Hey Tom,

There is a bit of research that has gone into the area of procedural geometry 
and parametric models, but there isn't really a complete high-level language 
(at least definitely not one that is cad-neutral) that serves the purpose you 
describe.

The closest that comes to mind is probably the GDML, which is a XML schema for 
geometry that CERN came up with a couple years ago.  It surprisingly includes 
primitives and booleans, but it doesn't even begin to touch procedures or 
parametric support.  Most of the procedural geometry research is very domain 
and application specific.

One of the TODO items I've had in mind [1] for several years now is to 
implement support in BRL-CAD for a procedural geometry "meta-primitive" where 
there are user-defined shapes with parameters.

 [1]: http://brlcad.org/~sean/ideas.html

That feature would allow users to define higher level objects and their 
parameters, similar to Cliff's tire tool, where a high level inputs are 
provided and those then cause more detailed geometry to be represented.  If I 
wanted a "medicine pill" primitive, for example, tha was defined simply by a 
simple radius, I'd write a little procedure that would create two spheres, 
connect them with a cone that was maybe twice the diameter, hollow it out with 
subtractions, and then fill the pill in with little spheres.

The idea is simply that the user specifies the parameters and writes the 
procedural logic.  BRL-CAD just have to have support for an evaluation script 
engine to understand how to represent, manage, invalidate, and reconstruct the 
geometry represented by a procedural geometry definition.

Cheers!
Sean


On Monday, January 25, 2010, at 09:28PM, "Tom Browder" <[email protected]> 
wrote:
>Over the years I've scripted BRL-CAD models (one was very large and
>detailed, others less so) with some fairly one-off Perl programs, some
>with a kind of geometry input language that was used eventually to
>produce another script that could be fed to mged.
>
>I find I am reinventing that wheel once again and don't want to waste
>the effort if there is a better way.
>
>I know others have used the same kinds of kludges, but is anyone aware
>of a serious effort toward some kind of high level, cad-neutral
>geometry description language as an interface for a filter of some
>kind that could output geometry in varying formats (kind of a geometry
>compiler)?  Would such a thing be of interest to other BRL-CAD users?
>
>Regards,
>
>-Tom
>
>Thomas M. Browder, Jr.
>Niceville, Florida
>USA
>
>------------------------------------------------------------------------------
>The Planet: dedicated and managed hosting, cloud storage, colocation
>Stay online with enterprise data centers and the best network in the business
>Choose flexible plans and management services without long-term contracts
>Personal 24x7 support from experience hosting pros just a phone call away.
>http://p.sf.net/sfu/theplanet-com
>_______________________________________________
>BRL-CAD Developer mailing list
>[email protected]
>https://lists.sourceforge.net/lists/listinfo/brlcad-devel
>
>

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
BRL-CAD Developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/brlcad-devel

Reply via email to