[OpenSCAD] Discuss manifoldness, co-incident faces edges etc

nop head nop.head at gmail.com
Wed Nov 13 15:39:47 EST 2019


If the STL is manifold than you can find each edge is shared by exactly two
triangles so you can reconstruct the mesh without any change in topology.
There is no explicit topology but it is preserved. Once read into OpenSCAD
all explicit topology is lost anyway because it uses a polygon soup itself.
Polygon soups are fine for representing a mesh until you have self
intersections or edges shared by more than two facets.

On Wed, 13 Nov 2019 at 20:34, Carsten Arnholm <arnholm at arnholm.org> wrote:

> On 13.11.2019 21:05, nop head wrote:
> > There is no guesswork needed to read an STL if the geometry it contains
> > is manifold.
>
> I am sorry but this is nonsensical. See below.
>
> > If STL files had a rule to say the geometry they contained has to be
> > manifold and writers and readers enforced it there would not be a
> problem.
>
> How many times must it be explained that STL is a polygon soup without
> any topological information whatsoever, and hence any topology
> (manifoldness relates to topology only) on the reader side must be
> guessed? It has been stated many times already. It is a fact.
>
> > OpenSCAD doesn't reject bad STLs. It just feeds the soup to CGAL and
> > CGAL barfs if it is not manifold and produces and incomprehensible
> > exception.
>
> And in this process, someone takes a guess at what the topology is
> supposed to be.
>
> Carsten Arnholm
>
> _______________________________________________
> OpenSCAD mailing list
> Discuss at lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscad.org/pipermail/discuss_lists.openscad.org/attachments/20191113/30e8ccfe/attachment.html>


More information about the Discuss mailing list