[Bf-committers] Its time for the CSG transform code
joeeagar at prodigy.net
Wed Sep 22 23:06:28 CEST 2004
Martin Poirier wrote:
>What part of transform will you work with? The new
>code uses mostly atomic functioncs, so just tell me
>what part you'll be working with and I can tell you as
>much as I can (assuming that part is done already).
A CSGObject has a list of trees (should be branches, must change) which
each in turn have a list of nodes. Each node can be scaled/rot'd/grab'd
(e.g. they each have their own individual rot/scale/loc members). Also,
each node is linked to an Object containing a Mesh, so I may change it
so that the basic rot/loc/size attributes are pulled from that Object.
Each node can also have it's own matrix. I copied the matrix pointers
from MetaElem in DNA_meta.h, but I have almost no idea how to actually
assign a matrix to one.
>The wiki isn't even all of it and the part that's
>there is only the basics principles.
P.S.: Why isn't Link structures used for DNA subobject structs? The
whole thing of using struct [subobject] *next, *prev is extreamly confusing.
More information about the Bf-committers