CGAL users discussion list

[Daniel Duque <>]

Hi,

I have had issues with precision at the stream before. I usually include lines like the third one when writing on a file:

  std::ofstream matrices;

  matrices.open("matrices.m");

  matrices.precision(15);

  matrices << "x = [" << endl;

  for(F_v_it vit=T.finite_vertices_begin();

      vit != T.finite_vertices_end();

      vit++)

    matrices << vit->point().x() << endl;

  matrices << "];" << endl;

This is part of a code which prints out stuff in matlab / octave format. The precision is there, it just doesn't get printed in full.

Best,

Daniel

On Thu, Nov 26, 2015 at 1:54 PM, Sebastien Loriot (GeometryFactory) <> wrote:

I don't think there is anything specific to CGAL according to the
typedefs you shown. Maybe first check that after inserting a vertex
the coordinates are what you expect. Also make sure the precision
of the stream is good enough.

Sebastien.

On 11/26/2015 01:29 PM, Pol Monsó Purtí Helimap wrote:

I am retrieving the coordinates x,y,z from a xyz file to compute a
delaunay. The values have double precision.

If I print or save the values of a point, x and y have no longer double
precision, but the z does. I've debugged the program and if I add a
watch _expression_ as *(double*)<address of the x> I get the original,
correct double value.

I've tried casting point().x() to *(double*) but of course x() returns a
double, not a double*. Somehow, the precision is lost when retrieved.

Any clue?

I am using

typedef CGAL::Exact_predicates_inexact_constructions_kernel K;
typedef CGAL::Projection_traits_xy_3<K>  Gt;

typedef CGAL::Triangulation_vertex_base_2<Gt> Vb;
typedef CGAL::Delaunay_mesh_face_base_2<Gt> Fb;
typedef CGAL::Triangulation_data_structure_2<Vb, Fb> Tds;
typedef CGAL::Constrained_Delaunay_triangulation_2<Gt, Tds> CDT;

Thanks!

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