Metacat

package edu.ucsb.nceas.metacat.index;

import java.io.Reader;

import edu.ucsb.nceas.metacat.index.D1IndexField.DataFormat;

import edu.ucsb.nceas.metacat.index.D1IndexField.Conversion;

/*

 * Simple xpath-based indexing. Accepts DOM-parseable data, extracts text data from all

 * nodes selected by an XPath expression concatenated into a single Lucene/SOLR field

 * body. Equivalent to the DataONE indexer's MergeSolrField bean.

 *

 * Currently just wraps DataONE SolrField code.

 */

public class XpathIndexField extends FieldSpec {

	private static final String textSelector = "text()";

	public final String xpath;

	public XpathIndexField(String name, String xp) {

		super(name);

		this.xpath = xp;

	}

	@Override

	public String[] extract(Reader in) {

		String s = this.xpath;

		if (!this.xpath.endsWith(textSelector)) {

			if (this.xpath.charAt(this.xpath.length()-1) != '/') {

				s = s + "/";

			}

			s = s + textSelector;

		}

		D1IndexField field = new D1IndexField(this.name, s, DataFormat.SINGLE, Conversion.NONE);

		return field.extract(in);

	}

}