/*
 * (c) Copyright 2004, 2005, 2006, 2007 Hewlett-Packard Development Company, LP
 * [See end of file]
 */

package arqo;

import java.util.* ;

import com.hp.hpl.jena.graph.Node;
import com.hp.hpl.jena.graph.Triple;
import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.rdf.model.StmtIterator;
import com.hp.hpl.jena.rdf.model.Statement;
import com.hp.hpl.jena.util.FileManager;
import com.hp.hpl.jena.vocabulary.OWL;
import com.hp.hpl.jena.vocabulary.RDF;
import com.hp.hpl.jena.vocabulary.RDFS;
import com.hp.hpl.jena.sparql.core.BasicPattern;
import com.hp.hpl.jena.sparql.engine.optimizer.core.BasicPatternJoin;
import com.hp.hpl.jena.sparql.engine.optimizer.probability.ProbabilityFactory;
import com.hp.hpl.jena.sparql.engine.optimizer.probability.Probability;

/**
 *
 * 
 * @author Markus Stocker
 * @version $Id$
 */

public class accuracy 
{
	private static String inGraphFileName = null ;
	private static String inIndexFileName = null ;
	private static boolean distinctPredicates = false ;
	private static boolean ignoreEmptyResultSet = false ;
	private static int runs = 20 ;
	private static int run = 0 ;
	private static int triples = 1 ;
	private static List subjects = new ArrayList() ; // List<Node>
	private static List predicates = new ArrayList() ; // List<Node>
	private static List objects = new ArrayList() ; // List<Node>
	
	/**
	 * Main program
	 * 
	 * @param args
	 */
	public static void main(String[] args) 
	{		
		List patterns = new ArrayList() ; // List<BasicPattern>
		List probabilities = new ArrayList() ; // List<Double>
		List selectivities = new ArrayList() ; // List<Double>
		
		try
		{
			readCmdParams(args) ;
			System.out.println("Data: " + inGraphFileName) ;
			System.out.println("Index: " + inIndexFileName) ;
			Model dataModel = FileManager.get().loadModel(inGraphFileName) ;
			Model indexModel = FileManager.get().loadModel(inIndexFileName) ;
			Probability probability = ProbabilityFactory.loadDefaultModel(dataModel, indexModel) ;
			
			// The the sets for SPO
			splitStatements(dataModel) ;
			
			int numOfSubjects = subjects.size() ;
			int numOfPredicates = predicates.size() ;
			int numOfObjects = objects.size() ;
			
			Random rand = new Random() ;
			
			while (run < runs)
			{
				double p, s ;
				Triple triple1 = null , triple2 = null ;
				Node subject1 = null, predicate1 = null, object1 = null, subject2 = null, predicate2 = null, object2 = null ;
				
				// Decide whether the nodes should be bound or unbound
				subject1 = boundOrUnbound((Node)subjects.get(rand.nextInt(numOfSubjects))) ;
				predicate1 = boundOrUnbound((Node)predicates.get(rand.nextInt(numOfPredicates))) ;
				object1 = boundOrUnbound((Node)objects.get(rand.nextInt(numOfObjects))) ;
				
				if (triples == 2)
				{
					// Only in the case of joined triple patterns
					subject2 = boundOrUnbound((Node)subjects.get(rand.nextInt(numOfSubjects))) ;
					predicate2 = boundOrUnbound((Node)predicates.get(rand.nextInt(numOfPredicates))) ;
					object2 = boundOrUnbound((Node)objects.get(rand.nextInt(numOfObjects))) ;
				
					if (distinctPredicates)
					{
						predicate1 = (Node)predicates.get(rand.nextInt(numOfPredicates)) ;
						predicate2 = (Node)predicates.get(rand.nextInt(numOfPredicates)) ;
						
						while (predicate1.equals(predicate2))
						{
							predicate1 = (Node)predicates.get(rand.nextInt(numOfPredicates)) ;
							predicate2 = (Node)predicates.get(rand.nextInt(numOfPredicates)) ;
						}
					}
				}
			
				BasicPattern pattern = new BasicPattern() ;
				
				if (triples == 1)
				{
					triple1 = new Triple(subject1, predicate1, object1) ;
					pattern.add(triple1) ;
					p = probability.getProbability(triple1) ;
					s = probability.getSelectivity(triple1) ;
				}
				else 
				{
					pattern.add(new Triple(subject1, predicate1, object1)) ;
					pattern.add(new Triple(subject2, predicate2, object2)) ;
					pattern = randJoinType(pattern) ;
					triple1 = pattern.get(0) ;
					triple2 = pattern.get(1) ;
					p = probability.getProbability(triple1, triple2) ;
					s = probability.getSelectivity(triple1, triple2) ;
				}
				
				if (!(ignoreEmptyResultSet && s == 0))
				{
					probabilities.add(new Double(p)) ;
					selectivities.add(new Double(s)) ;
					patterns.add(pattern) ;
					run++ ;
				}
			}
			
			for (Iterator iter = probabilities.iterator(); iter.hasNext(); )
			{
				System.out.println((Double)iter.next()) ;
			}
			
			System.out.println() ;
			
			for (Iterator iter = selectivities.iterator(); iter.hasNext(); )
			{
				System.out.println((Double)iter.next()) ;
			}
			
			System.out.println() ;
			
			for (Iterator iter = patterns.iterator(); iter.hasNext(); )
			{
				System.out.println((BasicPattern)iter.next()) ;
			}

		} catch (Exception e) { e.printStackTrace() ; }
	}
	
	private static BasicPattern randJoinType(BasicPattern pattern)
	{ 
		Random rand = new Random() ;
		int nextRand = rand.nextInt(4) ;
		BasicPattern p = new BasicPattern() ;
		Triple triple1 = pattern.get(0) ;
		Triple triple2 = pattern.get(1) ;
		Triple t1 = null, t2 = null ;
		
		if (! BasicPatternJoin.isJoined(triple1, triple2)) 
		{
			while (nextRand == 2 && !triple1.getObject().isURI())
			{
				// For the join OS, the object of the first triple has to be a URI
				nextRand = rand.nextInt(4) ;
			}
			
			t1 = triple1 ;
			
			if (nextRand == 0) // SS join
				t2 = new Triple(triple1.getSubject(), triple2.getPredicate(), triple2.getObject()) ;
			else if (nextRand == 1) // SO join
				t2 = new Triple(triple2.getSubject(), triple2.getPredicate(), triple1.getSubject()) ;
			else if (nextRand == 2) // OS join
				t2 = new Triple(triple1.getObject(), triple2.getPredicate(), triple2.getObject()) ;
			else // OO join
				t2 = new Triple(triple2.getSubject(), triple2.getPredicate(), triple1.getObject()) ;
		}
		else
		{
			t1 = triple1 ;
			t2 = triple2 ;
		}
		
		p.add(t1) ;
		p.add(t2) ;
		
		return p ;
	}
	
	private static Node boundOrUnbound(Node node)
	{
		Random rand = new Random() ;
		
		if (rand.nextInt(2) == 0)
			return node ;
		
		return Node.createVariable("x" + Math.abs(rand.nextInt())) ;
	}
	
	private static void splitStatements(Model dataModel)
	{		
		Set s = new HashSet() ; // Set<Node>
		Set p = new HashSet() ; // Set<Node>
		Set o = new HashSet() ; // Set<Node>
		
		StmtIterator stmtIter = dataModel.listStatements() ;

		while (stmtIter.hasNext())
		{
			Statement stmt = (Statement)stmtIter.nextStatement() ;
			Node subject = stmt.getSubject().asNode() ;
			Node predicate = stmt.getPredicate().asNode() ;
			Node object = stmt.getObject().asNode() ;
			
			s.add(subject) ;
			
			if (! exclude(predicate))
				p.add(predicate) ;
			
			o.add(object) ;
		}
			
		subjects.addAll(s) ;
		predicates.addAll(p) ;
		objects.addAll(o) ;
	}
	
	// Read the command line params
	private static void readCmdParams(String[] args) throws Exception
	{
		for (int i = 0; i < args.length; i++)
		{
			if (args[i].equals("--graph"))
				inGraphFileName = args[i+1] ;
			else if (args[i].equals("--index"))
				inIndexFileName = args[i+1] ;
			else if (args[i].equals("--distinct-predicates"))
				distinctPredicates = true ;
			else if (args[i].equals("--runs"))
				runs = new Integer(args[i+1]).intValue() ;
			else if (args[i].equals("--triples"))
				triples = new Integer(args[i+1]).intValue() ;
			else if (args[i].equals("--ignore-empty-rs"))
				ignoreEmptyResultSet = true ;
			else if (args[i].equals("--help"))
				usage() ;
		}

		if (inGraphFileName == null)
			usage() ;
	}
	
	// Print the usage of the main program
	private static void usage()
	{
		String usage = "arqo.accuracy [options]\n" ;
		usage += "--graph [file name]\n" ;
		usage += "--index [file name]\n" ;
		usage += "--distinct-predicates\n" ;
		usage += "--runs [default 100]\n" ;
		usage += "--triples [1 or 2]\n" ;
		usage += "--ignore-empty-rs\n" ;
		
		System.out.println(usage) ;
		
		System.exit(0) ;
	}
	
	private static boolean exclude(Node predicate)
	{
		String uri = predicate.getURI() ;
		
		if (uri.contains(RDF.getURI()) 
				|| uri.contains(RDFS.getURI()) 
				|| uri.contains(OWL.getURI()))
			return true ;
			
		return false ;
	}

}


/*
 *  (c) Copyright 2004, 2005, 2006, 2007 Hewlett-Packard Development Company, LP
 *  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. 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.
 * 3. The name of the author may not be used to endorse or promote products
 *    derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 AUTHOR 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.
 */