Getting KMess 2.1 into shape

[vdboor]

While I'm watching the development of KMess 2.1 from the sideline, it left me wondering with one question:

What do you need to get KMess 2.1 back in shape?

Basically: What things do we have to fix, so we can all run KMess from trunk again?
I think this step is important for many reasons, which I leave up to each one of you to imagine.

[fontknocker]

You know, I don't think there's really that much standing in the way although my usage needs will probably be different to the next person. For my every day stuff, like chatting, 2.1 is working (by working I mean it hasn't crashed massively on me...yet). As I use it I find things that need fixing here and there, but they're becoming less regular. Other things like DPs, file transfers, winks, etc are still out of action.

To solve the missing things, and to bring 2.1 back up to 2.0.x standard, I'd like to see the P2P stack fully integrated. Then we can get things like MsnObject download working again, file transfers, etc. Once we're back at the 2.0.x baseline we can go crazy and start value-adding

[vdboor]

This is good to hear! The sooner we can remove --iKnowThatTrunkIsBroken the better.

[fontknocker]

Anybody else with comments on this one? I know Valerio was/is working on the Debugging interface (which looks sexy, btw). Anybody know how the P2P integration is going?

[Valerio]

Nope. We *need* the P2P stack to work before we can even think about finishing to rewrite the missing components, unfortunately. In the feature tree, the P2P node that is still missing is root to an HUGE number of features...

Otherwise, the debug interface seems to be working correctly and AFAIK the library is fast and logging in and chatting work.. so the main bottleneck ATM is Gregg, who hasn't made himself heard in a while now (again) :\

[fontknocker]

.. so the main bottleneck ATM is Gregg, who hasn't made himself heard in a while now (again) :\

IMO we're going to have to learn to implement his library ourselves. He's proving to be particularly unreliable with such a core feature.

[Valerio]

IMO we're going to have to learn to implement his library ourselves. He's proving to be particularly unreliable with such a core feature.

Well, if anything it's got a good bit of documentation. I'm confident he'll find the time to finish it up though

[fontknocker]

Well, if anything it's got a good bit of documentation. I'm confident he'll find the time to finish it up though

Documentation is always good I don't doubt he'll find the time...it would just be reassuring if we saw him around #kmess or on the boards keeping us posted

PS: I got some internet access sorted out so I will be kicking around #kmess again

[vdboor]

Hi,

Without going through the finger-pointing menace, I think it's a really good idea to improve the API docs of the libmsntransport library. Would improve our understanding of the code too