Andreas Fabri
Andreas Fabri
@sloriot had proposed to also add a point property map
I had thought that the vertices of the decomposition are all input vertices. Is the current solution (intersection points between voxels and triangles) also what is done in the original...
So when using `Epick` things can go wrong with intersection points computed. Do you make sanity checks, .e.g. the intersection point is inside the dilated voxel ?
Maybe even manipulators would be practical, but one then has the problem to distinguish from a color information to write.
Warning in testsuite [CGAL-6.2-Ic-48CGAL-6.2-Ic-48](https://cgal.geometryfactory.com/CGAL/testsuite/CGAL-6.2-Ic-48/Stream_support_Examples/TestReport_Christo_MSVC-2022-Community-Release.gz) ``` Color_ostream.h(149,30): warning C4996: 'getenv': This function or variable may be unsafe. Consider using _dupenv_s instead. To disable deprecation, use _CRT_SECURE_NO_WARNINGS. See online help for details....
Is it necessary to document all classes? Isn't [`Basic_indenting_streambuf`](https://cgal.geometryfactory.com/CGAL/Manual_doxygen_test/CGAL-6.2-Ic-49/output2/Stream_support/classCGAL_1_1IO_1_1Basic__indenting__streambuf.html) juxst an implementation detail?
I would document what is needed from a user perspective, and I think as a user I would not need the stream buffer class.
Fails in 50% of today's testsuites https://cgal.geometryfactory.com/CGAL/testsuite/summary-6.2-Ic-40.html?package=STL_Extension_Examples
Segfault in [CGAL-6.2-Ic-41](https://cgal.geometryfactory.com/CGAL/testsuite/CGAL-6.2-Ic-41/STL_Extension_Examples/TestReport_lrineau_Debian-stable-Release.gz)