Hi Luc, I believe this issue is no longer relevant as computeLocalToWorldTransformFromXYZ() has been since re-written to address another issue, and no longer has any divisions. An update to svn/trunk will provide these changes.
For future reference, the place to report to bugs is on the osg-users mailing list, as osg-submissions is just for code submissions. Cheers, Robert. On Wed, Aug 12, 2009 at 10:21 AM, Luc Claustres<[email protected]> wrote: > Hello, > > In OpenSceneGraph-2.9.5 in the file include/osg/CoordinateSystemNode > inside the EllipsoidModel::computeLocalToWorldTransformFromXYZ() method > the inverse length of a vector is computed without testing if this > length is zero, leading to a division by zero : > > double length_XY = sqrt(X*X + Y*Y); > double inverse_length_XY = 1.0/length_XY; > > I think this occurs at the north pole point, but even if orientation may > not > be consistent it should return a valid matrix, don't it ? > > ------------------------------------------ > Luc Claustres - Project Manager > VEGA Technologies > Parc Technologique du Canal > 12 avenue de l'Europe > Villa San Diego > 31520 RAMONVILLE-SAINT-AGNE > FRANCE > > Tel : +33 (0)5.67.77.19.74 > Fax : +33 (0)5.61.53.42.12 > Mobile : +33 (0)6.15.34.01.50 > Email : [email protected] > Web : www.vega-group.com > ------------------------------------------ > > ______________________________________________________________________ > This email has been scanned by the MessageLabs Email Security System. > For more information please visit http://www.messagelabs.com/email > ______________________________________________________________________ > _______________________________________________ > osg-submissions mailing list > [email protected] > http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org > _______________________________________________ osg-submissions mailing list [email protected] http://lists.openscenegraph.org/listinfo.cgi/osg-submissions-openscenegraph.org
