Closed GoogleCodeExporter closed 9 years ago
[deleted comment]
I had just been implementing the algorithms inside of the robot and using a
switch to
go between. It would take a little doing to separate them, but if it's
necessary
then I'll do it. It does seem that it'd be easier that way if we implement the
scripting.
I'd suggest the robot model holding the algorithms instead of the algorithms
holding
the robot model because the view is displaying the robot and the robot is
affected by
the algorithm.
Is there a particular reason why you're making all of these model interfaces and
making the model classes inherit from them?
Original comment by frey...@gmail.com
on 30 Apr 2009 at 7:16
Attachments:
I can't think if any reasons at the moment that would make it necessary to
split up
the algorithms.
Holding the AI algorithm (a controller) inside the model I think would go
against the
MVC rules.
The model interface allows the actual model implementation to be changed
without the
views being none the wiser. Which of course I did when I switched from using
the stub
model to your model withing changing any code in the view.
Also the model and AI interactions need to be simple enough to work from the
scripting panel. Which I have working with Python now, it's pretty neat.
Original comment by lukelast@gmail.com
on 30 Apr 2009 at 8:14
Original comment by lukelast@gmail.com
on 9 Jun 2009 at 11:49
Original issue reported on code.google.com by
lukelast@gmail.com
on 29 Apr 2009 at 10:21