mirror of https://github.com/CGAL/cgal
164 lines
6.2 KiB
C++
164 lines
6.2 KiB
C++
// test whether box queries are computed correctly for random data
|
|
//
|
|
// 1) generate list of query points using report_all
|
|
// 2) remove and check reported points from these list
|
|
// 3) check if no remaining points should have been reported
|
|
|
|
#include <CGAL/Simple_cartesian.h>
|
|
|
|
#include "Point_with_info.h"
|
|
|
|
#include <CGAL/Kd_tree.h>
|
|
#include <CGAL/Search_traits_2.h>
|
|
#include <CGAL/Search_traits_adapter.h>
|
|
#include <CGAL/Fuzzy_iso_box.h>
|
|
|
|
#include <CGAL/iterator.h>
|
|
#include <CGAL/point_generators_2.h>
|
|
|
|
#include <cassert>
|
|
#include <vector>
|
|
#include <iostream>
|
|
|
|
typedef CGAL::Simple_cartesian<double> K;
|
|
typedef K::FT FT;
|
|
typedef K::Point_2 Point;
|
|
typedef K::Vector_2 Vector;
|
|
typedef K::Iso_rectangle_2 Iso_rectangle;
|
|
|
|
typedef CGAL::Random_points_in_square_2<Point> Random_points_iterator;
|
|
typedef CGAL::Counting_iterator<Random_points_iterator> N_Random_points_iterator;
|
|
typedef CGAL::Search_traits_2<K> Traits;
|
|
typedef Point_with_info_helper<Point>::type Point_with_info;
|
|
typedef Point_property_map<Point> Ppmap;
|
|
typedef CGAL::Search_traits_adapter<Point_with_info,Ppmap,Traits> Traits_with_info;
|
|
|
|
template <class SearchTraits, class Tree>
|
|
void run_with_fuzziness(std::list<Point> all_points, const Tree& tree,
|
|
const Point& p, const Point& q, const FT fuzziness)
|
|
{
|
|
typedef CGAL::Fuzzy_iso_box<SearchTraits> Fuzzy_box;
|
|
|
|
tree.search(CGAL::Emptyset_iterator(), Fuzzy_box(p,q)); //test compilation when Point != Traits::Point_d
|
|
|
|
typename SearchTraits::Point_d pp(p);
|
|
typename SearchTraits::Point_d qq(q);
|
|
|
|
std::cout << "test with box: [" << p << " || " << q << "] and eps: " << fuzziness << "... ";
|
|
|
|
// approximate range searching
|
|
std::list<typename SearchTraits::Point_d> result;
|
|
Fuzzy_box approximate_range(pp, qq, fuzziness);
|
|
tree.search(std::back_inserter(result), approximate_range);
|
|
std::cout << result.size() << " hits... Verifying correctness...";
|
|
|
|
// test the results of the approximate query
|
|
Iso_rectangle inner_ic(p + fuzziness*Vector(1,1), q - fuzziness*Vector(1,1));
|
|
Iso_rectangle outer_ic(p - fuzziness*Vector(1,1), q + fuzziness*Vector(1,1));
|
|
|
|
// If the fuziness is greater than half of the largest dimension of the box,
|
|
// then the inner box does not exist
|
|
const FT max_box_edge_length = (std::max)(q[1] - p[1], q[0] - p[0]);
|
|
const bool is_inner_c_empty = (fuzziness > 0.5 * max_box_edge_length);
|
|
if(is_inner_c_empty)
|
|
std::cout << " (empty inner box)... ";
|
|
|
|
for (typename std::list<typename SearchTraits::Point_d>::iterator pt=result.begin(); (pt != result.end()); ++pt)
|
|
{
|
|
// a point can only be reported if it is in the outer box
|
|
bool is_correct = outer_ic.has_on_bounded_side(get_point(*pt)) || outer_ic.has_on_boundary(get_point(*pt));
|
|
if(!is_correct)
|
|
std::cout << get_point(*pt) << " should have not been reported" << std::endl;
|
|
assert(is_correct);
|
|
all_points.remove(get_point(*pt));
|
|
}
|
|
|
|
// nothing to test if the inner box is empty because everything is on the unbounded side
|
|
if(!is_inner_c_empty)
|
|
{
|
|
for (std::list<Point>::iterator pt=all_points.begin(); (pt != all_points.end()); ++pt)
|
|
{
|
|
// all points that have not been reported must be outside the inner box
|
|
bool is_correct = inner_ic.has_on_unbounded_side(*pt);
|
|
if(!is_correct)
|
|
std::cout << *pt << " should have been reported" << std::endl;
|
|
|
|
assert(is_correct);
|
|
}
|
|
}
|
|
|
|
std::cout << "done" << std::endl;
|
|
}
|
|
|
|
template <class SearchTraits>
|
|
void run(std::list<Point> all_points) // intentional copy
|
|
{
|
|
// Insert also the N points in the tree
|
|
CGAL::Kd_tree<SearchTraits> tree(
|
|
boost::make_transform_iterator(all_points.begin(), Create_point_with_info<typename SearchTraits::Point_d>()),
|
|
boost::make_transform_iterator(all_points.end(), Create_point_with_info<typename SearchTraits::Point_d>())
|
|
);
|
|
|
|
// bigger box
|
|
Point p0(-10., -10.);
|
|
Point q0( 10., 10.);
|
|
|
|
// a subset
|
|
Point p1(-CGAL_PI/10., -CGAL_PI/10.);
|
|
Point q1( CGAL_PI/10., CGAL_PI/10.);
|
|
|
|
// another subset
|
|
Point p2(0.1, 0.2);
|
|
Point q2(0.3, 0.4);
|
|
|
|
// degenerate
|
|
Point p3(0., 0.);
|
|
Point q3(0., 0.);
|
|
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p0, q0, 0. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p0, q0, 0.1 /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p0, q0, 1. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p0, q0, 10. /*fuzziness*/);
|
|
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p1, q1, 0. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p1, q1, 0.1 /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p1, q1, 1. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p1, q1, 10. /*fuzziness*/);
|
|
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p2, q2, 0. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p2, q2, 0.1 /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p2, q2, 0.4 /*fuzziness*/);
|
|
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p3, q3, 0. /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p3, q3, 0.33 /*fuzziness*/);
|
|
run_with_fuzziness<SearchTraits>(all_points, tree, p3, q3, 1. /*fuzziness*/);
|
|
}
|
|
|
|
int main()
|
|
{
|
|
const int N=10000;
|
|
|
|
// generator for random data points in the square ( (-1,-1), (1,1) )
|
|
Random_points_iterator rpit(1.0);
|
|
|
|
// construct list containing N random points
|
|
std::list<Point> all_points(N_Random_points_iterator(rpit,0),
|
|
N_Random_points_iterator(N));
|
|
|
|
// add some interesting points
|
|
all_points.push_back(Point(0., 0.));
|
|
all_points.push_back(Point(-CGAL_PI/10.+0.1, -CGAL_PI/10.+0.1));
|
|
all_points.push_back(Point(1., 1.));
|
|
all_points.push_back(Point(0., 1.));
|
|
all_points.push_back(Point(0.3, 0.4));
|
|
all_points.push_back(Point(0.2, 0.3));
|
|
all_points.push_back(Point(0., 0.1));
|
|
|
|
run<Traits>(all_points);
|
|
run<Traits_with_info>(all_points);
|
|
|
|
return 0;
|
|
}
|
|
|
|
|