Java AccumuloInputFormat.setRanges方法代码示例

import org.apache.accumulo.core.client.mapreduce.AccumuloInputFormat; //导入方法依赖的package包/类
public Job getExpectedLoadJob(String inst, String zookeepers, String user,
        String password, String table, String start, String end,
        Authorizations authorizations,
        List<Pair<Text, Text>> columnFamilyColumnQualifierPairs)
        throws IOException {
    Collection<Range> ranges = new LinkedList<Range>();
    ranges.add(new Range(start, end));

    Job expected = new Job(new Configuration());

    try {
        AccumuloInputFormat.setConnectorInfo(expected, user,
                new PasswordToken(password));
    } catch (AccumuloSecurityException e) {
        Assert.fail(e.getMessage());
    }
    AccumuloInputFormat.setInputTableName(expected, table);
    AccumuloInputFormat.setScanAuthorizations(expected, authorizations);
    AccumuloInputFormat.setZooKeeperInstance(expected, inst, zookeepers);
    AccumuloInputFormat.fetchColumns(expected,
            columnFamilyColumnQualifierPairs);
    AccumuloInputFormat.setRanges(expected, ranges);

    return expected;
}
 

import org.apache.accumulo.core.client.mapreduce.AccumuloInputFormat; //导入方法依赖的package包/类
@Override
 public void setLocation(final String location, final Job job) throws IOException {
     if (logger.isDebugEnabled()) {
         logger.debug("Set Location[" + location + "] for job[" + job.getJobName() + "]");
     }
     conf = job.getConfiguration();
     setLocationFromUri(location, job);

     if (!ConfiguratorBase.isConnectorInfoSet(AccumuloInputFormat.class, conf)) {
         try {
	AccumuloInputFormat.setConnectorInfo(job, user, new PasswordToken(userP.getBytes(StandardCharsets.UTF_8)));
} catch (final AccumuloSecurityException e) {
	throw new RuntimeException(e);
}
         AccumuloInputFormat.setInputTableName(job, table);
 		AccumuloInputFormat.setScanAuthorizations(job, authorizations);
         if (!mock) {
             AccumuloInputFormat.setZooKeeperInstance(job, inst, zookeepers);
         } else {
             AccumuloInputFormat.setMockInstance(job, inst);
         }
     }
     if (columnFamilyColumnQualifierPairs.size() > 0) {
         AccumuloInputFormat.fetchColumns(job, columnFamilyColumnQualifierPairs);
     }
     logger.info("Set ranges[" + ranges + "] for job[" + job.getJobName() + "] on table[" + table + "] " +
             "for columns[" + columnFamilyColumnQualifierPairs + "] with authorizations[" + authorizations + "]");

     if (ranges.size() == 0) {
         throw new IOException("Accumulo Range must be specified");
     }
     AccumuloInputFormat.setRanges(job, ranges);
 }
 

import org.apache.accumulo.core.client.mapreduce.AccumuloInputFormat; //导入方法依赖的package包/类
/**
 * cloudbase props
 */

@Override
public int run(final String[] strings) throws Exception {
    conf.set(MRUtils.JOB_NAME_PROP, "Gather Evaluation Statistics");

    //initialize
    init();

    final Job job = new Job(conf);
    job.setJarByClass(AccumuloRdfCountTool.class);
    setupAccumuloInput(job);

    AccumuloInputFormat.setRanges(job, Lists.newArrayList(new Range(new Text(new byte[]{}), new Text(new byte[]{Byte.MAX_VALUE}))));
    // set input output of the particular job
    job.setMapOutputKeyClass(Text.class);
    job.setMapOutputValueClass(LongWritable.class);
    job.setOutputKeyClass(Text.class);
    job.setOutputValueClass(Mutation.class);

    // set mapper and reducer classes
    job.setMapperClass(CountPiecesMapper.class);
    job.setCombinerClass(CountPiecesCombiner.class);
    job.setReducerClass(CountPiecesReducer.class);

    final String outputTable = MRUtils.getTablePrefix(conf) + RdfCloudTripleStoreConstants.TBL_EVAL_SUFFIX;
    setupAccumuloOutput(job, outputTable);

    // Submit the job
    final Date startTime = new Date();
    System.out.println("Job started: " + startTime);
    final int exitCode = job.waitForCompletion(true) ? 0 : 1;

    if (exitCode == 0) {
        final Date end_time = new Date();
        System.out.println("Job ended: " + end_time);
        System.out.println("The job took "
                + (end_time.getTime() - startTime.getTime()) / 1000
                + " seconds.");
        return 0;
    } else {
        System.out.println("Job Failed!!!");
    }

    return -1;
}
 

import org.apache.accumulo.core.client.mapreduce.AccumuloInputFormat; //导入方法依赖的package包/类
@Override
public void setLocation(String location, Job job) throws IOException {
    setLocationFromUri(location);

    loadDependentJars(job.getConfiguration());

    Map<String, String> entries = getInputFormatEntries(job
            .getConfiguration());
    unsetEntriesFromConfiguration(job.getConfiguration(), entries);

    try {
        AccumuloInputFormat.setConnectorInfo(job, user, new PasswordToken(
                password));
    } catch (AccumuloSecurityException e) {
        throw new IOException(e);
    }

    AccumuloInputFormat.setInputTableName(job, table);
    AccumuloInputFormat.setScanAuthorizations(job, authorizations);
    AccumuloInputFormat.setZooKeeperInstance(job, inst, zookeepers);

    List<Pair<Text, Text>> inputFormatColumns = new LinkedList<Pair<Text, Text>>();
    int colfamPrefix = 0;

    for (Column c : columns) {
        switch (c.getType()) {
        case LITERAL:
            // Pull the colf[:colq] individually
            inputFormatColumns.add(makePair(c.getColumnFamily(),
                    c.getColumnQualifier()));
            break;
        case COLFAM_PREFIX:
            // Some colfams
            colfamPrefix++;
            break;
        case COLQUAL_PREFIX:
            // Some colquals in a given colfam
            inputFormatColumns.add(makePair(c.getColumnFamily(), null));
            break;
        default:
            log.info("Ignoring unhandled column type");
            break;
        }
    }

    // If we have colfam prefixes, we have to pull all columns and filter on
    // client-side
    // TODO Create an iterator that lets us push-down *all* of the filter
    // logic
    if (0 == colfamPrefix && !inputFormatColumns.isEmpty()) {
        AccumuloInputFormat.fetchColumns(job, inputFormatColumns);
    }

    Collection<Range> ranges = Collections.singleton(new Range(start, end));

    log.info("Scanning Accumulo for " + ranges + " for table " + table);

    AccumuloInputFormat.setRanges(job, ranges);

    configureInputFormat(job);
}