D-Net

import java.io.StringReader;

import java.io.StringWriter;

import java.util.Iterator;

import java.util.List;

import javax.xml.stream.events.Namespace;

import javax.xml.stream.events.StartElement;

import javax.xml.stream.events.XMLEvent;

import org.apache.solr.common.SolrInputDocument;

/**

 * Optimized version of the document parser, drop in replacement of InputDocumentFactory.

 * <p>

 * Faster because:

 * </p>

 * <ul>

 * <li>Quickly serialize the 'result' element directly in a string.</li>

 * <li>Uses less memory: less pressure on GC and allows more threads to process this in parallel</li>

 * </ul>

 * <p>

 * This class is fully reentrant and can be invoked in parallel.

 * </p>

 *

 * @author marko

 */

public class StreamingInputDocumentFactory extends InputDocumentFactory {

	protected static final String DEFAULTDNETRESULT = "dnetResult";

	protected static final String ROOT_ELEMENT = "indexRecord";

	protected ThreadLocal<XMLInputFactory> inputFactory = new ThreadLocal<XMLInputFactory>() {

		@Override

		}

	};

	@Override

	public SolrInputDocument parseDocument(final String version, final String inputDocument, final String dsId, final String resultName)

			throws XMLStreamException {

		return parseDocument(version, inputDocument, dsId, resultName, null);

	}

	@Override

	public SolrInputDocument parseDocument(final String version,

			final String inputDocument,

			final ResultTransformer resultTransformer) {

		final StringWriter results = new StringWriter();

		try {

			XMLEventReader parser = inputFactory.get().createXMLEventReader(new StringReader(inputDocument));

			while (parser.hasNext()) {

				final XMLEvent event = parser.nextEvent();

	}

	private List<Namespace> getNamespaces(final XMLEvent event) {

		@SuppressWarnings("unchecked")

		Iterator<Namespace> nsIter = event.asStartElement().getNamespaces();

		while (nsIter.hasNext()) {

	 * Helper used to get the string from a text element.

	 *

	 * @param text

	 */

	protected final String getText(final XMLEvent text) {

		if (text.isEndElement()) // log.warn("skipping because isEndOfElement " + text.asEndElement().getName().getLocalPart());

			return "";

	}

}