CGAL users discussion list

[Martin Uhrin <>]

I've just realised where my error is.

For anyone in this situation in the future: I hadn't read the manual [1]
properly, specifically:

"Note
    The coordinates have a rational representation, i.e. the first d
elements of the iterator range are the numerators and the (d+1)-st element
is the common denominator. "

So indeed the (d+1)^th element contained the denominator that when applied
gives the correct answer.

Apologies for jumping the gun on posting.

-Martin

[1]
http://doc.cgal.org/latest/Polytope_distance_d/classCGAL_1_1Polytope__distance__d.html#a8f75d3a4cfec3ea036c88b8fe8c37381
Martin Uhrin wrote
> I'm getting unexpected results from Polytope_distance_d running a
> similar query to this post [1].  Specifically, I'm looking for the
> nearest point on a polygon to a given point on the unbounded side of
> it.  A reduced example:
> 
> #include 
> <iostream>
> #include &lt;CGAL/MP_Float.h&gt;
> #include &lt;CGAL/Exact_predicates_inexact_constructions_kernel.h&gt;
> #include &lt;CGAL/Polygon_2.h&gt;
> #include &lt;CGAL/Polytope_distance_d.h&gt;
> #include &lt;CGAL/Polytope_distance_d_traits_2.h&gt;
> 
> typedef CGAL::Exact_predicates_inexact_constructions_kernel K;
> typedef CGAL::MP_Float ET;
> typedef CGAL::Polytope_distance_d_traits_2< K, ET, K::RT> DistTraits;
> typedef CGAL::Polytope_distance_d< DistTraits> PolyDist;
> 
> int main()
> {
>   using namespace CGAL;
>   using namespace std;
>   typedef K::Point_2 Point;
> 
>   Polygon_2
> <K>
>  p;
>   p.push_back(Point(1.2, 3.4)); // Construct square
>   p.push_back(Point(1.1, 3.4));
>   p.push_back(Point(1.1, 3.5));
>   p.push_back(Point(1.2, 3.5));
> 
>   Point q(1.15735, 3.33821);
> 
>   PolyDist dist(p.vertices_begin(), p.vertices_end(), &q, (&q) + 1);
> 
>   cout << dist;
>   PolyDist::Coordinate_iterator it =
> dist.realizing_point_p_coordinates_begin();
>   cout << "Realising p: " << *it;
>   cout << " " << *++it << "\n";
> 
>   it = dist.realizing_point_q_coordinates_begin();
>   cout << "Realising q: " << *it;
>   cout << " " << *++it << "\n";
> 
>   return 0;
> }
> 
> Output:
> 
> martin@pico:~/temp/polytope_dist$ g++ poly_dist.cpp -lCGAL
> -lboost_thread -lmpfr -lpthread -lgmp -frounding-math -o dist
> martin@pico:~/temp/polytope_dist$ ./dist
> 4
> 1.2 3.4
> 1.1 3.4
> 1.1 3.5
> 1.2 3.5
> 1
> 1.15735 3.33821
> Realising p: 0.0115735 0.034
> Realising q: 0.0115735 0.0333821
> 
> Neither of these realising points seems to have any relation to my input
> points.
> 
> g++ (Ubuntu/Linaro 4.8.1-10ubuntu9) 4.8.1
> CGAL 4.3.
> 
> I've tried:
> * Exact construction kernel (i.e. fully exact Polytope_distance)
> * Swapping p & q order in constructor
> * Gmpfr ET type
> 
> All of which produce the same output.
> 
> Any help you can provide is much appreciated,
> -Martin
> 
> [1]
> http://cgal-discuss.949826.n4.nabble.com/Distance-from-a-Convex-hull-3-to-Point-3-tt4656683.html#a4656704
> 
> -- 
> Martin Uhrin                                                  Tel: +44
> 207 679 3466
> Department of Physics & Astronomy               Fax:+44 207 679 0595
> University College London                         

> martin.uhrin.10@.ac

> Gower St, London, WC1E 6BT, U.K.         http://www.cmmp.ucl.ac.uk
> 
> -- 
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://sympa.inria.fr/sympa/info/cgal-discuss


Martin Uhrin wrote
> I'm getting unexpected results from Polytope_distance_d running a
> similar query to this post [1].  Specifically, I'm looking for the
> nearest point on a polygon to a given point on the unbounded side of
> it.  A reduced example:
> 
> #include 
> <iostream>
> #include &lt;CGAL/MP_Float.h&gt;
> #include &lt;CGAL/Exact_predicates_inexact_constructions_kernel.h&gt;
> #include &lt;CGAL/Polygon_2.h&gt;
> #include &lt;CGAL/Polytope_distance_d.h&gt;
> #include &lt;CGAL/Polytope_distance_d_traits_2.h&gt;
> 
> typedef CGAL::Exact_predicates_inexact_constructions_kernel K;
> typedef CGAL::MP_Float ET;
> typedef CGAL::Polytope_distance_d_traits_2< K, ET, K::RT> DistTraits;
> typedef CGAL::Polytope_distance_d< DistTraits> PolyDist;
> 
> int main()
> {
>   using namespace CGAL;
>   using namespace std;
>   typedef K::Point_2 Point;
> 
>   Polygon_2
> <K>
>  p;
>   p.push_back(Point(1.2, 3.4)); // Construct square
>   p.push_back(Point(1.1, 3.4));
>   p.push_back(Point(1.1, 3.5));
>   p.push_back(Point(1.2, 3.5));
> 
>   Point q(1.15735, 3.33821);
> 
>   PolyDist dist(p.vertices_begin(), p.vertices_end(), &q, (&q) + 1);
> 
>   cout << dist;
>   PolyDist::Coordinate_iterator it =
> dist.realizing_point_p_coordinates_begin();
>   cout << "Realising p: " << *it;
>   cout << " " << *++it << "\n";
> 
>   it = dist.realizing_point_q_coordinates_begin();
>   cout << "Realising q: " << *it;
>   cout << " " << *++it << "\n";
> 
>   return 0;
> }
> 
> Output:
> 
> martin@pico:~/temp/polytope_dist$ g++ poly_dist.cpp -lCGAL
> -lboost_thread -lmpfr -lpthread -lgmp -frounding-math -o dist
> martin@pico:~/temp/polytope_dist$ ./dist
> 4
> 1.2 3.4
> 1.1 3.4
> 1.1 3.5
> 1.2 3.5
> 1
> 1.15735 3.33821
> Realising p: 0.0115735 0.034
> Realising q: 0.0115735 0.0333821
> 
> Neither of these realising points seems to have any relation to my input
> points.
> 
> g++ (Ubuntu/Linaro 4.8.1-10ubuntu9) 4.8.1
> CGAL 4.3.
> 
> I've tried:
> * Exact construction kernel (i.e. fully exact Polytope_distance)
> * Swapping p & q order in constructor
> * Gmpfr ET type
> 
> All of which produce the same output.
> 
> Any help you can provide is much appreciated,
> -Martin
> 
> [1]
> http://cgal-discuss.949826.n4.nabble.com/Distance-from-a-Convex-hull-3-to-Point-3-tt4656683.html#a4656704
> 
> -- 
> Martin Uhrin                                                  Tel: +44
> 207 679 3466
> Department of Physics & Astronomy               Fax:+44 207 679 0595
> University College London                         

> martin.uhrin.10@.ac

> Gower St, London, WC1E 6BT, U.K.         http://www.cmmp.ucl.ac.uk
> 
> -- 
> You are currently subscribed to cgal-discuss.
> To unsubscribe or access the archives, go to
> https://sympa.inria.fr/sympa/info/cgal-discuss





--
View this message in context: 
http://cgal-discuss.949826.n4.nabble.com/Unexpected-resulst-from-Polytope-distance-tp4658871p4658872.html
Sent from the cgal-discuss mailing list archive at Nabble.com.