turtleparser.java

这是外国一个开源推理机

/*  Sesame - Storage and Querying architecture for RDF and RDF Schema
 *  Copyright (C) 2001-2005 Aduna
 *
 *  Contact: 
 *  	Aduna
 *  	Prinses Julianaplein 14 b
 *  	3817 CS Amersfoort
 *  	The Netherlands
 *  	tel. [phone]
 *  	fax. [phone]
 *
 *  	http://aduna.biz/
 *  	http://www.openrdf.org/
 *  
 *  This library is free software; you can redistribute it and/or
 *  modify it under the terms of the GNU Lesser General Public
 *  License as published by the Free Software Foundation; either
 *  version 2.1 of the License, or (at your option) any later version.
 *
 *  This library 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
 *  Lesser General Public License for more details.
 *
 *  You should have received a copy of the GNU Lesser General Public
 *  License along with this library; if not, write to the Free Software
 *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
 */

package org.openrdf.rio.turtle;

import java.io.IOException;
import java.io.InputStream;
import java.io.InputStreamReader;
import java.io.LineNumberReader;
import java.io.PushbackReader;
import java.io.Reader;
import java.io.UnsupportedEncodingException;
import java.util.HashMap;
import java.util.Map;

import org.openrdf.util.ASCIIUtil;
import org.openrdf.util.xml.XmlDatatypeUtil;
import org.openrdf.vocabulary.RDF;
import org.openrdf.vocabulary.XmlSchema;

import org.openrdf.model.BNode;
import org.openrdf.model.Literal;
import org.openrdf.model.Resource;
import org.openrdf.model.URI;
import org.openrdf.model.Value;
import org.openrdf.model.ValueFactory;
import org.openrdf.model.impl.ValueFactoryImpl;

import org.openrdf.rio.NamespaceListener;
import org.openrdf.rio.ParseErrorListener;
import org.openrdf.rio.ParseException;
import org.openrdf.rio.ParseLocationListener;
import org.openrdf.rio.Parser;
import org.openrdf.rio.StatementHandler;
import org.openrdf.rio.StatementHandlerException;

/**
 * Parser for Turtle files. A specification of Turtle can be found
 * <a href="http://www.ilrt.bris.ac.uk/discovery/2004/01/turtle/">in this document</a>.
 * This parser is not thread-safe, therefore its public methods are
 * synchronized.
 **/
public class TurtleParser implements Parser {

/*--------------+
| Variables     |
+--------------*/

	private StatementHandler _statementHandler;

	private NamespaceListener _nsListener;

	private ParseErrorListener _errListener;

	private ParseLocationListener _locListener;

	/** The base URI for resolving relative URIs. **/
	private org.openrdf.util.uri.URI _baseURI;

	private LineNumberReader _lineReader;
	private PushbackReader _reader;
	
	private ValueFactory _valFactory;

	/**
	 * Mapping from bNode ID's as used in the RDF document to the
	 * object created for it by the ValueFactory.
	 **/
	private Map _bNodeIdMap;

	private Map _namespaceTable;

	private Resource _subject;
	private URI _predicate;
	private Value _object;

	/** Flag indicating whether the parser should check the data it parses. **/
	boolean _verifyData = true;

	/** 
	 * Flag indicating whether the parser should preserve bnode identifiers
	 * specified in the source.
	 */
	boolean _preserveBNodeIds = false;
	
	/**
	 * Indicates how datatyped literals should be handled. Legal
	 * values are <tt>DT_IGNORE</tt>, <tt>DT_VERIFY</tt> and
	 * <tt>DT_NORMALIZE</tt>.
	 **/
	private int _datatypeHandling;

/*--------------+
| Constructors  |
+--------------*/

	/**
	 * Creates a new TurtleParser that will use a <tt>ValueFactoryImpl</tt> to
	 * create object for resources, bNodes and literals.
	 * @see org.openrdf.model.impl.ValueFactoryImpl
	 **/
	public TurtleParser() {
		this(new ValueFactoryImpl());
	}

	/**
	 * Creates a new TurtleParser that will use the supplied ValueFactory to
	 * create objects for resources, bNodes and literals.
	 *
	 * @param valueFactory A ValueFactory.
	 **/
	public TurtleParser(ValueFactory valueFactory) {
		_valFactory = valueFactory;
		_bNodeIdMap = new HashMap();
		_datatypeHandling = DT_VERIFY;
		_namespaceTable = new HashMap(16);
	}

/*--------------+
| Methods       |
+--------------*/

	// Implements Parser.setStatementHandler(StatementHandler)
	public synchronized void setStatementHandler(StatementHandler sh) {
		_statementHandler = sh;
	}

	// Implements Parser.setNamespaceListener(NamespaceListener)
	public void setNamespaceListener(NamespaceListener nl) {
		_nsListener = nl;
	}

	// Implements Parser.setParseErrorListener(ParseErrorListener)
	public synchronized void setParseErrorListener(ParseErrorListener el) {
		_errListener = el;
	}

	// Implements Parser.setParseLocationListener(ParseLocationListener)
	public synchronized void setParseLocationListener(ParseLocationListener el) {
		_locListener = el;
	}

	// Implements Parser.setVerifyData(boolean)
	public synchronized void setVerifyData(boolean verifyData) {
		_verifyData = verifyData;
	}

	// Implements Parser.setPreserveBNodeIds(boolean)
	public void setPreserveBNodeIds(boolean preserveBNodeIds) {
		_preserveBNodeIds = preserveBNodeIds;
	}

	// Implements Parser.setStopAtFirstError(boolean)
	public synchronized void setStopAtFirstError(boolean stopAtFirstError) {
		// ignore
	}

	// Implements Parser.setDatatypeHandling(int)
	public void setDatatypeHandling(int datatypeHandling) {
		_datatypeHandling = datatypeHandling;
	}

	/**
	 * Implementation of the <tt>parse(InputStream, String)</tt> method defined
	 * in the Parser interface. 	
	 * 
	 * @param in The InputStream from which to read the data. The InputStream is
	 * supposed to contain UTF-8 encoded Unicode characters, as per the Turtle
	 * specification.
	 * @param baseURI The URI associated with the data in the InputStream.
	 * @exception IOException If an I/O error occurred while data was read
	 * from the InputStream.
	 * @exception ParseException If the parser has found an unrecoverable
	 * parse error.
	 * @exception StatementHandler If the configured statement handler
	 * encountered an unrecoverable error.
	 * @exception IllegalArgumentException If the supplied input stream or
	 * base URI is <tt>null</tt>.
	 **/
	public synchronized void parse(InputStream in, String baseURI)
		throws IOException, ParseException, StatementHandlerException
	{
		if (in == null) {
			throw new IllegalArgumentException("Input stream can not be 'null'");
		}
		// Note: baseURI will be checked in parse(Reader, String)

		try {
			parse(new InputStreamReader(in, "UTF-8"), baseURI);
		}
		catch (UnsupportedEncodingException e) {
			// Every platform should support the UTF-8 encoding...
			throw new RuntimeException(e);
		}
	}

	/**
	 * Implementation of the <tt>parse(Reader, String)</tt> method defined in
	 * the Parser interface. 	
	 * 
	 * @param reader The Reader from which to read the data.
	 * @param baseURI The URI associated with the data in the Reader.
	 * @exception IOException If an I/O error occurred while data was read
	 * from the InputStream.
	 * @exception ParseException If the parser has found an unrecoverable
	 * parse error.
	 * @exception StatementHandler If the configured statement handler
	 * encountered an unrecoverable error.
	 * @exception IllegalArgumentException If the supplied reader or base URI
	 * is <tt>null</tt>.
	 **/
	public synchronized void parse(Reader reader, String baseURI)
		throws IOException, ParseException, StatementHandlerException
	{
		if (reader == null) {
			throw new IllegalArgumentException("Reader can not be 'null'");
		}
		if (baseURI == null) {
			throw new IllegalArgumentException("base URI can not be 'null'");
		}

		_lineReader = new LineNumberReader(reader);
		// Start counting lines at 1:
		_lineReader.setLineNumber(1);

		// Allow at most 2 characters to be pushed back:
		_reader = new PushbackReader(_lineReader, 2);

		// Store normalized base URI
		_baseURI = new org.openrdf.util.uri.URI(baseURI);
		_baseURI.normalize();

		_reportLocation();

		try {
			_skipWhitespace();

			int c = _peek();
			while (c != -1) {
				if (c == '#') {
					// Comment, ignore line
				 	_skipLine();
				}
				else if (c == '@') {
					_parsePrefix();
				}
				else {
					_parseTriple();
				}

				_skipWhitespace();
				c = _peek();
			}
		}
		finally {
			_bNodeIdMap.clear();
			_namespaceTable.clear();
		}
	}

	/**
	 * Reads characters from _reader until it finds a character that is not
	 * a space, tab, line feed or newline.
	 **/
	private void _skipWhitespace()
		throws IOException
	{
		int c = _reader.read();
		while (TurtleUtil.isWhitespace(c)) {
			c = _reader.read();
		}

		_unread(c);
	}

	/**
	 * Reads characters from _reader until the first EOL has been read. The
	 * first character after the EOL is returned. In case the end of the
	 * character stream has been reached, -1 is returned.
	 **/
	private void _skipLine()
		throws IOException
	{
		int c = _reader.read();
		while (c != -1 && c != 0xD && c != 0xA) {
			c = _reader.read();
		}

		// c is equal to -1, \r or \n.
		// In case c is equal to \r, we should also read a following \n.
		if (c == 0xD) {
			c = _reader.read();

			if (c != 0xA) {
				_unread(c);
			}
		}

		_reportLocation();
	}

	private void _parsePrefix()
		throws IOException, ParseException
	{
		// Verify that the first characters form the string "prefix"
		_verifyCharacter(_reader.read(), "@");
		_verifyCharacter(_reader.read(), "p");
		_verifyCharacter(_reader.read(), "r");
		_verifyCharacter(_reader.read(), "e");
		_verifyCharacter(_reader.read(), "f");
		_verifyCharacter(_reader.read(), "i");
		_verifyCharacter(_reader.read(), "x");

		_skipWhitespace();

		// Read prefix ID (e.g. "rdf:" or ":")
		StringBuffer prefixID = new StringBuffer(8);

		int c = _reader.read();
		while (c != ':') {
			if (c == -1) {
				_throwEOFException();
			}
			prefixID.append( (char)c );
			c = _reader.read();
		}

		// c now contains the ':', ignore it
		_reader.read();

		_skipWhitespace();

		// Read the namespace URI
		URI namespace = _parseUriRef();

		_skipWhitespace();

		// Read closing dot
		_verifyCharacter(_reader.read(), ".");

		// Store namespace mapping and report it to the namespace listener (if any).
		String prefixStr = prefixID.toString();
		String namespaceStr = namespace.getURI();

		_namespaceTable.put(prefixStr, namespaceStr);

		if (_nsListener != null) {
			_nsListener.handleNamespace(prefixStr, namespaceStr);
		}
	}

	private void _parseTriple()
		throws IOException, ParseException, StatementHandlerException
	{
		_parseSubject();

		_skipWhitespace();

		_parsePredicateObjectList();

		_skipWhitespace();

		_verifyCharacter(_reader.read(), ".");

		_subject = null;
		_predicate = null;
		_object = null;
	}

	private void _parsePredicateObjectList()
		throws IOException, ParseException, StatementHandlerException
	{
		_predicate = _parsePredicate();
		_skipWhitespace();
		_parseObjectList();
		_skipWhitespace();

		int c = _peek();

		while (c == ';') {
			_reader.read();
			_skipWhitespace();

			c = _peek();
			if (c == '.' || // end of triple
				c == ']') // end of predicateObjectList inside blank node
			{
				break;
			}

			_predicate = _parsePredicate();
			_skipWhitespace();
			_parseObjectList();
			_skipWhitespace();

			c = _peek();
		}
	}

	private void _parseObjectList()
		throws IOException, ParseException, StatementHandlerException
	{
		_parseObject();
		
		_skipWhitespace();
		int c = _peek();

		while (c == ',') {
			_reader.read();
			_skipWhitespace();

			_parseObject();

			_skipWhitespace();
			c = _peek();
		}
	}

	private void _parseSubject()
		throws IOException, ParseException, StatementHandlerException
	{
		int c = _peek();
		if (c == '<' || c == ':' || TurtleUtil.isPrefixStartChar(c)) {
			// subject is an uriref or a qname
			_subject = _parseResource();
		}
		else if (c == '_' || c == '[' || c == '(') {
			// subject is a blank node (or an empty list)
			_subject = _parseBlank();
		}
		else if (c == -1) {
			_throwEOFException();
		}
		else {
			_throwParseException("Expected an uriref, qname or node ID, found '" + (char)c + "'");
		}
	}

	private URI _parsePredicate()
		throws IOException, ParseException
	{
		// Check the short-cut 'a' is used
		int c1 = _reader.read();

		if (c1 == 'a') {
			int c2 = _reader.read();

			if (TurtleUtil.isWhitespace(c2)) {
				// Short-cut is used, return the rdf:type URI
				return _createURI(RDF.TYPE);
			}

			// Short-cut is not used, unread all characters
			_unread(c2);
		}
		_unread(c1);

		// Predicate is a normal resource
		return _parseResource();
	}

	private void _parseObject()
		throws IOException, ParseException, StatementHandlerException
	{
		int c = _peek();
		if (c == '<' || c == ':' || TurtleUtil.isPrefixStartChar(c)) {
			// object is an uriref or a qname
			_object = _parseResource();
		}
		else if (c == '_' || c == '[' || c == '(') {
			// object is a blank node (or an empty list)
			_object = _parseBlank();
		}
		else if (c == '"') {
			// object is a literal
			_object = _parseLiteral();
		}
		else if (ASCIIUtil.isNumber(c)) {
			// object is an integer
			_object = _parseInteger();
		}
		else if (c == -1) {
			_throwEOFException();
		}
		else {
			_throwParseException("Expected resource, node ID, literal or integer, found '" + (char)c + "'");
		}

		_statementHandler.handleStatement(_subject, _predicate, _object);
	}

	private Literal _parseLiteral()
		throws IOException, ParseException
	{
		// First character should be '"'
		_verifyCharacter(_reader.read(), "\"");

		// Read up to the next '"' character
		StringBuffer label = new StringBuffer(32);
		int c = _reader.read();
		while (c != '"') {
			if (c == -1) {
				_throwEOFException();
			}
			label.append( (char)c );