001/*
002 * (C) Copyright 2006-2016 Nuxeo SA (http://nuxeo.com/) and others.
003 *
004 * Licensed under the Apache License, Version 2.0 (the "License");
005 * you may not use this file except in compliance with the License.
006 * You may obtain a copy of the License at
007 *
008 *     http://www.apache.org/licenses/LICENSE-2.0
009 *
010 * Unless required by applicable law or agreed to in writing, software
011 * distributed under the License is distributed on an "AS IS" BASIS,
012 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
013 * See the License for the specific language governing permissions and
014 * limitations under the License.
015 *
016 * Contributors:
017 *     Nuxeo - initial API and implementation
018 *
019 */
020package org.nuxeo.ecm.core.convert.plugins.text.extractors;
021
022import java.io.File;
023import java.io.FileInputStream;
024import java.io.FileOutputStream;
025import java.io.IOException;
026import java.io.InputStream;
027import java.io.OutputStream;
028import java.io.Serializable;
029import java.util.Map;
030
031import org.apache.poi.POITextExtractor;
032import org.apache.poi.extractor.ExtractorFactory;
033import org.apache.poi.openxml4j.exceptions.OpenXML4JException;
034import org.apache.xmlbeans.XmlException;
035
036import org.nuxeo.ecm.core.api.Blob;
037import org.nuxeo.ecm.core.api.Blobs;
038import org.nuxeo.ecm.core.api.blobholder.BlobHolder;
039import org.nuxeo.ecm.core.convert.api.ConversionException;
040import org.nuxeo.ecm.core.convert.cache.SimpleCachableBlobHolder;
041import org.nuxeo.ecm.core.convert.extension.Converter;
042import org.nuxeo.ecm.core.convert.extension.ConverterDescriptor;
043import org.nuxeo.runtime.api.Framework;
044
045public class MSOffice2TextConverter implements Converter {
046
047    @Override
048    public BlobHolder convert(BlobHolder blobHolder, Map<String, Serializable> parameters) throws ConversionException {
049
050        File f = null;
051
052        try (POITextExtractor extractor = ExtractorFactory.createExtractor(blobHolder.getBlob().getStream())) {
053            // TODO: find a way to distinguish headings from paragraphs using
054            // WordExtractor#getParagraphText()?
055
056            // Get extracted text with Unix end of line characters
057            String extractedText = extractor.getText().replace("\r\n", "\n");
058
059            byte[] bytes = extractedText.getBytes("UTF-8");
060            f = Framework.createTempFile("po-msoffice2text", ".txt");
061            try (OutputStream fas = new FileOutputStream(f)) {
062                fas.write(bytes);
063            }
064
065            try (InputStream is = new FileInputStream(f)) {
066                Blob blob = Blobs.createBlob(is, "text/plain", "UTF-8");
067                return new SimpleCachableBlobHolder(blob);
068            }
069        } catch (IOException | OpenXML4JException | XmlException e) {
070            throw new ConversionException("Error during MSOffice2Text conversion", e);
071        } finally {
072            if (f != null) {
073                f.delete();
074            }
075        }
076    }
077
078    @Override
079    public void init(ConverterDescriptor descriptor) {
080    }
081
082}