D-Net

package eu.dnetlib.data.mapreduce.hbase.dataimport;

import org.apache.commons.lang3.StringUtils;

import org.apache.commons.lang3.exception.ExceptionUtils;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.hadoop.io.Text;

import org.apache.hadoop.mapreduce.Mapper;

import org.dom4j.Document;

import org.dom4j.DocumentException;

import org.dom4j.io.DocumentResult;

import org.dom4j.io.DocumentSource;

import org.dom4j.io.SAXReader;

import javax.xml.transform.Transformer;

import javax.xml.transform.TransformerConfigurationException;

import javax.xml.transform.TransformerFactory;

import java.io.IOException;

import java.io.StringReader;

import java.util.regex.Matcher;

import java.util.regex.Pattern;

public class GetInvalidXmlRecordsMapper extends Mapper<Text, Text, Text, Text> {

	private static final Log log = LogFactory.getLog(GetInvalidXmlRecordsMapper.class); // NOPMD by marko on 11/24/08 5:02 PM

	public static final String DOI_REGEX = "(10[.][0-9]{4,}(?:[.][0-9]+)*/(?:(?![\\\"&\\'])\\S)+)";

	private Transformer transformer;

	private SAXReader saxReader;

	private Text valueOut;

	private final static String xslt =

			"<xsl:stylesheet version=\"1.0\" xmlns:xsl=\"http://www.w3.org/1999/XSL/Transform\">\n"

					+ "  <xsl:template match=\"@*|node()\">\n"

					+ "    <xsl:copy>\n"

					+ "      <xsl:apply-templates select=\"@*|node()\"/>\n"

					+ "    </xsl:copy>\n"

					+ "  </xsl:template>\n"

					+ "</xsl:stylesheet>";

	@Override

	protected void setup(final Context context) throws IOException, InterruptedException {

		super.setup(context);

		valueOut = new Text();

		saxReader = new SAXReader();

		log.info("using xslt:\n" + xslt);

		try {

			transformer = TransformerFactory.newInstance().newTransformer(new DocumentSource((new SAXReader()).read(new StringReader(xslt))));

		} catch (TransformerConfigurationException | DocumentException e) {

			log.error(e);

			throw new RuntimeException(e);

		}

		log.info("using trasformer: '" + transformer.getClass().getName() + "'");

	}

	@Override

	protected void map(final Text key, final Text value, final Context context) throws IOException, InterruptedException {

		try {

			final DocumentResult result = new DocumentResult();

			final Document document = saxReader.read(new StringReader(value.toString()));

			transformer.transform(new DocumentSource(document), result);

			result.getDocument().asXML();

		} catch (final Throwable e) {

			//log.error("error parsing record\n" + value.toString(), e);

			context.getCounter("error", e.getClass().getName()).increment(1);

			final String c = getInvalidXmlChar(e);

			if (StringUtils.isNotBlank(c)) {

				context.getCounter("invalid char", c).increment(1);

			}

			String doi = getDoi(value.toString());

			if (StringUtils.isNotBlank(doi)) {

				valueOut.set(doi);

				context.write(key, valueOut);

				context.getCounter("output", "doi").increment(1);

			}

		}

	}

	public static String getInvalidXmlChar(final Throwable e) {

		final String error = ExceptionUtils.getRootCauseMessage(e);

		if (StringUtils.contains(error, "An invalid XML character")) {

			final Pattern p = Pattern.compile(".*\\(.*:\\s?(?<char>.*)\\).*");

			final Matcher m = p.matcher(error);

			if (m.matches()) {

				final String c = m.group("char");

				if (StringUtils.isNotBlank(c)) {

					return c;

				}

			}

		}

		return null;

	}

	public static String getDoi(final String url) {

		final Pattern pattern = Pattern.compile(DOI_REGEX);

		final Matcher matcher = pattern.matcher(url);

		if (matcher.find())

			return matcher.group(0);

		return null;

	}

}