def __points_within_distance(node, results):
"""Build a list of indexes which are within distance of the point and vertex."""
if not node:
return
vertex_index = node.vertex_index
this_position = positions[vertex_index]
this_uv = uvs[vertex_index]
if vmath.v3equal(this_position, position,
position_tolerance) and vmath.v2equal(
this_uv, uv, uv_tolerance):
results.append(vertex_index)
split_axis = node.split_axis
v_axis = this_position[split_axis]
p_axis = position[split_axis]
if (p_axis + position_tolerance) < v_axis:
__points_within_distance(node.left_child, results)
elif (p_axis - position_tolerance) > v_axis:
__points_within_distance(node.right_child, results)
else:
__points_within_distance(node.left_child, results)
__points_within_distance(node.right_child, results)
def __points_within_distance(node, results):
"""Build a list of indexes which are within distance of the point and vertex."""
if not node:
return
vertex_index = node.vertex_index
vertex = vertexes[vertex_index]
if vmath.v3equal(point, vertex, distance):
results.append(vertex_index)
split_axis = node.split_axis
v_axis = vertex[split_axis]
p_axis = point[split_axis]
if (p_axis + distance) < v_axis:
__points_within_distance(node.left_child, results)
elif (p_axis - distance) > v_axis:
__points_within_distance(node.right_child, results)
else:
__points_within_distance(node.left_child, results)
__points_within_distance(node.right_child, results)
def __points_within_distance(node, results):
"""Build a list of indexes which are within distance of the point and vertex."""
if not node:
return
vertex_index = node.vertex_index
vertex = vertexes[vertex_index]
if vmath.v3equal(point, vertex, distance):
results.append(vertex_index)
split_axis = node.split_axis
v_axis = vertex[split_axis]
p_axis = point[split_axis]
if (p_axis + distance) < v_axis:
__points_within_distance(node.left_child, results)
elif (p_axis - distance) > v_axis:
__points_within_distance(node.right_child, results)
else:
__points_within_distance(node.left_child, results)
__points_within_distance(node.right_child, results)
def __points_within_distance(node, results):
"""Build a list of indexes which are within distance of the point and vertex."""
if not node:
return
vertex_index = node.vertex_index
this_position = positions[vertex_index]
this_uv = uvs[vertex_index]
if vmath.v3equal(this_position, position, position_tolerance) and vmath.v2equal(this_uv, uv, uv_tolerance):
results.append(vertex_index)
split_axis = node.split_axis
v_axis = this_position[split_axis]
p_axis = position[split_axis]
if (p_axis + position_tolerance) < v_axis:
__points_within_distance(node.left_child, results)
elif (p_axis - position_tolerance) > v_axis:
__points_within_distance(node.right_child, results)
else:
__points_within_distance(node.left_child, results)
__points_within_distance(node.right_child, results)
split_axis = depth % 3
# Sort point list and choose median as pivot element
indexes.sort(key=lambda v: vertexes[v][split_axis])
median = len(indexes) / 2 # choose median
# Create node and construct subtrees
node = Node(indexes[median], split_axis)
node.left_child = build_kdtree_nodes(vertexes, indexes[0:median], depth + 1)
node.right_child = build_kdtree_nodes(vertexes, indexes[median+1:], depth + 1)
return node
#######################################################################################################################
if __name__ == "__main__":
import random
NUM = 1000
VERTEXES = [ (random.random(), random.random(), random.random()) for x in range(NUM) ]
ROOT = build_kdtree(VERTEXES)
POINT = (0.25, 0.5, 0.75)
DISTANCE = 0.1
RESULTS = ROOT.points_within_distance(VERTEXES, POINT, DISTANCE)
RESULTS.sort()
for r in RESULTS:
print "Result: %i %s is close to %s." % (r, VERTEXES[r], POINT)
print "=" * 80
for i, x in enumerate(VERTEXES):
if vmath.v3equal(x, POINT, DISTANCE):
print "Result: %i %s is close to %s." % (i, x, POINT)
median = len(indexes) / 2 # choose median
# Create node and construct subtrees
node = Node(indexes[median], split_axis)
node.left_child = build_kdtree_nodes(vertexes, indexes[0:median],
depth + 1)
node.right_child = build_kdtree_nodes(vertexes, indexes[median + 1:],
depth + 1)
return node
#######################################################################################################################
if __name__ == "__main__":
import random
NUM = 1000
VERTEXES = [(random.random(), random.random(), random.random())
for x in range(NUM)]
ROOT = build_kdtree(VERTEXES)
POINT = (0.25, 0.5, 0.75)
DISTANCE = 0.1
RESULTS = ROOT.points_within_distance(VERTEXES, POINT, DISTANCE)
RESULTS.sort()
for r in RESULTS:
print "Result: %i %s is close to %s." % (r, VERTEXES[r], POINT)
print "=" * 80
for i, x in enumerate(VERTEXES):
if vmath.v3equal(x, POINT, DISTANCE):
print "Result: %i %s is close to %s." % (i, x, POINT)