_hd(self, fm):
hd_id_dict = {
2015-06-15 10:41:24 -04:00
'3gp': '0',
'3gphd': '1',
2015-12-12 02:24:58 -05:00
'flv': '0',
'flvhd': '0',
2015-12-12 02:24:58 -05:00
'mp4': '1',
'mp4hd': '1',
2015-12-12 02:24:58 -05:00
'mp4hd2': '1',
2015-12-12 02:49:19 -05:00
'mp4hd3': '1',
2015-12-12 02:24:58 -05:00
'hd2': '2',
'hd3': '3',
2015-05-28 05:00:09 -04:00
}
return hd_id_dict[fm]
def parse_ext_l(self, fm):
ext_dict = {
2015-12-12 02:24:58 -05:00
'3gp': 'flv',
'3gphd': 'mp4',
2015-06-15 10:41:24 -04:00
'flv': 'flv',
2015-12-12 02:24:58 -05:00
'flvhd': 'flv',
2015-06-15 10:41:24 -04:00
'mp4': 'mp4',
2015-12-11 06:18:14 -05:00
'mp4hd': 'mp4',
'mp4hd2': 'flv',
'mp4hd3': 'flv',
2015-06-15 10:41:24 -04:00
'hd2': 'flv',
'hd3': 'flv',
2015-05-28 05:00:09 -04:00
}
return ext_dict[fm]
2013-06-23 16:01:02 -04:00
2015-05-29 22:03:32 -04:00
def get_format_name(self, fm):
_dict = {
2015-06-15 10:41:24 -04:00
'3gp': 'h6',
'3gphd': 'h5',
'flv': 'h4',
2015-12-12 02:24:58 -05:00
'flvhd': 'h4',
2015-06-15 10:41:24 -04:00
'mp4': 'h3',
'mp4hd': 'h3',
2015-12-12 02:24:58 -05:00
'mp4hd2': 'h4',
'mp4hd3': 'h4',
2015-12-12 02:24:58 -05:00
'hd2': 'h2',
'hd3': 'h1',
2015-05-29 22:03:32 -04:00
}
return _dict[fm]
2013-06-23 16:01:02 -04:00
def _real_extract(self, url):
2015-06-15 11:31:30 -04:00
video_id = self._match_id(url)
2013-06-23 16:01:02 -04:00
2015-12-31 00:05:46 -05:00
self._set_cookie('youku.com', '__ysuid', self.get_ysuid())
def retrieve_data(req_url, note):
headers = {
2015-12-12 01:41:53 -05:00
'Referer': req_url,
}
headers.update(self.geo_verification_headers())
2015-12-12 01:41:53 -05:00
self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
2013-06-23 16:01:02 -04:00
raw_data = self._download_json(req_url, video_id, note=note, headers=headers)
return raw_data['data']
video_password = self._downloader.params.get('videopassword')
2015-08-31 12:19:58 -04:00
# request basic data
2016-02-14 04:37:17 -05:00
basic_data_url = 'http://play.youku.com/play/get.json?vid=%s&ct=12' % video_id
2015-08-31 12:19:58 -04:00
if video_password:
basic_data_url += '&pwd=%s' % video_password
2015-08-31 23:07:47 -04:00
2015-12-12 02:19:14 -05:00
data = retrieve_data(basic_data_url, 'Downloading JSON metadata')
2014-09-25 03:58:09 -04:00
2015-12-11 22:26:15 -05:00
error = data.get('error')
if error:
error_note = error.get('note')
if error_note is not None and '因版权原因无法观看此视频' in error_note:
2015-06-15 11:54:55 -04:00
raise ExtractorError(
'Youku said: Sorry, this video is available in China only', expected=True)
elif error_note and '该视频被设为私密' in error_note:
raise ExtractorError(
'Youku said: Sorry, this video is private', expected=True)
2015-06-15 11:54:55 -04:00
else:
msg = 'Youku server reported error %i' % error.get('code')
2015-12-17 06:51:50 -05:00
if error_note is not None:
msg += ': ' + error_note
2015-06-15 11:54:55 -04:00
raise ExtractorError(msg)
2015-05-28 05:00:09 -04:00
2015-12-12 01:41:53 -05:00
# get video title
2015-12-11 22:26:15 -05:00
title = data['video']['title']
2015-05-28 05:00:09 -04:00
# generate video_urls_dict
2015-12-11 22:26:15 -05:00
video_urls_dict = self.construct_video_urls(data)
2015-05-28 05:00:09 -04:00
# 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
2015-12-11 22:26:15 -05:00
} for i in range(max(len(v.get('segs')) for v in data['stream']))]
for stream in data['stream']:
if stream.get('channel_type') == 'tail':
continue
fm = stream.get('stream_type')
2015-05-28 05:00:09 -04:00
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 12:15:09 -04:00
'format_id': self.get_format_name(fm),
'ext': self.parse_ext_l(fm),
'filesize': int(seg['size']),
2016-05-29 01:54:05 -04:00
'width': stream.get('width'),
'height': stream.get('height'),
2015-06-15 12:15:09 -04:00
})
2015-05-28 05:00:09 -04:00
return {
'_type': 'multi_video',
'id': video_id,
'title': title,
'entries': entries,
}
2016-06-08 11:45:46 -04:00
class YoukuShowIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?youku\.com/show_page/id_(?P<id>[0-9a-z]+)\.html'
IE_NAME = 'youku:show'
_TEST = {
'url': 'http://www.youku.com/show_page/id_zc7c670be07ff11e48b3f.html',
'info_dict': {
'id': 'zc7c670be07ff11e48b3f',
'title': '花千骨 未删减版',
'description': 'md5:578d4f2145ae3f9128d9d4d863312910',
},
'playlist_count': 50,
}
_PAGE_SIZE = 40
def _find_videos_in_page(self, webpage):
videos = re.findall(
r'<li><a[^>]+href="(?P<url>https?://v\.youku\.com/[^"]+)"[^>]+title="(?P<title>[^"]+)"', webpage)
return [
self.url_result(video_url, YoukuIE.ie_key(), title)
for video_url, title in videos]
def _real_extract(self, url):
show_id = self._match_id(url)
webpage = self._download_webpage(url, show_id)
entries = self._find_videos_in_page(webpage)
playlist_title = self._html_search_regex(
r'<span[^>]+class="name">([^<]+)</span>', webpage, 'playlist title', fatal=False)
detail_div = get_element_by_attribute('class', 'detail', webpage) or ''
playlist_description = self._html_search_regex(
r'<span[^>]+style="display:none"[^>]*>([^<]+)</span>',
detail_div, 'playlist description', fatal=False)
for idx in itertools.count(1):
episodes_page = self._download_webpage(
'http://www.youku.com/show_episode/id_%s.html' % show_id,
show_id, query={'divid': 'reload_%d' % (idx * self._PAGE_SIZE + 1)},
note='Downloading episodes page %d' % idx)
new_entries = self._find_videos_in_page(episodes_page)
entries.extend(new_entries)
if len(new_entries) < self._PAGE_SIZE:
break
return self.playlist_result(entries, show_id, playlist_title, playlist_description)