pengfork-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Pengfork-devel] Task making protocol independant of the core


From: Jean-Charles Salzeber
Subject: Re: [Pengfork-devel] Task making protocol independant of the core
Date: Wed, 4 Sep 2002 01:54:41 +0200
User-agent: Mutt/1.5.1i

On Tue, Sep 03, 2002 at 18:48, Nicolas Burrus wrote:
> I think this task is quite important, we should start enhancing tunneling 
> features, etc .. on a clean core loop, which has nothing to do with prot30.c 
> I think.

Yes, I think too... I've thinked of an engine/multiplexor from which we
can register callback when data is ready to send/receive.

> 
> I may do this at the end of this week, do you have suggestions or 
> requirements 
> for code organisation (filenames, functional division, etc ...) ?
> 
> We could maybe add a directory for AOL 3.0 protocol stuff, to have source 
> tree  
> cleaner ?

In fact, there is not much activity from me because I'm trying to see
difference between versions of AOL.
I do not have looked at AOL 7.0 protocol but for AOL >=3.0 & <=6.0 I can
say:
  * they all use the P3 protocol (the prot30.c source)
  * from aol 4.0, the login process seems crypted
  * aol 4.0 & 5.0 are identical
  * aol 6.0 use (at least) a very different init packet

The VMware idea that I posted few days ago work quite well...
Does someone have try to look at the AOL 7.0 protocol?
I'm waiting for an AOL 7.0 CD to start reverse engineering.

So most of the code is shared from all differents version but not all,
it is difficult to make a good directory structure.
perhaps a P3
           +aol30
           +aol40
           +aol60

But as I said some code is shared between versions.

For filenames&functions perhaps should we use p3_* instead of prot30_*
in prot30.c and just delete the prot30_* on others files.

I have much more knowledge on the protocol now.
The tunneling work better now, you can ping IP...
DNS, and others stuff don't work yet.
But it's not the real issue, actually the logon process do not work
well but I know why ;)

So I believe I will make many change to the protocol in some few days.

The biggest issue is actually to know what cipher aol use for the login
process that is my work for tomorrow ;)

> -- 
> Nicolas Burrus

JC

Attachment: pgpc8S0FsfDUT.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]