SolrMapper.java

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.io.*;
import org.apache.hadoop.mapreduce.Mapper;


public class SolrMapper extends Mapper<LongWritable, Text, Text, Text> {
  private HttpSolrServer solrServer;
  private Text outKey = new Text();
  
  @Override
  protected void setup(Context context) throws IOException, InterruptedException {
    super.setup(context);
    Configuration config = context.getConfiguration();
    solrServer = new HttpSolrServer(config.get("solr.url"));
  }
  
  @Override
  protected void map(LongWritable key, Text value, Context context) throws IOException, InterruptedException {
    try {
      File fileToProcess = new File(value.toString());
      //build SolrInputDocument from file
      SolrInputDocument solrInputDocument = null; //todo
      String docKey = null; //hang on to the document id for reporting
      solrServer.add(solrInputDocument);
      outKey.set(docKey);
      context.write(docKey, docKey);
    } catch (Exception e) {
      System.err.println("Caught exception processing " + value.toString());
      e.printStackTrace();
    }
  }
}