code.delx.au - monosys/blob - ripping/youtube.cgi

   1 #!/usr/bin/env python
   2
   3 import cgi
   4 from lxml.html import document_fromstring
   5 import os
   6 import re
   7 import shutil
   8 import subprocess
   9 import sys
  10 import urllib
  11
  12
  13 class VideoUnavailable(Exception):
  14         pass
  15
  16 urllib.URLopener.version = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0.1) Gecko/20100101 Firefox/4.0.1"
  17
  18 fmt_quality = [
  19         (38, ".mp4"),  # 4096x3072
  20         (37, ".mp4"),  # 1920x1080
  21         (22, ".mp4"),  # 1280x720
  22 ###     (45, ".webm"), # 1280x720
  23 ###     (43, ".webm"), # 640x360
  24         (35, ".flv"),  # 854x480
  25         (34, ".flv"),  # 640x360
  26         (18, ".mp4"),  # 480x360
  27         (5,  ".flv"),  # 400x240
  28         (17, ".3gp"),  # 176x144
  29 ]
  30
  31
  32 def print_form(url="", msg=""):
  33         script_url = "http://%s%s" % (os.environ["HTTP_HOST"], os.environ["REQUEST_URI"])
  34         print "Content-Type: application/xhtml+xml\r\n\r\n"
  35         print """
  36 <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd">
  37 <html xmlns="http://www.w3.org/1999/xhtml">
  38 <head>
  39         <title>delx.net.au - YouTube Scraper</title>
  40         <link rel="stylesheet" type="text/css" href="/style.css"/>
  41         <style type="text/css">
  42                 input[type="text"] {
  43                         width: 100%;
  44                 }
  45                 .error {
  46                         color: red;
  47                 }
  48         </style>
  49 </head>
  50 <body>
  51         <h1>delx.net.au - YouTube Scraper</h1>
  52         {0}
  53         <form action="" method="get">
  54         <p>This page will let you easily download YouTube videos to watch offline. It
  55         will automatically grab the highest quality version.</p>
  56         <div><input type="text" name="url" value="{1}"/></div>
  57         <div><input type="submit" value="Download!"/></div>
  58         </form>
  59         <p>Tip! Use this bookmarklet: <a href="javascript:(function(){window.location='{2}?url='+escape(location);})()">YouTube Download</a>
  60         to easily download videos. Right-click the link and add it to bookmarks,
  61         then when you're looking at a YouTube page select that bookmark from your
  62         browser's bookmarks menu to download the video straight away.</p>
  63 </body>
  64 </html>
  65 """.replace("{0}", msg).replace("{1}", url).replace("{2}", script_url)
  66
  67 def parse_url(url):
  68         f = urllib.urlopen(url)
  69         doc = document_fromstring(f.read())
  70         f.close()
  71         return doc
  72
  73 def get_video_url(doc):
  74         unavailable = doc.xpath("//div[@id='unavailable-message']/text()")
  75         if unavailable:
  76                 raise VideoUnavailable(unavailable[0].strip())
  77         embed = doc.xpath("//embed")[0]
  78         flashvars = embed.attrib["flashvars"]
  79         flashvars = cgi.parse_qs(flashvars)
  80         fmt_url_map = {}
  81         for url_desc in flashvars["url_encoded_fmt_stream_map"][0].split(","):
  82                 url_desc_map = cgi.parse_qs(url_desc)
  83                 key = int(url_desc_map["itag"][0])
  84                 fmt_url_map[key] = url_desc_map["url"][0]
  85         for fmt, extension in fmt_quality:
  86                 try:
  87                         video_url = fmt_url_map[fmt]
  88                         break
  89                 except KeyError:
  90                         continue
  91         else:
  92                 return None, None, None
  93
  94         title = doc.xpath("/html/head/title/text()")[0]
  95         title = re.sub("\s+", " ", title.strip())
  96         valid_chars = frozenset("-_.() abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789")
  97         filename = "".join(c for c in title.encode("ascii", "ignore") if c in valid_chars)
  98         filename += extension
  99
 100         return video_url, filename
 101
 102 def cgimain():
 103         args = cgi.parse()
 104         try:
 105                 url = args["url"][0]
 106         except:
 107                 print_form(url="http://www.youtube.com/watch?v=FOOBAR")
 108                 return
 109
 110         try:
 111                 doc = parse_url(url)
 112                 video_url, filename = get_video_url(doc)
 113                 data = urllib.urlopen(video_url)
 114                 httpinfo = data.info()
 115                 sys.stdout.write("Content-Disposition: attachment; filename=\"%s\"\r\n" % filename)
 116                 sys.stdout.write("Content-Length: %s\r\n" % httpinfo.getheader("Content-Length"))
 117                 sys.stdout.write("\r\n")
 118                 shutil.copyfileobj(data, sys.stdout)
 119                 data.close()
 120         except VideoUnavailable, e:
 121                 print_form(
 122                         url=url,
 123                         msg="<p class='error'>Sorry, there was an error: %s</p>" % cgi.escape(e.message)
 124                 )
 125         except Exception, e:
 126                 print_form(
 127                         url=url,
 128                         msg="<p class='error'>Sorry, there was an error. Check your URL?</p>"
 129                 )
 130                 return
 131
 132 def main():
 133         try:
 134                 url = sys.argv[1]
 135         except:
 136                 print >>sys.stderr, "Usage: %s http://youtube.com/watch?v=FOOBAR" % sys.argv[0]
 137                 sys.exit(1)
 138         doc = parse_url(url)
 139         video_url, filename = get_video_url(doc)
 140         data = urllib.urlopen(video_url)
 141         outfile = open(filename, "w")
 142         shutil.copyfileobj(data, outfile)
 143         data.close()
 144         outfile.close()
 145
 146
 147 if __name__ == "__main__":
 148         if os.environ.has_key("SCRIPT_NAME"):
 149                 cgimain()
 150         else:
 151                 main()
 152