001/**
002 * Copyright (c) 2011, The University of Southampton and the individual contributors.
003 * All rights reserved.
004 *
005 * Redistribution and use in source and binary forms, with or without modification,
006 * are permitted provided that the following conditions are met:
007 *
008 *   *  Redistributions of source code must retain the above copyright notice,
009 *      this list of conditions and the following disclaimer.
010 *
011 *   *  Redistributions in binary form must reproduce the above copyright notice,
012 *      this list of conditions and the following disclaimer in the documentation
013 *      and/or other materials provided with the distribution.
014 *
015 *   *  Neither the name of the University of Southampton nor the names of its
016 *      contributors may be used to endorse or promote products derived from this
017 *      software without specific prior written permission.
018 *
019 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
020 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
021 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
022 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
023 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
024 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
025 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
026 * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
027 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
028 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
029 */
030package org.openimaj.text.nlp.textpipe.annotators;
031
032import java.io.IOException;
033import java.io.InputStream;
034import java.util.ArrayList;
035import java.util.List;
036
037import opennlp.tools.tokenize.TokenizerME;
038import opennlp.tools.tokenize.TokenizerModel;
039
040import org.openimaj.text.nlp.textpipe.annotations.RawTextAnnotation;
041import org.openimaj.text.nlp.textpipe.annotations.TokenAnnotation;
042
043/**
044 * @author Sina Samangooei (ss@ecs.soton.ac.uk)
045 * 
046 */
047public class OpenNLPTokenAnnotator extends AbstractTokenAnnotator {
048
049        /**
050         *
051         */
052        public static final String TOKEN_MODEL_PROP = "org.openimaj.text.opennlp.models.token";
053        private static final String TOKEN_MODEL_DEFAULT = "/org/openimaj/text/opennlp/models/en-token.bin";
054        TokenizerME tokenizer;
055
056        /**
057         *
058         */
059        public OpenNLPTokenAnnotator() {
060                super();
061                InputStream modelIn = null;
062                modelIn = OpenNLPTokenAnnotator.class.getResourceAsStream(System.getProperty(TOKEN_MODEL_PROP,
063                                TOKEN_MODEL_DEFAULT));
064                TokenizerModel model = null;
065                try {
066                        model = new TokenizerModel(modelIn);
067                } catch (final IOException e) {
068                        e.printStackTrace();
069                } finally {
070                        if (modelIn != null) {
071                                try {
072                                        modelIn.close();
073                                } catch (final IOException e) {
074                                }
075                        }
076                }
077                tokenizer = new TokenizerME(model);
078        }
079
080        @Override
081        void checkForRequiredAnnotations(RawTextAnnotation annotation)
082                        throws MissingRequiredAnnotationException
083        {
084
085        }
086
087        @Override
088        public List<TokenAnnotation> tokenise(String text) {
089                final List<TokenAnnotation> tla = new ArrayList<TokenAnnotation>();
090                int currentOff = 0;
091                for (final String token : tokenizer.tokenize(text)) {
092                        final int start = currentOff + (text.substring(currentOff).indexOf(token));
093                        final int stop = start + token.length();
094                        tla.add(new TokenAnnotation(token, text.substring(currentOff, stop), start, stop));
095                        currentOff = stop;
096                }
097                return tla;
098        }
099
100}