/**
    * Copyright (c) 2011, The University of Southampton and the individual contributors.
    * All rights reserved.
    *
    * Redistribution and use in source and binary forms, with or without modification,
    * are permitted provided that the following conditions are met:
    *
    *   * 	Redistributions of source code must retain the above copyright notice,
    * 	this list of conditions and the following disclaimer.
   *
   *   *	Redistributions in binary form must reproduce the above copyright notice,
   * 	this list of conditions and the following disclaimer in the documentation
   * 	and/or other materials provided with the distribution.
   *
   *   *	Neither the name of the University of Southampton nor the names of its
   * 	contributors may be used to endorse or promote products derived from this
   * 	software without specific prior written permission.
   *
   * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
   * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
   * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
   * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
   * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
   * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
   * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
   * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
   * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
   * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
   */
  package org.openimaj.ml.clustering.dbscan;
  
  import gnu.trove.list.TIntList;
  import gnu.trove.list.array.TIntArrayList;
  import gnu.trove.map.hash.TIntObjectHashMap;
  import gnu.trove.procedure.TIntObjectProcedure;
  import gnu.trove.procedure.TIntProcedure;
  import gnu.trove.set.hash.TIntHashSet;
  
  import java.util.List;
  
  import org.openimaj.ml.clustering.dbscan.neighbourhood.RegionMode;
  import org.openimaj.util.pair.IntDoublePair;
  
  /**
   * Implementation of DBSCAN (http://en.wikipedia.org/wiki/DBSCAN) using
   * a
   * @author Sina Samangooei (ss@ecs.soton.ac.uk)
   *
   */
  public class DBSCAN {
  	protected boolean noiseAsClusters = false;
  	/**
  	 *
  	 * @author Sina Samangooei (ss@ecs.soton.ac.uk)
  	 */
  	public static class State{
  
  		TIntHashSet visited = new TIntHashSet();
  		TIntHashSet noise = new TIntHashSet();
  		TIntHashSet addedToCluster = new TIntHashSet();
  		TIntObjectHashMap<TIntList> clusters = new TIntObjectHashMap<TIntList>();
  		private RegionMode<IntDoublePair> regionMode;
  		private int length;
  		private boolean noiseAsClusters;
  		
  		/**
  		 * 
  		 * @param length
  		 * @param regionMode
  		 */
  		public State(int length, RegionMode<IntDoublePair> regionMode)
  		{
  			this(length, regionMode, false);
  		}
  		
  		/**
  		 * @param length
  		 * @param regionMode
  		 * @param noiseAsClusters treat noise as isolated clusters
  		 */
  		public State(int length, RegionMode<IntDoublePair> regionMode, boolean noiseAsClusters){
  			this.regionMode = regionMode;
  			this.length = length;
  			this.noiseAsClusters = noiseAsClusters;
  		}
  	}
  
  	DoubleDBSCANClusters dbscan(final State state) {
  		final int[] clusterIndex = new int[]{0};
  		for (int p = 0; p < state.length; p++) {
  			if(state.visited.contains(p))continue;
  			state.visited.add(p);
  			List<IntDoublePair> region = state.regionMode.regionQuery(p);
  			if(!state.regionMode.validRegion(region)){
  				state.noise.add(p);
  			}
  			else{
  				TIntList cluster = new TIntArrayList();
  				state.clusters.put(clusterIndex[0], cluster);
 				expandCluster(p,region,cluster,state);
 				clusterIndex[0]++;
 			}
 		}
 		
 		if(state.noiseAsClusters){
 			state.noise.forEach(new TIntProcedure() {
 				
 				@Override
 				public boolean execute(int value) {
 					TIntArrayList arr = new TIntArrayList();
 					arr.add(value);
 					state.clusters.put(clusterIndex[0]++, arr);
 					return true;
 				}
 			});
 		}
 		final int[][] clusterMembers = new int[state.clusters.size()][];
 		final int[] nEntries = new int[1];
 		state.clusters.forEachEntry(new TIntObjectProcedure<TIntList>() {
 			@Override
 			public boolean execute(int cluster, TIntList b) {
 				clusterMembers[cluster] = b.toArray();
 				nEntries[0] += clusterMembers[cluster].length; 
 				return true;
 			}
 		});
 		int[] noise = state.noise.toArray();
 		
 		final DoubleDBSCANClusters dbscanClusters = new DoubleDBSCANClusters(noise, clusterMembers);
 		return dbscanClusters;
 	}
 
 	private void expandCluster(int p, List<IntDoublePair> region, TIntList cluster, State state) {
 		addToCluster(p,cluster,state);
 		for (int regionIndex = 0; regionIndex < region.size(); regionIndex++) {
 			int pprime = region.get(regionIndex).first;
 			if (!state.visited.contains(pprime)){
 				state.visited.add(pprime);
 				List<IntDoublePair> regionPrime = state.regionMode.regionQuery(pprime);
 				if(state.regionMode.validRegion(regionPrime)) 
 					region.addAll(regionPrime);
 				else
 					state.noise.add(pprime);
 			}
 
 			addToCluster(pprime, cluster, state);
 		}
 	}
 
 	private void addToCluster(int p, TIntList cluster, State state) {
 		if(!state.addedToCluster.contains(p)){
 			// This point is now certainly at the very least DENSITY reachable and must not be noise
 			state.noise.remove(p); 
 			cluster.add(p);
 			state.addedToCluster.add(p);
 		}
 	}
 	
 	/**
 	 * Treat noise as clusters on their own
 	 * @param b
 	 */
 	public void setNoiseAsClusters(boolean b) {
 		this.noiseAsClusters  = b;
 	}
 }