/*
    Copyright 2009 Olaf Hartig

    This file is free software: you can redistribute it and/or modify it
    under the terms of the GNU General Public License as published by the
    Free Software Foundation, either version 3 of the License, or (at your
    option) any later version.

    This file is distributed in the hope that it will be useful, but WITHOUT
    ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
    FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
    more details. You find the GNU General Public License at
    <http://www.gnu.org/licenses/>.
*/

import java.io.File;
import java.io.FileOutputStream;
import java.io.FileNotFoundException;
import java.io.OutputStream;

import java.util.Iterator;

import com.hp.hpl.jena.rdf.model.Model;
import com.hp.hpl.jena.shared.JenaException;
import com.hp.hpl.jena.util.FileManager;

import arq.cmdline.ArgDecl;
import arq.cmdline.ArgModuleGeneral;
import arq.cmdline.CmdArgModule;
import arq.cmdline.CmdGeneral;

import com.hp.hpl.jena.query.QueryExecution;
import com.hp.hpl.jena.query.QueryExecutionFactory;
import com.hp.hpl.jena.query.ResultSet;
import com.hp.hpl.jena.sparql.util.Utils;

import de.fuberlin.wiwiss.ng4j.NamedGraph;
import de.fuberlin.wiwiss.ng4j.NamedGraphSet;
import de.fuberlin.wiwiss.ng4j.impl.NamedGraphImpl;
import de.fuberlin.wiwiss.ng4j.impl.NamedGraphSetImpl;


/**
 * Creates a sliced graphset from a dataset generated by the Berlin SPARQL
 * Benchmark (BSBM).
 * The created graphset provides a sliced view of the original dataset as if
 * the original has been retrieved using linked data principles. Thus, the
 * created graphset contains single RDF graphs for each product type, product
 * feature, producer, product, vendor, offer, person, and review in the BSBM-
 * generated dataset. To slice the dataset this program simply queries the
 * dataset for all relevant resources, issues a DESCRIBE query for each of
 * these resources, and adds the graphs resulting from the DESCRIBE queries
 * to a set of named graphs.
 *
 * To compile this program you need NG4J (including Jena and ARQ provided with
 * NG4J).
 *
 * Version: 1.0
 * @author Olaf Hartig
 */
public class convert extends CmdGeneral
{
	final protected ModConvert modConvert = new ModConvert ();
	final private NamedGraphSet ngs = new NamedGraphSetImpl ();
	private long countTriples;


	// initialization

	protected convert ( String[] args )
	{
		super( args );
		addModule( modConvert );
	}

	static public void main ( String [] args )
	{
		new convert( args ).mainRun();
	}

	// implementation of the CmdArgModule interface

	protected void processModulesAndArgs ()
	{
	}


	// implementation of the CmdMain interface

	final protected void exec ()
	{
		ngs.clear();
		countTriples = 0;

		createGraphsForProductTypes();
		createGraphsForProductFeatures();
		createGraphsForProducers();
		createGraphsForProducts();
		createGraphsForVendors();
		createGraphsForOffers();
		createGraphsForPersons();
		createGraphsForReviews();

		System.out.println( String.valueOf(countTriples) + " triples in " + ngs.countGraphs() + " graphs" );
		ngs.write( modConvert.getOutputStream(), "TRIG", null );
	}

	protected String getCommandName ()
	{
		return Utils.className( this );
	}


	// implementation of the CmdGeneral interface

	protected String getSummary ()
	{
		return getCommandName() + " --dataset=<file> [--output=<file>]";
	}


	// helpers

	protected void createGraphsForProductTypes ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?t \n" +
		               "WHERE { \n" +
		               "   ?t rdf:type bsbm:ProductType . \n" +
		               "}";
		addGraphsFromQueryResults( query, "t" );
	}

	protected void createGraphsForProductFeatures ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?f \n" +
		               "WHERE { \n" +
		               "   ?f rdf:type bsbm:ProductFeature . \n" +
		               "}";
		addGraphsFromQueryResults( query, "f" );
	}

	protected void createGraphsForProducers ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?p \n" +
		               "WHERE { \n" +
		               "   ?p rdf:type bsbm:Producer . \n" +
		               "}";
		addGraphsFromQueryResults( query, "p" );
	}

	protected void createGraphsForProducts ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?p \n" +
		               "WHERE { \n" +
		               "   ?p rdf:type bsbm:Product . \n" +
		               "}";
		addGraphsFromQueryResults( query, "p" );
	}

	protected void createGraphsForVendors ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?v \n" +
		               "WHERE { \n" +
		               "   ?v rdf:type bsbm:Vendor . \n" +
		               "}";
		addGraphsFromQueryResults( query, "v" );
	}

	protected void createGraphsForOffers ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?o \n" +
		               "WHERE { \n" +
		               "   ?o rdf:type bsbm:Offer . \n" +
		               "}";
		addGraphsFromQueryResults( query, "o" );
	}

	protected void createGraphsForPersons ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX foaf: <http://xmlns.com/foaf/0.1/> \n" +
		               "SELECT ?p \n" +
		               "WHERE { \n" +
		               "   ?p rdf:type foaf:Person . \n" +
		               "}";
		addGraphsFromQueryResults( query, "p" );
	}

	protected void createGraphsForReviews ()
	{
		String query = "PREFIX rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> \n" +
		               "PREFIX bsbm: <http://www4.wiwiss.fu-berlin.de/bizer/bsbm/v01/vocabulary/> \n" +
		               "SELECT ?r \n" +
		               "WHERE { \n" +
		               "   ?r rdf:type bsbm:Review . \n" +
		               "}";
		addGraphsFromQueryResults( query, "r" );
	}

	protected void addGraphsFromQueryResults ( String query, String varName )
	{
		Iterator<String> it = execSingleElmtQuery( query, varName );
		while ( it.hasNext() )
		{
			String resourceURI = it.next();
			String graphURI = resourceURI.replace( "/instances/", "/data/" );
			NamedGraph ng = new NamedGraphImpl( graphURI, describe(resourceURI).getGraph() );
			ngs.addGraph( ng );
			countTriples += ng.size();
		}
	}

	protected Iterator<String> execSingleElmtQuery ( String query, String varName )
	{
		QueryExecution qe = QueryExecutionFactory.create( query, modConvert.getBSBMDataset() );
		ResultSet results = qe.execSelect();
		return new SingleBindingIterator ( results, varName );
	}

	protected Model describe ( String uri )
	{
		QueryExecution qe = QueryExecutionFactory.create( "DESCRIBE <" + uri + ">", modConvert.getBSBMDataset() );
		return qe.execDescribe();
	}


	/** The parameter module for this program. */
	static class ModConvert implements ArgModuleGeneral
	{
		final protected ArgDecl datasetDecl = new ArgDecl( ArgDecl.HasValue, "dataset" );
		final protected ArgDecl outputDecl = new ArgDecl( ArgDecl.HasValue, "output" );
		private Model bsbmDataset;
		private OutputStream out;

		// implementation of the ArgModuleGeneral interface

		public void registerWith ( CmdGeneral cmdLine )
		{
			cmdLine.getUsage().startCategory( "Dataset Conversion" );
			cmdLine.add( datasetDecl,
			             "--dataset",
			             "File containing the benchmark dataset as generated by BSBM (mandatory)") ;
			cmdLine.add( outputDecl,
			             "--output",
			             "The filename for the converted dataset (optional; print to stdout if omitted)") ;
		}

		// implementation of the ArgModule interface

		public void processArgs ( CmdArgModule cmdLine )
		{
			bsbmDataset = null;

			if ( cmdLine.contains(datasetDecl) )
			{
				String fileName = cmdLine.getValue( datasetDecl );
				try {
					bsbmDataset = FileManager.get().loadModel( fileName );
				} catch ( JenaException e ) {
					cmdLine.cmdError( "Loading the given dataset file failed (" + Utils.className(e) + ": " + e.getMessage() + ")." );
				}
			}
			else {
				cmdLine.cmdError( "No dataset specified" );
			}

			if ( cmdLine.contains(outputDecl) )
			{
				File outputFile = new File( cmdLine.getValue(outputDecl) );
				try {
					out = new FileOutputStream( outputFile );
				} catch ( FileNotFoundException e ) {
					cmdLine.cmdError( "Writing to the given output file is impossible: " + e.getMessage() );
				}
			}
			else {
				out = System.out;
			}
		}

		// accessors

		public Model getBSBMDataset ()
		{
			return bsbmDataset;
		}

		public OutputStream getOutputStream ()
		{
			return out;
		}
	}


	/**
	 * An iterator that returns the URIs bound to a given variable in a set of
	 * query results.
	 */
	static class SingleBindingIterator implements Iterator<String>
	{
		final private ResultSet results;
		final private String varName;
		public SingleBindingIterator ( ResultSet results, String varName ) { this.results = results; this.varName = varName; }
		public boolean hasNext () { return results.hasNext(); }
		public String next () { return results.nextSolution().getResource(varName).getURI(); }
		public void remove () { throw new UnsupportedOperationException(); }
	}
}