parse tudou url like http://www.tudou.com/albumplay/TenTw_JgiPM/PzsAs5usU9A.html

youtube_dl/extractor/tudou.py

@@ -2,12 +2,17 @@
 
 import re
 import json
+import urllib2
+
+from time import time
+from random import randint
 
 from .common import InfoExtractor
 
 
 class TudouIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(?:listplay|programs)/(?:view|(.+?))/(?:([^/]+)|([^/]+))(?:\.html)?'
+	#_VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/((?:listplay|programs)/(?:view|(.+?)))|(?:albumplay)/(?:([^/]+)|([^/]+))(?:\.html)?'
+	_VALID_URL = r'(?:http://)?(?:www\.)?tudou\.com/(((?:listplay|programs)/(?:view|(.+?)))|((?:albumplay)(?:/[^/]*)))/(?:([^/]+)|([^/]+))(?:\.html)?'
 	_TEST = {
 	u'url': u'http://www.tudou.com/listplay/zzdE77v6Mmo/2xN2duXMxmw.html',
 	u'file': u'159448201.f4v',
@@ -25,13 +30,106 @@ class TudouIE(InfoExtractor):
 		final_url = self._html_search_regex('>(.+?)</f>',webpage, 'video url')
 		return final_url
 
+	def get_page(self,url):
+		request=urllib2.urlopen(url)
+		html=request.read()
+		content_type=request.headers.get('Content-Type')
+		m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
+		if m:
+			encoding=m.group(1)
+			html=html.decode(encoding,"replace")
+			return html
+		else:
+			return None
+
+
+	def isyouku(self,url):
+		request=urllib2.urlopen(url)
+		html=request.read()
+		content_type=request.headers.get('Content-Type')
+		m = re.match(r'[a-zA-Z0-9_.-]+/[a-zA-Z0-9_.-]+\s*;\s*charset=(.+)', content_type)
+		if m:
+			encoding=m.group(1)
+			html=html.decode(encoding,"replace")
+		vcode=re.search(r'vcode\s*[:=]\s*\'([^\']+)\'',html).group(1)
+		title=re.search(r'kw\s*[:=]\s*[\'\"]([^\']+?)[\'\"]',html).group(1)
+		if vcode:
+			return (vcode,title)
+		else:
+			return None
+
+	def downloadYouku_by_id(self,videoId,title):
+		info=self.get_youkuinfo(videoId)
+		result=[]
+		urls,sizes=zip(*self.find_video(info,None))
+		pattern=re.compile(r'/st/([^/]+)/')
+		ext=str(re.search(pattern, urls[0]).group(1))
+		for i,url in enumerate(urls):
+			part_info={
+					'id':i,
+					'url':url,
+					'ext':ext,
+					'title':title,
+					'thumbnail':None,
+					}
+			result.append(part_info)
+
+		return result
+
+
+	def get_youkuinfo(self,videoId):
+		return json.loads(self.get_page('http://v.youku.com/player/getPlayList/VideoIDS/' + videoId + '/timezone/+08/version/5/source/out/Sc/2'))
+
+	def find_video(self,info, stream_type = None):
+		#key = '%s%x' % (info['data'][0]['key2'], int(info['data'][0]['key1'], 16) ^ 0xA55AA5A5)
+		segs = info['data'][0]['segs']
+		types = segs.keys()
+		if not stream_type:
+			for x in ['hd2', 'mp4', 'flv']:
+				if x in types:
+					stream_type = x
+					break
+			else:
+				raise NotImplementedError()
+		assert stream_type in ('hd2', 'mp4', 'flv')
+		file_type = {'hd2': 'flv', 'mp4': 'mp4', 'flv': 'flv'}[stream_type]
+
+		seed = info['data'][0]['seed']
+		source = list("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ/\\:._-1234567890")
+		mixed = ''
+		while source:
+			seed = (seed * 211 + 30031) & 0xFFFF
+			index = seed * len(source) >> 16
+			c = source.pop(index)
+			mixed += c
+
+		ids = info['data'][0]['streamfileids'][stream_type].split('*')[:-1]
+		vid = ''.join(mixed[int(i)] for i in ids)
+
+		sid = '%s%s%s' % (int(time() * 1000), randint(1000, 1999), randint(1000, 9999))
+
+		urls = []
+		for s in segs[stream_type]:
+			no = '%02x' % int(s['no'])
+			url = 'http://f.youku.com/player/getFlvPath/sid/%s_%s/st/%s/fileid/%s%s%s?K=%s&ts=%s' % (sid, no, file_type, vid[:8], no.upper(), vid[10:], s['k'], s['seconds'])
+			urls.append((url, int(s['size'])))
+		return urls
+
 	def _real_extract(self, url):
 		mobj = re.match(self._VALID_URL, url)
 		video_id = mobj.group(2)
+		if video_id is None:
+				vcode,title=self.isyouku(url)
+				if not vcode:
+					print "Not transferring to Youku"
+					return None
+				return self.downloadYouku_by_id(vcode,title)
+
 		webpage = self._download_webpage(url, video_id)
 		title = re.search(",kw:\"(.+)\"",webpage)
 		if title is None:
 			title = re.search(",kw: \'(.+)\'",webpage)
+
 		title = title.group(1)
 		thumbnail_url = re.search(",pic: \'(.+?)\'",webpage)
 		if thumbnail_url is None: