[Bf-python] RE: action / NLA api

[Toni Alatalo]

On Monday 14 November 2005 15:26, Gilbert, Joseph T. wrote:
> NLA strips are not exposed. If you want to bind properties to them you
> need to expose them and define it's relationship to BPy_Action.
> Actions are not about armatures anymore, yes. We can move them to
> Blender.NLA eventually.

i've been thinking this a bit more, and this is the current conclusion:
move Blender.Armature.NLA to Blender.NLA, but *also* have a separate 
Blender.Action module/type. rationale below .. i guess i should draft this on 
wiki and then start implementing while listening to comments from you people.

13:15 < antont> .. now actions are gotten via Armature.NLA
13:15 < antont> but now that any object can have actions they dont belong to
                under Armature anymore
13:15 < antont> we need to change the API somehow
13:15 < antont> i'm a bit confused..
13:16 < antont> actions exist independently of course
13:16 < antont> and e.g. when exporting them to a game,
13:16 < antont> the NLA is not used at all
13:16 < antont> so it would not actually make sense to have
                Blender.Armature.NLA moved to Blender.NLA as a way to get
                actions .. or would it?
13:17 < antont> perhaps we should have Blender.Action and Blender.NLA
                separately - actions can be used without knowing what NLA is,
                and NLA would have strips
13:17 < broken> IMO, actions can work in conjunction with NLA, but they aren't
                really a subset of it
13:17 < antont> right
13:17 < broken> yes
13:17 < broken> and strips != actions
13:17 < antont> yep
13:17 < broken> strips are modified representatiosn of thm
13:17 < broken> mm

~Toni

>
> -----Original Message-----
> From: bf-python-bounces at projects.blender.org
> [mailto:bf-python-bounces at projects.blender.org] On Behalf Of Toni
> Alatalo
> Sent: Sunday, November 13, 2005 1:06 PM
> To: Blender Foundation Python list
> Subject: Re: [Bf-python] RE: action / NLA api
>
> On Friday 11 November 2005 16:57, Gilbert, Joseph wrote:
> > The object is called BPy_Action that we are talking about (same thing
>
> as
>
> > action strip). I don't like the word strip because it has more
> > connotations than Action.
>
> strips are different from actions.
>
> i first thought the matching would be somehow done with strips, not
> modifying
> the actual actions themselves.
>
> now i dont know how the implementation works exactly: is the plan to
> somehow
> offset the use of the action by a strip, and if yes, how? or modify the
> action itself?
>
> > unique names. I would not use BPy_Bone objects as input.
>
> hm .. i usually think that object references are better than name
> strings. are
> you referring to some implementational difficulties in doing that?
>
> > It might be better to extract the ipo channels from an Action and mark
> > one as a matchbone or however.
> > Action.ipochannels['bone1'].setMatchbone().
>
> perhaps. except that here Roland Hess says something contradicting:
> > The match bone is a property of the Action Strip. It uses the already
>
> so the matching is not about actions per se, but about how they are in
> NLA via
> the strips. or?
>
> btw, now it is official: object actions in bf-blender cvs, so actions
> are not
> about armatures anymore. i guess we should still work the 2.4 api to
> match
> that somehow.
>
> a bit cluelessly still,
> ~Toni
> _______________________________________________
> Bf-python mailing list
> Bf-python at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-python
> _______________________________________________
> Bf-python mailing list
> Bf-python at projects.blender.org
> http://projects.blender.org/mailman/listinfo/bf-python