[Bf-committers] Anim system recode

[bjornmose]

Ton Roosendaal wrote:
> Hi,
> 
> (For those who don't have cvs mails, here's the cvs log!)
> 
> Result of 2 weeks of quiet coding work in Greece :)
> 
> Aim was to get a total refresh of the animation system. 
..
> - Depgraph is aware of all relationships, including meta balls,
>   constraints, bevelcurve, and so on
..
> TODO NOW;
> 
..

> - Depgraph code needs revision, cleanup, can be much faster!
>   (But, compliments for Jean-Luc, it works like a charm!)
..
compliments for Jean-Luc, yes, great work!
Hum, still i guess there seems to be a problem with 'action 
constraints'. /* i know things may get nasty there, since all bones /** 
even parents **/ may move in the action as well /** allowing this is 
allowing recursion --> dependancy circles, AKA infinite loops  **/ :) */
Well at least the rig given there --> 
http://www.elysiun.com/forum/viewtopic.php?t=31839&start=0
and mine using actions don't behave. /* if nozzy and me use stuff like 
that i'm almost sure bassam does too :) */
> -Ton-
> 
wheee that scientist is writing riddles again!

an again may be totally wrong since i needed to change code here to make 
  it compile on msvc6:

1. declaring the variable before code in block ..
		else if ELEM(G.obedit->type, OB_CURVE, OB_SURF) {
			Nurb *nu;
			DAG_object_flush_update(G.scene, G.obedit, OB_RECALC_DATA);  /* sets 
recalc flags */
			
			nu= editNurb.first;
			while(nu) {


2. commentig off the MT_assert since verify_boneptr(..) is not declared 
in the msvc6 world
float BL_ArmatureObject::GetBoneLength(Bone* bone) const
{
//	MT_assert(verify_boneptr((bArmature*) GetArmature(), bone) && "Bone 
is not part of this armature.");
	
	return (MT_Point3(bone->head) - MT_Point3(bone->tail)).length();
}

-BM-