D-Net

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

import com.google.gson.JsonObject;

import com.google.gson.JsonParser;

import eu.dnetlib.actionmanager.actions.ActionFactory;

import eu.dnetlib.actionmanager.actions.AtomicAction;

import eu.dnetlib.actionmanager.common.Agent;

import org.apache.hadoop.io.LongWritable;

import org.apache.hadoop.io.Text;

import org.apache.hadoop.mapreduce.Mapper;

import java.io.IOException;

public class CrossRefImportMapper extends Mapper<LongWritable, Text, Text, Text> {

    private String setName;

    private Agent agent;

    private Text keyout;

    private Text valueOut;

    private JsonParser parser;

    private ActionFactory factory;

    private boolean invisible;

    @Override

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

        setName = context.getConfiguration().get("setName");

        agent = new Agent(context.getConfiguration().get("agentId"), context.getConfiguration().get("agentName"), Agent.AGENT_TYPE.service);

        keyout = new Text("");

        valueOut = new Text("");

        factory = new ActionFactory();

        parser = new JsonParser();

        invisible = context.getConfiguration().getBoolean("invisible", false);

    }

    @Override

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

        final String inputJson = value.toString();

        final JsonObject rootElement = parser.parse(inputJson).getAsJsonObject();

        try {

            final AtomicAction action = CrossRefToActions.generateActionsFromDump(rootElement, factory, setName, agent, invisible);

            keyout.set(action.getRowKey());

            valueOut.set(action.toJSON());

            context.write(keyout, valueOut);

        } catch (Throwable e) {

            System.err.println(inputJson);

            throw e;

        }

    }

}