Is it a bug on calculating Bnd_Box of an untessellated bspline face ?

Hi All,

I have a face which is a trimmed bspline surface. When the face is not tessellated I get a very large bounding box, and when it's tessellated I get the correct smaller bounding box. I think it's a bug BndLib_AddSurface, the code is:

if (Abs(UMin-S.FirstUParameter()) Abs(VMin-S.FirstVParameter()) Abs(UMax-S.LastUParameter ()) Abs(VMax-S.LastVParameter ())

the "if" sentence always give true back.

-Ding

Roman Lygin's picture

Yep, it's not really a bug but an effect of whether tesellation is available or not.

Check this post of April - http://opencascade.blogspot.com/2009/04/all-inclusive-bounding-box.html

Roman

Cauchy Ding's picture

Hi Romain,

Thank you so much for your reply. I can't log onto your blog for a long time, that's very very pity, maybe it has some IP restriction or our government block it.
In my system, when I import an iges file, I try to stitch them before they are loaded into opengl window. To do a stitching operation, I need a proper stitching tolerance. So I evaluate it from the bounding box of the raw shapes from file. Now, I modified the "if" condition and it works well now.
Thank you all the same.

-Ding