Python3 Automatically download Youku video applet

We are generally in Youku to see some fun video, and sometimes see the wonderful download to the local save it as a souvenir, in the win can be used to download the software, such as Victoria, but suffering the children with Linux. Although some of Chrome and Firefox plug-ins can be downloaded, but sometimes download it is segmented video, but also to manually use FFmpeg to synthesize a video is too cumbersome, and plug-in download to click a few times the mouse, too does not conform to Linuxer's character ha ~ So I wrote a youkudown.py with Python3, the code is as follows:

[Python]View PlainCopy 

2. Import re
4. Import Urllib.request
6. Import JSON
8. Import time
10. Import Random
12. Import Sys
16. def getvideoinfo (URL):
18. Ruletitle=re.compile (' <title> (. *) </title> ')
20. Ruleid=re.compile (' http://v.youku.com/v_show/id_ (. *). html ')
22. Videotitle=ruletitle.findall (Urllib.request.urlopen (URL). read (). Decode (' UTF8 '))
24. Videoid=ruleid.findall (URL)
26. return videotitle[0],videoid[0]
30. def gettruelink (videoid):
32. Data=urllib.request.urlopen (' http://v.youku.com/player/getPlayList/VideoIDS/' +videoid)
34. Info=json.loads (Data.read (). Decode (' UTF8 '))
38. segs=info[' data ' [0][' segs ']
40. Types=segs.keys ()
44. seed=info[' data ' [0][' seed ']
46. Source=list ("abcdefghijklmnopqrstuvwxyzabcdefghijklmnopqrstuvwxyz/\\:._-1234567890")
48. mixed="
50. While Source:
52. seed= (seed*211+30031) &0xFFFF
54. Index=seed*len (source) >>
56. C=source.pop (Index)
58. Mixed+=c
62. ids=info[' data ' [0][' streamfileids '] [' flv '].split (' * ') [:-1]
64. Vid=". Join (Mixed[int (i)] for i in IDs)
68. sid='%s%s%s '% (int (time.time () *), Random.randint (1999), Random.randint (9999 ))  
72. Urls=[]
74. For s in segs[' flv ']:
76. no='%02x '%int (s[' no ')
78. Url=' http://f.youku.com/player/getFlvPath/sid/%s_%s/st/flv/fileid/%s%s%s? k=%s&ts=%s '% (sid,no,vid[:8],no.upper (), vid[10:],s[' k '],s['seconds '])
80. Urls.append (URL)
82. return URLs
86. def down2file (urls,filename):
88. F=open (filename,' WB ')
90. Filenum=len (URLs)
92. count=0
94. For URL in URLs:
96. count+=1
98. print (' downloading file%d/%d '% (count,filenum))
100. Req=urllib.request.request (url,headers={' Referer ':' http://www.youku.com ')
102. Data=urllib.request.urlopen (req). Read ()
104. F.write (data)
106. F.close ()
108. print (' Download ' +filename+' ok! ')
112. def youkudown (link):
114. Videotitle,videoid=getvideoinfo (link)
116. Urls=gettruelink (VideoID)
118. Down2file (urls,videotitle+'. flv ')
122. If __name__==' __main__ ':
124. If Len (SYS.ARGV) <2:
126. print (' Example usage:python3 youkudown.py http://v.youku.com/v_show/id_XNjUzNDQwMzIw.html ')
128. print (")
130. Exit ()
132. Youkudown (sys.argv[1])

Briefly, because the true link of Youku video is dynamic, and the ID of each video is constant, for example, in link http://v.youku.com/v_show/id_XNjUzNDQwMzIw.html, "Xnjuzndqwmziw" Is the unique ID identifier of the video, through which the ID can request a new link generated by the "http://v.youku.com/player/getPlayList/VideoIDS/" +id, and get the video information in JSON format. Then assembled into a video of the real link, this real link is also time-sensitive, and to this real link to initiate the request must be the head of the Referer set to www.youku.com fooled the server to download, so I did not like before with Os.system (' wget ' + URL). Also, Youku on the big video is segmented, so get the real URL I have a list inside, download the time will show a total of a few paragraphs and real-time download how many paragraphs. Here are some examples of my use:

To save to a local video playback situation:

In Linux want to download Youku video of the pot friends enjoy it~~ at the same time I wish you a happy New Year ~ ha haha

Python3 Automatically download Youku video applet