CGAL users discussion list

["Sebastien Loriot (GeometryFactory)" <>]

I confirm that even with exact predicates the algorithm is looping.
Here is a smaller example:
  myPoints.push_back(Point_2(0, 0));
  myPoints.push_back(Point_2(0, 1));
  myPoints.push_back(Point_2(1, 1));
  myPoints.push_back(Point_2(1, 0));
  myPoints.push_back(Point_2(0.5, 0));
  myPoints.push_back(Point_2(0.25, 0));
  myPoints.push_back(Point_2(0.15, 0));

the bug appears when several points are aligned along x=cst or y=cst.


Sebastien.


On 06/23/2014 09:35 AM, adrian wrote:
> I am using random_polygon_2() in a bigger project to calculate a polygon over
> a set of points. Now the program sometimes seems to get stuck in
> random_polygon_2 and never returns. In the task manager I see that one onf
> the CPU cores is busy. To me this looks like an bug in CGAL, as for other
> sets of points it works perfectly fine. Can anybody confirm the problem or
> propose a solution?
> For smaller sets of points changing the order of points within the list
> solved the issue, but that's not a real solution because normally the points
> are generated by the program.
> I am running the code on Windows 7, 64 bit (compiled for 64 bit) in Visual
> Studio 2008.
>
> The coordinates are normally generated in the program, but for simplicity I
> made an example where the coordinates are hard coded:
>
> #include <CGAL/Cartesian.h>
> #include <CGAL/Polygon_2.h>
> #include <CGAL/random_polygon_2.h>
>
> typedef CGAL::Cartesian<double> KP;
> typedef KP::Point_2 Point_2;
> typedef CGAL::Polygon_2<KP> Polygon_2;
>
> myFunction()
> {
>         std::list<Point_2> myPoints;
>         Polygon_2 myPolygon;
>
>         myPoints.push_back(Point_2(1439.24, -90.5695));
>         myPoints.push_back(Point_2(1439.24, -90.7099));
>         myPoints.push_back(Point_2(1439.25, -90.4293));
>         myPoints.push_back(Point_2(1439.38, -89.2577));
>         myPoints.push_back(Point_2(1439.24, -90.7242));
>         myPoints.push_back(Point_2(1439.24, -90.5695));
>         myPoints.push_back(Point_2(1441.88, -90.5439));
>         myPoints.push_back(Point_2(1439.24, -90.6255));
>         myPoints.push_back(Point_2(1439.24, -90.6351));
>         myPoints.push_back(Point_2(1439.24, -90.5933));
>         myPoints.push_back(Point_2(1441.59, -91.7365));
>         myPoints.push_back(Point_2(1441.8, -91.3991));
>         myPoints.push_back(Point_2(1441.64, -91.7477));
>         myPoints.push_back(Point_2(1441.8, -91.3991));
>         myPoints.push_back(Point_2(1439.24, -90.7019));
>         myPoints.push_back(Point_2(1439.24, -90.7019));
>         myPoints.push_back(Point_2(1439.24, -90.6819));
>         myPoints.push_back(Point_2(1439.24, -90.6769));
>         myPoints.push_back(Point_2(1439.24, -90.6748));
>         myPoints.push_back(Point_2(1439.24, -90.6709));
>         myPoints.push_back(Point_2(1439.24, -90.67));
>         myPoints.push_back(Point_2(1439.24, -90.6678));
>         myPoints.push_back(Point_2(1439.24, -90.666));
>         myPoints.push_back(Point_2(1439.24, -90.6634));
>         myPoints.push_back(Point_2(1439.24, -90.6362));
>         myPoints.push_back(Point_2(1439.24, -90.6324));
>         myPoints.push_back(Point_2(1439.24, -90.6299));
>         myPoints.push_back(Point_2(1439.24, -90.5995));
>         myPoints.push_back(Point_2(1439.24, -90.5978));
>         myPoints.push_back(Point_2(1439.54, -89.2892));
>         myPoints.push_back(Point_2(1441.88, -90.5439));
>         myPoints.push_back(Point_2(1441.64, -91.7477));
>
>         std::cout << "Calculating polygon" << std::endl;
>         CGAL::random_polygon_2(myPoints.size(), std::back_inserter(myPolygon),
> myPoints.begin());
>         std::cout << "Done calculating polygon" << std::endl;
> }
>
>
>
> --
> View this message in context: 
> http://cgal-discuss.949826.n4.nabble.com/random-polygon-2-gets-stuck-possible-CGAL-bug-tp4659470.html
> Sent from the cgal-discuss mailing list archive at Nabble.com.