Nmap Development mailing list archives

Re: [NSE] http library cookie bug


From: Ryan Dewhurst <ryandewhurst () gmail com>
Date: Tue, 15 Mar 2011 21:46:48 +0000

Just playing devils advocate here.

What if the server expects cookie attributes to be in the same case it sent
them in?

On Tue, Mar 15, 2011 at 9:43 PM, Patrik Karlsson <patrik () cqure net> wrote:


On Mar 15, 2011, at 22:37 , David Fifield wrote:

On Tue, Mar 15, 2011 at 09:18:04PM +0100, John Bond wrote:
On 15 March 2011 20:33, Patrik Karlsson <patrik () cqure net> wrote:
Hi,

I think I ran into a bug when testing some code that handles cookies.
As far as I can understand from RFC 2109 "Attributes (names) (attr) are
case-insensitive."

When the http-library parses the cookie, each name value pair is
processed and a table field is dynamically created using the following code:
cookie[name] = value

I'm testing against two different servers. One returns the cookie path
value all lowercase and the second using a leading capital P.
I think it would make more sense if the cookie attributes were always
stored using lowercase table field names.
So I propose the following patch:

- cookie[name] = value
+ cookie[name:lower()] = value

Im not familiar with the specifics of the cookie rfc's however in http
1.1 paths should be considered case-sensitive[1] so i suspect it is
the same for the path part of the http cookie

Patrik is talking about the string "Path" or "path", not the path
itself. Patrik, I think your change makes sense and you can do it. It
will be just like the way we handle header fields.

I've commited the change.


David Fifield
_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/


//Patrik
--
Patrik Karlsson
http://www.cqure.net
http://www.twitter.com/nevdull77

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://seclists.org/nmap-dev/


Current thread: