package algorithms.geneticAlgorithm.holegGA;

import java.awt.Color;
import java.util.ArrayList;

import algorithms.geneticAlgorithm.holegGA.Components.HolegGeneration;
import classes.AbstractCpsObject;
import classes.Category;
import classes.CpsEdge;
import classes.HolonBattery;
import classes.HolonObject;
import classes.HolonSwitch;
import classes.Position;
import ui.controller.Control;
import ui.model.Model;

public class GenAlgoInit {
	ParameterArray params = new ParameterArray();
	Control controller;
	Model model;
	GenAlgoWindow window;

	public GenAlgoInit(Control controller, Model model){
		this.controller = controller;
		this.model = model;
	}
	
	public void start(){
		
		window = new GenAlgoWindow(controller, model);
		window.frame.setVisible(true);
		
	}
	
	public void evaluate(){
		Thread t = new Thread(){
			ArrayList<ArrayList<HolegGeneration>> runHistory = new ArrayList<ArrayList<HolegGeneration>>();
			ArrayList<Double> avgFitness = new ArrayList<Double>();
			ArrayList<Double> bestFitness = new ArrayList<Double>();
			ArrayList<Double> maxFitness =  new ArrayList<Double>();
			ArrayList<Double> maxFound = new ArrayList<Double>();
			int runs = 30;
			Double maxValue = 100.0;
			public void run(){
				Evaluator plotAvg = new Evaluator("average from all parameters");
				Evaluator plotBest = new Evaluator("best from all parameters");
				plotAvg.initPlottedRendering("Average of all parameters");
				plotBest.initPlottedRendering("Best of all parameters");
				
				for(int b = 1; b <= 3; b++){
					avgFitness = new ArrayList<Double>();
					bestFitness = new ArrayList<Double>();
					maxFitness = new ArrayList<Double>();
					maxFound = new ArrayList<Double>();
					ParameterArray params = getParameter(b);
					for(int j = 0; j < 100; j++){
						avgFitness.add((double) 0);
						bestFitness.add(0.0);
						maxFitness.add(maxValue);
						maxFound.add(0.0);
					}
					Evaluator evalAvg = new Evaluator("average " + b);
					Evaluator evalBest = new Evaluator("best set " + b);
					Evaluator evalAvgAvg = new Evaluator("avgavg " + b);
					Evaluator evalAvgBest = new Evaluator("avgBest " + b);
					evalAvg.initRendering("Average of each run for set " + b);
					evalBest.initRendering("Best of each run for set " + b);
					evalAvgAvg.initRendering("Average of all runs for set " + b);
					evalAvgBest.initRendering("Best of all runs for set " + b);
					for(int i = 0; i  < runs; i++){
						GenAlgoWindow tmp = new GenAlgoWindow(controller,model);
						ArrayList<HolegGeneration> oneRun = tmp.evaluationRun(params);
						ArrayList<Double> bestOneRun = new ArrayList<Double>();
						ArrayList<Double> averageOneRun = new ArrayList<Double>();
						ArrayList<Double> bestFoundOneRun = new ArrayList<Double>();
						for(int k = 0; k < oneRun.size(); k++){
							avgFitness.set(k, avgFitness.get(k)+oneRun.get(k).getAvgFittness());
							bestFitness.set(k, bestFitness.get(k)+oneRun.get(k).getMaxFittness());
							averageOneRun.add(oneRun.get(k).getAvgFittness());
							bestOneRun.add(oneRun.get(k).getMaxFittness());
							if(k != 0){
								if(oneRun.get(k).getMaxFittness() > bestFoundOneRun.get(k-1)){
									bestFoundOneRun.add(oneRun.get(k).getMaxFittness());
								}else{
									bestFoundOneRun.add(bestFoundOneRun.get(k-1));
								}
							}else{
								bestFoundOneRun.add(oneRun.get(k).getMaxFittness());
							}
							maxFound.set(k, maxFound.get(k)+bestFoundOneRun.get(k));
						}
						evalAvg.addSeries(averageOneRun, null, "");
						evalBest.addSeries(bestFoundOneRun, null, "");
						System.out.println("At run " + i);
					}
					for(int z = 0; z < 100; z++){
						avgFitness.set(z, avgFitness.get(z)/runs);
						bestFitness.set(z, bestFitness.get(z)/runs);
						maxFound.set(z, maxFound.get(z)/runs);
					}
					evalAvgAvg.addSeries(avgFitness, Color.GRAY, "");
					evalAvgBest.addSeries(maxFound, Color.BLUE, "");
					plotAvg.addSeries(avgFitness, getColor(b), "Parameter Set "+ b);
					plotBest.addSeries(maxFound, getColor(b), "Parameter Set " + b);
					evalAvg.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
					evalAvg.drawChart();
					evalBest.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
					evalBest.drawChart();
					evalAvgAvg.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
					evalAvgAvg.drawChart();
					evalAvgBest.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
					evalAvgBest.drawChart();
				}
				plotAvg.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
				plotAvg.drawChart();
				plotBest.addSeries(maxFitness, Color.BLACK, "Maximum Fitness");
				plotBest.drawChart();
			}
		};
		t.start();
	}
	
	public Color getColor(int i){
		if(i == 1){
			return Color.BLUE;
		}
		if(i == 2){
			return Color.GREEN;
		}else{
			return Color.RED;
		}
	}
	
	public ParameterArray getParameter(int i){
		if(i == 1){
			return getParameter1();
		}
		if(i == 2){
			return getParameter2();
		}else{
			return getParameter3();
		}
	}
	
	public ParameterArray getParameter1(){
		ParameterArray params = new ParameterArray();
		params = getSameParameters();
		params.set(params.MAX_EDGES, 39);
		params.set(params.POPULATION_SIZE, 100);
		params.set(params.TOURNAMENT_SIZE, 50);
		params.set(params.TOURNAMENT_PROB, 0.15);
		params.set(params.EDGE_MUTATION_PROB, 0.01);
		params.set(params.WILDCARD_MUTATION_PROB, 0.01);
		return params;
	}
	
	public ParameterArray getParameter2(){
		ParameterArray params = new ParameterArray();
		params = getSameParameters();
		params.set(params.MAX_EDGES, 39);
		params.set(params.POPULATION_SIZE, 100);
		params.set(params.TOURNAMENT_SIZE, 50);
		params.set(params.TOURNAMENT_PROB, 0.01);
		params.set(params.EDGE_MUTATION_PROB, 0.01);
		params.set(params.WILDCARD_MUTATION_PROB, 0.01);

		return params;
	}
	
	public ParameterArray getParameter3(){
		ParameterArray params = new ParameterArray();
		params = getSameParameters();
		params.set(params.MAX_EDGES, 39);
		params.set(params.POPULATION_SIZE, 100);
		params.set(params.TOURNAMENT_SIZE, 1);
		params.set(params.TOURNAMENT_PROB, 1.0);
		params.set(params.EDGE_MUTATION_PROB, 0.01);
		params.set(params.WILDCARD_MUTATION_PROB, 0.01);

		return params;
	}
	
	public ParameterArray getSameParameters(){
		ParameterArray params = new ParameterArray();
		params.set(params.MAX_EDGES, 50);
		params.set(params.EDIT_EDGES, true);
		params.set(params.POPULATION_SIZE, 100);
		params.set(params.TOURNAMENT_SIZE, 30);
		params.set(params.EDGE_BREAK_AMOUNT, 4);
		params.set(params.SIM_ITERATIONS, 1);
		params.set(params.EDGE_MUTATION_PROB, 0.0);
		params.set(params.SUPPLIED_POINTS, 10.0);
		params.set(params.OVERSUPPLIED_POINTS, 10.0);
		params.set(params.UNDERSUPPLIED_POINTS, -10.0);
		params.set(params.LENGTH_POINTS, -0.5);
		params.set(params.WILDCARD_USAGE, -5.0);
		params.set(params.WILDCARD_MUTATION_PROB, 0.0);
		params.set(params.PARTIALSUPPLIED_POINTS, 3.0);
		params.set(params.OVERPRODUCTION, -10.0);	
		params.set(params.NODES_IN_WILDCARDS, true);
		return params;
		
	}
	
}