D-Net

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

import com.google.common.base.Joiner;

import com.google.gson.Gson;

import com.googlecode.protobuf.format.JsonFormat;

import eu.dnetlib.data.mapreduce.util.OafDecoder;

import eu.dnetlib.data.proto.OafProtos;

import eu.dnetlib.dhp.schema.oaf.Oaf;

import eu.dnetlib.dhp.schema.util.ProtoConverter;

import eu.dnetlib.dhp.schema.util.ProtoUtils;

import org.apache.commons.logging.Log;

import org.apache.commons.logging.LogFactory;

import org.apache.hadoop.hbase.client.Result;

import org.apache.hadoop.hbase.io.ImmutableBytesWritable;

import org.apache.hadoop.hbase.mapreduce.TableMapper;

import org.apache.hadoop.io.Text;

import java.io.IOException;

import java.util.Map;

import java.util.Map.Entry;

import java.util.NavigableMap;

/**

 * Exports Oaf objects as their json serialization.

 *

 * @author claudio

 *

 */

public class ExportInformationSpaceMapper2DHP extends TableMapper<Text, Text> {

	/**

	 * logger.

	 */

	private static final Log log = LogFactory.getLog(ExportInformationSpaceMapper2DHP.class);

	private Text keyOut;

	private Text valueOut;

	private Gson g;

	@Override

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

		super.setup(context);

		keyOut = new Text();

		valueOut = new Text();

		g = new Gson();

	}

	@Override

	protected void map(final ImmutableBytesWritable keyIn, final Result value, final Context context) throws IOException, InterruptedException {

		try {

			Map<byte[], NavigableMap<byte[], byte[]>> row = value.getNoVersionMap();

			for (byte[] cf : row.keySet()) {

				for (Entry<byte[], byte[]> q : row.get(cf).entrySet()) {

					final OafProtos.Oaf oaf = OafDecoder.decode(q.getValue()).getOaf();

					Oaf result = null;

					try {

						result = ProtoConverter.convert(JsonFormat.printToString(oaf));

					} catch (Throwable e) {

						context.getCounter("Convert", "error" ).increment(1);

					}

					if (result != null) {

						emit(result.getClass().getName(), result, context);

					}

				}

			}

		} catch (final Throwable e) {

			log.error("error exporting the following record from HBase: " + value.toString(), e);

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

			throw new RuntimeException(e);

		}

	}

	private void emit(final String type,  final Oaf result, final Context context) throws IOException, InterruptedException {

		keyOut.set(type);

		valueOut.set(g.toJson(result));

		context.write(keyOut, valueOut);

	}

}