Deleting just the primitives is not for the faint of heart - I think that was 
covered in the nuke-dev email thread from last year "Removing faces from GEO".  
Basically it's not a well tested part of the API.

When you delete the primitives in the vector (GeoInfo::cache.primitives) and 
clear it you also need to clear the overall vertex count 
(GeoInfo::cache.vertices) and any attributes that are bound to either 
Group_Primitive or Group_Vertices as they are also now out of date 
(GeoInfo::clear_attributes()).

Plus, any GeometryList references to the GeoInfo cache pointers are now out of 
date so you can try GeometryList::synchronize_objects() and see if that cleans 
it up the references.

Good luck,

-jonathan

On Jan 7, 2012, at 10:50 AM, Ivan Busquets wrote:

> Hi again,
> 
> Ok, so I managed to do the primitive-deletion thing by hacking the GeoInfo 
> header, and making clear_primitives() public instead of private.
> 
> I'd be interested to know what's in the clear_primitives() method, so I can 
> try and roll my own instead of hacking the GeoInfo header.
> 
> Any ideas?
> 
> Thanks,
> Ivan
> 
> On Sat, Jan 7, 2012 at 1:26 AM, Ivan Busquets <[email protected]> wrote:
> Hi,
> 
> Is there a way to delete all primitives from a GeoInfo (in a 
> SourceGeo-derived op) ?
> 
> So far I've tried to get the Cache for the GeoInfo and clear the primitives, 
> but any attempt at reading/writing "geoinfo.get_cache_pointer()->primitives" 
> is crashing on me.
> 
> Alternatively, I could do with deleting a single object from a GeometryList, 
> but I only see a GeometryList::delete_objects() method?
> Is there a way to delete a single one?
> 
> 
> Thanks,
> Ivan
> 
> 
> _______________________________________________
> Nuke-dev mailing list
> [email protected], http://forums.thefoundry.co.uk/
> http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev

_______________________________________________
Nuke-dev mailing list
[email protected], http://forums.thefoundry.co.uk/
http://support.thefoundry.co.uk/cgi-bin/mailman/listinfo/nuke-dev

Reply via email to