private static void createMapReduceJob(
String tableNameToIndex, Configuration conf, int caching, int versions)
throws IOException, InterruptedException, ClassNotFoundException {
// Set the details to TableInputFormat
Scan s = new Scan();
s.setCaching(caching);
s.setMaxVersions(versions);
conf.set(TableInputFormat.INPUT_TABLE, tableNameToIndex);
Set<Entry<String, List<String>>> entrySet = cfs.entrySet();
for (Entry<String, List<String>> entry : entrySet) {
List<String> quals = entry.getValue();
addColumn(quals, Bytes.toBytes(entry.getKey()), s);
}
Job job = new Job(conf, "CreateIndex");
String hfileOutPath = conf.get(BULK_OUTPUT_CONF_KEY);
TableMapReduceUtil.initTableMapperJob(
tableNameToIndex, // input table
s, // Scan instance to control CF and attribute selection
IndexCreationMapper.class, // mapper class
ImmutableBytesWritable.class, // mapper output key
Put.class, // mapper output value
job);
TableMapReduceUtil.initTableReducerJob(
IndexUtils.getIndexTableName(tableNameToIndex), // output
// table
null, // reducer class
job);
if (hfileOutPath != null) {
HTable table = new HTable(conf, tableNameToIndex);
job.setReducerClass(KeyValueSortReducer.class);
Path outputDir = new Path(hfileOutPath);
FileOutputFormat.setOutputPath(job, outputDir);
HFileOutputFormat.configureIncrementalLoad(job, table);
} else {
job.setNumReduceTasks(0);
}
TableMapReduceUtil.addDependencyJars(
job.getConfiguration(), com.google.common.base.Preconditions.class);
job.waitForCompletion(true);
assert job.isComplete() == true;
}
