001/*
002 * (C) Copyright 2002-2007 Nuxeo SAS (http://nuxeo.com/) and contributors.
003 *
004 * All rights reserved. This program and the accompanying materials
005 * are made available under the terms of the GNU Lesser General Public License
006 * (LGPL) version 2.1 which accompanies this distribution, and is available at
007 * http://www.gnu.org/licenses/lgpl.html
008 *
009 * This library is distributed in the hope that it will be useful,
010 * but WITHOUT ANY WARRANTY; without even the implied warranty of
011 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
012 * Lesser General Public License for more details.
013 *
014 * Contributors:
015 *     Nuxeo - initial API and implementation
016 *
017 */
018package org.nuxeo.ecm.core.convert.plugins.text.extractors;
019
020import java.io.File;
021import java.io.FileInputStream;
022import java.io.FileOutputStream;
023import java.io.IOException;
024import java.io.InputStream;
025import java.io.OutputStream;
026import java.io.Serializable;
027import java.util.Map;
028
029import org.apache.poi.POITextExtractor;
030import org.apache.poi.extractor.ExtractorFactory;
031import org.apache.poi.openxml4j.exceptions.OpenXML4JException;
032import org.apache.xmlbeans.XmlException;
033import org.nuxeo.ecm.core.api.Blob;
034import org.nuxeo.ecm.core.api.Blobs;
035import org.nuxeo.ecm.core.api.blobholder.BlobHolder;
036import org.nuxeo.ecm.core.convert.api.ConversionException;
037import org.nuxeo.ecm.core.convert.cache.SimpleCachableBlobHolder;
038import org.nuxeo.ecm.core.convert.extension.Converter;
039import org.nuxeo.ecm.core.convert.extension.ConverterDescriptor;
040
041public class MSOffice2TextConverter implements Converter {
042
043    @Override
044    public BlobHolder convert(BlobHolder blobHolder, Map<String, Serializable> parameters) throws ConversionException {
045
046        File f = null;
047        OutputStream fas = null;
048
049        try {
050            POITextExtractor extractor = ExtractorFactory.createExtractor(blobHolder.getBlob().getStream());
051            // TODO: find a way to distinguish headings from paragraphs using
052            // WordExtractor#getParagraphText()?
053
054            // Get extracted text with Unix end of line characters
055            String extractedText = extractor.getText().replace("\r\n", "\n");
056
057            byte[] bytes = extractedText.getBytes("UTF-8");
058            f = File.createTempFile("po-msoffice2text", ".txt");
059            fas = new FileOutputStream(f);
060            fas.write(bytes);
061
062            try (InputStream is = new FileInputStream(f)) {
063                Blob blob = Blobs.createBlob(is, "text/plain", "UTF-8");
064                return new SimpleCachableBlobHolder(blob);
065            }
066        } catch (IOException | OpenXML4JException | XmlException e) {
067            throw new ConversionException("Error during MSOffice2Text conversion", e);
068        } finally {
069            if (fas != null) {
070                try {
071                    fas.close();
072                } catch (IOException e) {
073                }
074            }
075            if (f != null) {
076                f.delete();
077            }
078        }
079    }
080
081    @Override
082    public void init(ConverterDescriptor descriptor) {
083    }
084
085}