> Variations were giving me a major headache with the recursive function
that now needs to stay on top of CurrentPos the whole time in and out of
variations. I will look into that. I was actually overthinking it at first
and messing it up, then i took out some code that i put in and thought i
had it working with a little tweak but i guess it needs some more attention
for subvariations(?) I will look at ticket

I wonder if the position being tied to a game will ever give us headaches.
This is the most novel thing you do.. And as positions are thrown around
everywhere in Scid, we'll have to keep an eye on this.
One thing comes to mind.. Games are always getting pushed and pulled of a
stack.
with sc_game_push. Just check this is not breaking anything.

> I haven't yet done any setup board stuff yet. I can get to that. Should
just be a matter of attaching a proper Game * to those positions wherever
they are made.

I can do that this weekend. Shouldnt be much work . Maybe just relaxing
Castling sanity checks. Handling the king +-2 will be harder i think, but i
should have time for it.

> Just curious, if you guys are interested in the work and i keep
producing, how would you guys handle it on the backend? Like make a branch
at r2569 and keep putting this stuff there and then merge later from time
to time? I can just keep a 2569 around for myself locally and make patches
from that to give you if that works.

I'm thinking just to keep it as a patch for a while, maybe posting a
windows beta binary.
Normally, merging patches isn't much of an issue nowadays.
If it is solid,  we can inline it as an #ifdef, and a build configuration
option,
and post separate binaries for Scidvs960 (or whatever).
Though it can handle normal chess... hard to say now whether it could ever
become the only default code.

Thanks
------------------------------------------------------------------------------
_______________________________________________
Scidvspc-users mailing list
Scidvspc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scidvspc-users

Reply via email to