bug in http.py caused by IE

[radix]

I've been reviewing the http.py code. It seems there's a bug
in the
following, when there is an IE connect.

def lineReceived(self, line):
if self.__first_line:
# if this connection is not persistent, drop any data which
# the client (illegally) sent after the last request.
if not self.persistent:
self.dataReceived = self.lineReceived = lambda *args: None
return

# create a new Request object
request = self.requestFactory(self, len(self.requests))
self.requests.append(request)

# IE sends an extraneous empty line (\r\n) after a POST
request;
# eat up such a line, but only ONCE
if not line and self.__first_line == 1:
self.__first_line = 2
return


As written, IE with the bug will create two Request objects.

Here's a bugfix, which swaps the last two sections

def lineReceived(self, line):
if self.__first_line:
# if this connection is not persistent, drop any data which
# the client (illegally) sent after the last request.
if not self.persistent:
self.dataReceived = self.lineReceived = lambda *args: None
return

# IE sends an extraneous empty line (\r\n) after a POST
request;
# eat up such a line, but only ONCE
if not line and self.__first_line == 1:
self.__first_line = 2
return

# create a new Request object
request = self.requestFactory(self, len(self.requests))
self.requests.append(request)


Andrew

[radix]

implemented your fix