[Kde-games-devel] Re: KPoker cleanup phase 1 finished

[Inge Wallin]

On Wed, 1 Sep 2004, Andreas Beckermann wrote:

> On Tuesday 31 August 2004 22:21, Inge Wallin wrote:
> > After I am satisfied that I have introduced no new bugs, I will remove the
> > old code that is still there and the trace messages and backport this to
> > KDE_3_3_BRANCH as there are several bugs in bugzilla that were fixed by
> > the new code.
>
> I would propose to wait with the backport to BRANCH until you are sure that
> you didn't introduce new bugs. However that's your decision.

I am pretty sure at this stage.  I have tested it pretty extensively, and
have only managed to find a few small bugs that I fixed immediately.  Not
100%, of course, but you can never be 100% sure.

> On old code: in my opinion there is no need to keep old code around at all.
> Using cvs diff or cvs up -r we can easily see what you changed anyway.

Right, I agree.  I will remove the old stuff tonight.

> > But after that there are still a lot of things to do:
> >
> > * Separate the poker game code from the view code.
> >
> > * Make the AI possible to use in a game with more than 2 players.
> >
> > * Port KPoker to use KPlayer and KGame so that we can have network games.
>
> This is the point where I would like you to stop.
> It's cool that you fix the bugs and cleaned the code up, but I think it
> doesn't make much sense if we have 2 rewrite branches around (for porting to
> network and AI support you will have to rewrite nearly everything).

First, I am not sure what you mean here.  Do you mean stop before those
items of after?  I assume you mean before.

Second, I am pretty sure that you overestimate the difficulties here.
Ok, I don't know how difficult the port to KGame is, but regarding the AI,
that would surely be pretty small work.  My only aim in this first phase
would be so that you as a person could play against several computer
players...

> My KGame port is basically finished (giving AI support for free - you just
> need to write an AI input), so I don't think we win anything if you redo
> everything, too.

... and third, I get a bit curious here.  What are your short and long
term plans?  I really hope that you don't plan to throw out all the old
code and just replace it with your new stuff?  The old code used to be
really ugly, but it is shaping up quickly, and it actually works.

I suspect that throwing out all that and replacing it with new, untested
stuff will lead to more work and less stability.  I hope that I have shown
that working with an existing code base need not be very difficult even
when you start at a low level.

The bottom line here is that I hope to be able to persuade you to take
your new code and put it into the current program.

You mentioned earlier that you are willing to show the new stuff even
though it is not finished yet.  Why don't you do that, and I can rewrite
parts of the current KPoker to "pave the way" as it were for the new
and improved parts of the engine.

> Hehe :)
> I don't care too much about the CVS version anymore. The code in CVS is just
> plain _bad_. Or at least was, before you changed things. I will need to take
> a closer look at your changed when I get a chance to work on my own version
> again, but that's not a big problem.
>
> However as said above - I would really dislike to see a rewrite by you, as my
> own rewrite is already mostly done concerning game logic (it's the UI that is
> in an experimental state).

So, let's work within the current GUI and improve the game logic from the
inside.  I have a number of ideas on how the current GUI can be improved,
but I don't want to do everything simultaneously.  First the game logic /
AI engine, then the GUI.  Besides, I want to discuss it with you and hear
your idea before I take any action.

> CU
> Andi

	-Inge


Inge Wallin               | Thus spake the master programmer:               |
                          |      "After three days without programming,     |
inge at lysator.liu.se       |       life becomes meaningless."                |
                          | Geoffrey James: The Tao of Programming.         |