CGAL users discussion list

[Andreas Fabri <>]

Hi Bob

one PR later I now understand where your initial problem comes from.
You commented refine_Delaunay_mesh_2().
This function internally does a traversal
as the mark_domain() function.
When not called there is no domain marked.

Best,

Andreas

ps.: The above mentioned PR is here    https://github.com/CGAL/cgal/pull/6412

On 3/14/2022 12:46 PM, Andreas Fabri wrote:

This should do it, although not yet elegant

https://gist.github.com/afabri/82e9680181eea814fb488f96ceca8c12

On 3/14/2022 12:09 PM, Bob Bill ( via cgal-discuss Mailing List) wrote:

          Thanks Sebastien and Andreas,
          
          
          I see what Andreas means, indeed if I modify my
            example accordingly and then call refine_Delaunay_mesh_2() I
            have the expected error " no member named 'set_in_domain'...".
          
          
          I tried to adapt it by adding typedef
                CGAL::Delaunay_mesh_face_base_2<K> Fb;  in
              place of  typedef
              CGAL::Constrained_triangulation_face_base_2<K,Fbb>  
               Fb, but I get an error "no member named 'info' in..." coming
            from the start->info(), so I don't really know
            how to adapt it in order to have both functions. 
          
          
          Should I only change the typedefs?
          
          
          
          
          Best,
          Bob

Il lunedì 14 marzo 2022, 08:59:26 CET, Andreas Fabri <> ha scritto:

Hello,

Note that write_VTU is documented in the  CGAL package 2D Conforming
Triangulations and Meshes.

When you call a function like refine_Delaunay_mesh()  you define the
domain (for example the interior
of your polygon, as a domain, otherwise the convex hull of the points is
the domain.

When you follow the example Sebastien just gave you  must adapt it, as
the Delaunay_mesh_face_base_2
class has a function to set the domain.

https://doc.cgal.org/latest/Mesh_2/classDelaunayMeshFaceBase__2.html#ac4a80317799961e7830b9690121d6bdc
Best,

Andreas


On 3/14/2022 8:40 AM, Sebastien Loriot ( via
cgal-discuss Mailing List) wrote:
> A triangulation of a point set is a triangulation of the convex hull
> of the points. If some constrain edges are added, you then have the
> guarantee that those edges will be present in the triangulation of the
> convex hull of your points.
>
> If your constraints define a closed shape then you need a flooding like
> algorithm to only pick the triangles that are inside the shape, like in
> this example:
>
> https://doc.cgal.org/latest/Triangulation_2/Triangulation_2_2polygon_triangulation_8cpp-example.html
>
>
> Best,
>
> Sebastien.
>
> On 3/13/22 02:46, Bob Bill ( via cgal-discuss
> Mailing List) wrote:
>> I am given the vertices of a polygon, and I need to mesh it. I am
>> following one of the first examples in the CDT package documentation.
>>
>> The code is the following:
>>
>> #include <CGAL/Exact_predicates_inexact_constructions_kernel.h> >> #include <CGAL/Constrained_Delaunay_triangulation_2.h> >> #include <CGAL/Delaunay_mesher_2.h> >> #include <CGAL/Delaunay_mesh_face_base_2.h>
>> #include <CGAL/Delaunay_mesh_size_criteria_2.h> >> #include <CGAL/IO/write_VTU.h> >>
>> #include <iostream>
>> typedef CGAL::Exact_predicates_inexact_constructions_kernel K; >> typedef CGAL::Triangulation_vertex_base_2<K> Vb; >> typedef CGAL::Delaunay_mesh_face_base_2<K> Fb; >> typedef CGAL::Triangulation_data_structure_2<Vb, Fb> Tds; >> typedef CGAL::Constrained_Delaunay_triangulation_2<K, Tds> CDT;
>> typedef CGAL::Delaunay_mesh_size_criteria_2<CDT> Criteria; >> typedef CDT::Vertex_handle Vertex_handle; >> typedef CDT::Point Point;
>>
>> int main()
>> {
>>    CDT cdt;
>>    Vertex_handle va = cdt.insert(Point(-1,-1)); >>    Vertex_handle vb = cdt.insert(Point(-1,-0.25714));
>>    Vertex_handle vc = cdt.insert(Point(-1.5,-1.4));
>>    Vertex_handle vd = cdt.insert(Point(1.3,-1.2));
>>    Vertex_handle ve = cdt.insert(Point(1.3,0.5));
>>    Vertex_handle vf = cdt.insert(Point(1, 0.45714));
>>    Vertex_handle vg = cdt.insert(Point(1, -1)); >>
>>    cdt.insert_constraint(va, vb); >>    cdt.insert_constraint(vb, vc); >>    cdt.insert_constraint(vc, vd); >>    cdt.insert_constraint(vd, ve); >>    cdt.insert_constraint(ve,vf); >>    cdt.insert_constraint(vf,vg); >>    cdt.insert_constraint(vg,va); >>    std::cout << "Number of vertices: " << cdt.number_of_vertices() << >> std::endl;
>>    std::cout << "Meshing the triangulation..." << std::endl;
>>    // CGAL::refine_Delaunay_mesh_2(cdt, Criteria(0.125, 0.5));
>>    std::cout << "Number of vertices: " << cdt.number_of_vertices() << >> std::endl;
>>
>>    std::ofstream output("intersected_mesh.vtu");
>>    CGAL::write_vtu(output, cdt); >>
>> //Show the vertices of the triangulation >>      for(auto it =
>> cdt.finite_faces_begin();it!=cdt.finite_faces_end();++it){ >>      std::cout << cdt.triangle(it) <<'\n';
>>    }
>> }
>>
>> If I look at the .vtu file, it seems to have the correct shape (see
>> attached image). However, after I printed the vertices for each
>> triangle, it appears that one element of the triangulation is also
>> the one with vertices
>> (-1, -0.25714)
>> (1, -1)
>> (1, 0.45714)
>>
>> which should not be there since is not inside that polygon. Isn't
>> this a bug? If not, how can I create a mesh for which this is
>> avoided? I mean, I'd like to have all the triangles inside the domain.
>>
>> Best regards,
>> Bob
>>
>> --
>> You are currently subscribed to cgal-discuss. >> To unsubscribe or access the archives, go to >> https://sympa.inria.fr/sympa/info/cgal-discuss >>
>
--
Andreas Fabri, PhD
Chief Officer, GeometryFactory
Editor, The CGAL Project

--
You are currently subscribed to cgal-discuss. To unsubscribe or access the archives, go to https://sympa.inria.fr/sympa/info/cgal-discuss

          -- 
          You are currently subscribed to cgal-discuss.
          To unsubscribe or access the archives, go to
          https://sympa.inria.fr/sympa/info/cgal-discuss
          
        
      
-- 
Andreas Fabri, PhD
Chief Officer, GeometryFactory
Editor, The CGAL Project
      
        -- 
        You are currently subscribed to cgal-discuss.
        To unsubscribe or access the archives, go to
        https://sympa.inria.fr/sympa/info/cgal-discuss
        
      
    
-- 
Andreas Fabri, PhD
Chief Officer, GeometryFactory
Editor, The CGAL Project