001/*
002 * (C) Copyright 2006-2007 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.IOException;
023import java.io.InputStream;
024import java.io.Serializable;
025import java.util.Map;
026
027import javax.xml.parsers.ParserConfigurationException;
028
029import org.apache.commons.logging.Log;
030import org.apache.commons.logging.LogFactory;
031import org.nuxeo.ecm.core.api.Blobs;
032import org.nuxeo.ecm.core.api.blobholder.BlobHolder;
033import org.nuxeo.ecm.core.convert.api.ConversionException;
034import org.nuxeo.ecm.core.convert.cache.SimpleCachableBlobHolder;
035import org.nuxeo.ecm.core.convert.extension.Converter;
036import org.nuxeo.ecm.core.convert.extension.ConverterDescriptor;
037import org.xml.sax.SAXException;
038
039public class XML2TextConverter implements Converter {
040
041    private static final Log log = LogFactory.getLog(XML2TextConverter.class);
042
043    @Override
044    public BlobHolder convert(BlobHolder blobHolder, Map<String, Serializable> parameters) throws ConversionException {
045
046        InputStream stream = null;
047        try {
048            stream = blobHolder.getBlob().getStream();
049            Xml2TextHandler xml2text = new Xml2TextHandler();
050            String text = xml2text.parse(stream);
051
052            return new SimpleCachableBlobHolder(Blobs.createBlob(text));
053        } catch (IOException | SAXException | ParserConfigurationException e) {
054            throw new ConversionException("Error during XML2Text conversion", e);
055        } finally {
056            if (stream != null) {
057                try {
058                    stream.close();
059                } catch (IOException e) {
060                    log.error("Error while closing Blob stream", e);
061                }
062            }
063        }
064    }
065
066    @Override
067    public void init(ConverterDescriptor descriptor) {
068    }
069
070}