1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
|
#!/usr/bin/python -t
import pwd, os, sys, logging, logging.handlers, string
import asynchat, asyncore, socket, httplib, urlparse
try:
import cStringIO as StringIO
except ImportError:
import StringIO
endpoints = {
{'host':'10.1.0.1', 'port':8080, 'speed':220, 'name':'Proxy 10.1'},
{'host':'10.2.2.11', 'port':8081, 'speed':340, 'name':'Proxy 10.2'},
{'host':'10.3.0.99', 'port':8080, 'speed':340, 'name':'Proxy 10.3'},
}
class HTTPResponseProducer(object):
def __init__(self, resp, amt=512):
self.resp = resp
self.amt = amt
def more(self):
return self.resp.read(self.amt)
class HTTPChannel(asynchat.async_chat):
def __init__(self, server, sock, addr):
asynchat.async_chat.__init__(self, sock)
self.server = server
self.set_terminator("\r\n\r\n")
self.request = None
self.data = StringIO.StringIO()
def collect_incoming_data(self, data):
self.data.write(data)
if self.data.tell() > 16384:
self.close_when_done()
def found_terminator(self):
if not self.request:
# parse http header
self.data.seek(0)
self.request = string.split(self.data.readline(), None, 2)
if len(self.request) != 3:
# badly formed request; just shut down
self.close_when_done()
else:
self.server.handle_request(self, self.request[0], self.request[1])
self.close_when_done()
else:
pass # ignore body data, for now
class HTTPProxyServer(asyncore.dispatcher):
def __init__(self):
asyncore.dispatcher.__init__(self)
self.port = 8080
self.create_socket(socket.AF_INET, socket.SOCK_STREAM)
self.set_reuse_addr()
self.bind(("", 8080))
self.listen(5)
def handle_accept(self):
conn, addr = self.accept()
HTTPChannel(self, conn, addr)
def handle_request(self, channel, method, path):
url = urlparse.urlparse(path)
print method, path
if method != "GET" or url.query != "":
#do not handle non-GET or GET with Query (?foo=bla) requests
return self._bypass_request(channel, method, url)
#check for content-length header with a HEAD request
conn = httplib.HTTPConnection(url.hostname, url.port or 80)
conn.request("HEAD", url.path)
resp = conn.getresponse()
content_length = filter(lambda it: it[0] == "content-length", resp.getheaders())
if len( content_length ) == 0:
# no content length given, bypass this request
return self._bypass_request(channel, method, url)
else:
content_length = content_length[0][1]
if content_length < 524288:
# do not handle requests smaller than 512kb
return self._bypass_request(channel, method, url)
print "Content-Length: %s" % (content_length)
# XXX an dieser stelle muss de request aufgeteilt werden
return self._bypass_request(channel, method, url)
#print "do some magic for " +str(url)
#channel.push("HTTP/1.0 200 OK\r\nX-Proxy: Magicproxy (request handled in boost mode)\r\n")
#channel.close_when_done()
def _bypass_request(self, channel, method, url):
#XXX hier sollte nicht proxy gespielt werden sondern
#die daten 1-zu-1 durchgereicht werden.
#Weiterhin sollte sichergestellt werden, dass die requests
#zu Host X1 immer über Proxy Y1 geroutet werden
# etwa proxy=proxies[ stuff(hostname) % len(proxies) ]
conn = httplib.HTTPConnection(url.hostname, url.port or 80)
conn.request(method, url.path)
resp = conn.getresponse()
channel.push("HTTP/1.0 200 OK\r\nX-Proxy: Magicproxy (request handled in standard mode)\r\n")
channel.push( "\r\n".join(map(lambda k: "%s: %s" % (k[0],k[1]), resp.getheaders())) )
channel.push("\r\n\r\n")
channel.push_with_producer( HTTPResponseProducer(resp) )
channel.close_when_done()
if __name__ == "__main__":
proxy = HTTPProxyServer()
asyncore.loop()
|