Preliminary project structuring and Networking

[Berin Loritsch <bl...@d-haven.org>]

I started the process of the preliminary bagging and tagging of what we 
have--focusing
primarily on new stuff that is not part of the current release.

Trustin, I did not move the MINA stuff to the new network location as I 
figured it would
be rude to move what you are working on out from underneath you.  The 
place to move it
is created, and a first cut at the new protocol API layer is done.

The protocol-api has a first version done and tagged--but it is not 
final by any means.
Since now there is some code to try compiling against and play with 
please let me know
any feedback.  Also, keep in mind that this is for SEDANG and MINA, and 
not the
current crop of stuff.

The protocol-api is designed for asynchronous IO, and identifying 
clients via the socket
connection.  As far as the network layer is concerned that is the best 
that can be done--
anything beyond that is protocol specific.  To make it easier to manage 
the process a
bit, I included a way for the network layer to respond to protocol 
initiated events without
knowing anything specific about the network layer.  I did try to be 
generous with the
comments, esp. since almost everything is interfaces.

Please let me know any comments so that we can verify what is there will 
work for
the foreseeable future.  If not, we can refine it as necessary.

-- 

"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."
                - Rich Cook

Re: Preliminary project structuring and Networking

[Trustin Lee <tr...@gmail.com>]

Hi,

> Trustin, I did not move the MINA stuff to the new network location as I
> figured it would
> be rude to move what you are working on out from underneath you.  The
> place to move it
> is created, and a first cut at the new protocol API layer is done.

Ah, that's OK.  I'll move there. :)

> The protocol-api has a first version done and tagged--but it is not
> final by any means.
> Since now there is some code to try compiling against and play with
> please let me know
> any feedback.  Also, keep in mind that this is for SEDANG and MINA, and
> not the
> current crop of stuff.

> The protocol-api is designed for asynchronous IO, and identifying
> clients via the socket
> connection.  As far as the network layer is concerned that is the best
> that can be done--
> anything beyond that is protocol specific.  To make it easier to manage
> the process a
> bit, I included a way for the network layer to respond to protocol
> initiated events without
> knowing anything specific about the network layer.  I did try to be
> generous with the
> comments, esp. since almost everything is interfaces.

It is interesting that the API looks like using callback heavily. 
I'll be looking forward the next commit.

-- 
what we call human nature is actually human habit
--
http://gleamynode.net/

Re: Preliminary project structuring and Networking

[Alex Karasulu <ao...@bellsouth.net>]

Berin Loritsch wrote:

> I started the process of the preliminary bagging and tagging of what 
> we have--focusing
> primarily on new stuff that is not part of the current release.
>
Excellent I've been watching.  Good stuff going in there. 

>
> The protocol-api has a first version done and tagged--but it is not 
> final by any means.
> Since now there is some code to try compiling against and play with 
> please let me know
> any feedback.  Also, keep in mind that this is for SEDANG and MINA, 
> and not the
> current crop of stuff.
>
Yep I'll be taking a look soon.  Trying to wrap up with present release 
details.  I'm ready to do something more worth while like this.

> The protocol-api is designed for asynchronous IO, and identifying 
> clients via the socket
> connection.  As far as the network layer is concerned that is the best 
> that can be done--
> anything beyond that is protocol specific.  To make it easier to 
> manage the process a
> bit, I included a way for the network layer to respond to protocol 
> initiated events without
> knowing anything specific about the network layer.  I did try to be 
> generous with the
> comments, esp. since almost everything is interfaces.

> Please let me know any comments so that we can verify what is there 
> will work for
> the foreseeable future.  If not, we can refine it as necessary.

This is great I can't wait.  Expect to hear from me soon ....

Alex