D-Net

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

import com.google.common.base.Joiner;

import com.google.protobuf.InvalidProtocolBufferException;

import com.googlecode.protobuf.format.JsonFormat;

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

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

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

import eu.dnetlib.data.proto.OafProtos;

import eu.dnetlib.data.proto.TypeProtos;

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.hbase.util.Bytes;

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 ExportInformationSpaceMergedUpdatesMapper extends TableMapper<Text, Text> {

	/**

	 * logger.

	 */

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

	private static final String SEPARATOR = "@";

	private Text keyOut;

	private Text valueOut;

	@Override

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

		super.setup(context);

		keyOut = new Text();

		valueOut = new Text();

	}

	@Override

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

		try {

			byte[] rowKey = keyIn.copyBytes();

			final OafRowKeyDecoder rkd = OafRowKeyDecoder.decode(keyIn.get());

			final OafProtos.Oaf oaf = mergeUpdates(value, context, rkd.getType());

			if (oaf == null) {

				return;

			}

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

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

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

					if(Bytes.toString(q.getKey()).startsWith("update")) {

						continue;

					}

					if (Bytes.toString(q.getKey()).equals("body")) {

						emit(rowKey, cf, q.getKey(), oaf.toByteArray(), context);

					} else {

						emit(rowKey, cf, q.getKey(), q.getValue(), 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 byte[] rowKey, final byte[] cf, final byte[] q, final byte[] value, final Context context) throws IOException, InterruptedException {

		keyOut.set(Joiner.on(SEPARATOR).join(new String(rowKey), new String(cf), new String(q)));

		if ((value == null) || (value.length == 0)) {

			valueOut.set("");

		} else {

			valueOut.set(new JsonFormat().printToString(OafDecoder.decode(value).getOaf()));

		}

		context.write(keyOut, valueOut);

	}

	private OafProtos.Oaf mergeUpdates(final Result value, final Context context, final TypeProtos.Type type)

			throws InvalidProtocolBufferException {

		try {

			return UpdateMerger.mergeBodyUpdates(context, value.getFamilyMap(Bytes.toBytes(type.toString())));

		} catch (final InvalidProtocolBufferException e) {

			throw e;

		}

	}

}