All Downloads are FREE. Search and download functionalities are using the official Maven repository.

org.apache.commons.math3.geometry.partitioning.Region Maven / Gradle / Ivy

Go to download

The Apache Commons Math project is a library of lightweight, self-contained mathematics and statistics components addressing the most common practical problems not immediately available in the Java programming language or commons-lang.

There is a newer version: 3.6.1
Show newest version
/*
 * Licensed to the Apache Software Foundation (ASF) under one or more
 * contributor license agreements.  See the NOTICE file distributed with
 * this work for additional information regarding copyright ownership.
 * The ASF licenses this file to You under the Apache License, Version 2.0
 * (the "License"); you may not use this file except in compliance with
 * the License.  You may obtain a copy of the License at
 *
 *      http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
package org.apache.commons.math3.geometry.partitioning;

import org.apache.commons.math3.geometry.Space;
import org.apache.commons.math3.geometry.Vector;

/** This interface represents a region of a space as a partition.

 * 

Region are subsets of a space, they can be infinite (whole * space, half space, infinite stripe ...) or finite (polygons in 2D, * polyhedrons in 3D ...). Their main characteristic is to separate * points that are considered to be inside the region from * points considered to be outside of it. In between, there * may be points on the boundary of the region.

*

This implementation is limited to regions for which the boundary * is composed of several {@link SubHyperplane sub-hyperplanes}, * including regions with no boundary at all: the whole space and the * empty region. They are not necessarily finite and not necessarily * path-connected. They can contain holes.

*

Regions can be combined using the traditional sets operations : * union, intersection, difference and symetric difference (exclusive * or) for the binary operations, complement for the unary * operation.

* @param Type of the space. * @version $Id: Region.java 1244107 2012-02-14 16:17:55Z erans $ * @since 3.0 */ public interface Region { /** Enumerate for the location of a point with respect to the region. */ public static enum Location { /** Code for points inside the partition. */ INSIDE, /** Code for points outside of the partition. */ OUTSIDE, /** Code for points on the partition boundary. */ BOUNDARY; } /** Build a region using the instance as a prototype. *

This method allow to create new instances without knowing * exactly the type of the region. It is an application of the * prototype design pattern.

*

The leaf nodes of the BSP tree must have a * {@code Boolean} attribute representing the inside status of * the corresponding cell (true for inside cells, false for outside * cells). In order to avoid building too many small objects, it is * recommended to use the predefined constants * {@code Boolean.TRUE} and {@code Boolean.FALSE}. The * tree also must have either null internal nodes or * internal nodes representing the boundary as specified in the * {@link #getTree getTree} method).

* @param newTree inside/outside BSP tree representing the new region * @return the built region */ Region buildNew(BSPTree newTree); /** Copy the instance. *

The instance created is completely independant of the original * one. A deep copy is used, none of the underlying objects are * shared (except for the underlying tree {@code Boolean} * attributes and immutable objects).

* @return a new region, copy of the instance */ Region copySelf(); /** Check if the instance is empty. * @return true if the instance is empty */ boolean isEmpty(); /** Check if the sub-tree starting at a given node is empty. * @param node root node of the sub-tree (must have {@link * Region Region} tree semantics, i.e. the leaf nodes must have * {@code Boolean} attributes representing an inside/outside * property) * @return true if the sub-tree starting at the given node is empty */ boolean isEmpty(final BSPTree node); /** Check if the instance entirely contains another region. * @param region region to check against the instance * @return true if the instance contains the specified tree */ boolean contains(final Region region); /** Check a point with respect to the region. * @param point point to check * @return a code representing the point status: either {@link * Location#INSIDE}, {@link Location#OUTSIDE} or {@link Location#BOUNDARY} */ Location checkPoint(final Vector point); /** Get the underlying BSP tree. *

Regions are represented by an underlying inside/outside BSP * tree whose leaf attributes are {@code Boolean} instances * representing inside leaf cells if the attribute value is * {@code true} and outside leaf cells if the attribute is * {@code false}. These leaf attributes are always present and * guaranteed to be non null.

*

In addition to the leaf attributes, the internal nodes which * correspond to cells split by cut sub-hyperplanes may contain * {@link BoundaryAttribute BoundaryAttribute} objects representing * the parts of the corresponding cut sub-hyperplane that belong to * the boundary. When the boundary attributes have been computed, * all internal nodes are guaranteed to have non-null * attributes, however some {@link BoundaryAttribute * BoundaryAttribute} instances may have their {@link * BoundaryAttribute#plusInside plusInside} and {@link * BoundaryAttribute#plusOutside plusOutside} fields both null if * the corresponding cut sub-hyperplane does not have any parts * belonging to the boundary.

*

Since computing the boundary is not always required and can be * time-consuming for large trees, these internal nodes attributes * are computed using lazy evaluation only when required by setting * the {@code includeBoundaryAttributes} argument to * {@code true}. Once computed, these attributes remain in the * tree, which implies that in this case, further calls to the * method for the same region will always include these attributes * regardless of the value of the * {@code includeBoundaryAttributes} argument.

* @param includeBoundaryAttributes if true, the boundary attributes * at internal nodes are guaranteed to be included (they may be * included even if the argument is false, if they have already been * computed due to a previous call) * @return underlying BSP tree * @see BoundaryAttribute */ BSPTree getTree(final boolean includeBoundaryAttributes); /** Get the size of the boundary. * @return the size of the boundary (this is 0 in 1D, a length in * 2D, an area in 3D ...) */ double getBoundarySize(); /** Get the size of the instance. * @return the size of the instance (this is a length in 1D, an area * in 2D, a volume in 3D ...) */ double getSize(); /** Get the barycenter of the instance. * @return an object representing the barycenter */ Vector getBarycenter(); /** Compute the relative position of the instance with respect to an * hyperplane. * @param hyperplane reference hyperplane * @return one of {@link Side#PLUS Side.PLUS}, {@link Side#MINUS * Side.MINUS}, {@link Side#BOTH Side.BOTH} or {@link Side#HYPER * Side.HYPER} (the latter result can occur only if the tree * contains only one cut hyperplane) */ Side side(final Hyperplane hyperplane); /** Get the parts of a sub-hyperplane that are contained in the region. *

The parts of the sub-hyperplane that belong to the boundary are * not included in the resulting parts.

* @param sub sub-hyperplane traversing the region * @return filtered sub-hyperplane */ SubHyperplane intersection(final SubHyperplane sub); }