Java MultiRowRangeFilter.RowRange方法代码示例

import org.apache.hadoop.hbase.filter.MultiRowRangeFilter; //导入方法依赖的package包/类
private static List<MultiRowRangeFilter.RowRange> parseRowRangeParameter(
  String arg, String rangeSwitch) {
  final String[] ranges = arg.substring(rangeSwitch.length()).split(";");
  final List<MultiRowRangeFilter.RowRange> rangeList = new ArrayList<>();
  for (String range : ranges) {
    String[] startEnd = range.split(",", 2);
    if (startEnd.length != 2 || startEnd[1].contains(",")) {
      printUsage("Please specify range in such format as \"--range=a,b\" " +
          "or, with only one boundary, \"--range=,b\" or \"--range=a,\"");
      throw new IllegalArgumentException("Wrong range specification: " + range);
    }
    String startKey = startEnd[0];
    String endKey = startEnd[1];
    rangeList.add(new MultiRowRangeFilter.RowRange(
      Bytes.toBytesBinary(startKey), true,
      Bytes.toBytesBinary(endKey), false));
  }
  return rangeList;
}
 

import org.apache.hadoop.hbase.filter.MultiRowRangeFilter; //导入方法依赖的package包/类
/**
 * Sets filter {@link FilterBase} to the {@link Scan} instance.
 * If provided rowRangeList contains more than one element,
 * method sets filter which is instance of {@link MultiRowRangeFilter}.
 * Otherwise, method sets filter which is instance of {@link FirstKeyOnlyFilter}.
 * If rowRangeList contains exactly one element, startRow and stopRow are set to the scan.
 * @param scan
 * @param rowRangeList
 */
private static void setScanFilter(Scan scan, List<MultiRowRangeFilter.RowRange> rowRangeList) {
  final int size = rowRangeList == null ? 0 : rowRangeList.size();
  if (size <= 1) {
    scan.setFilter(new FirstKeyOnlyFilter());
  }
  if (size == 1) {
    MultiRowRangeFilter.RowRange range = rowRangeList.get(0);
    scan.setStartRow(range.getStartRow()); //inclusive
    scan.setStopRow(range.getStopRow());   //exclusive
  } else if (size > 1) {
    scan.setFilter(new MultiRowRangeFilter(rowRangeList));
  }
}
 

import org.apache.hadoop.hbase.filter.MultiRowRangeFilter; //导入方法依赖的package包/类
private CloseableIterable<Result> createScanner() {
    // End of input ids
    if (null != idsIterator && !idsIterator.hasNext()) {
        return null;
    }

    Table table = null;
    try {
        final Scan scan = new Scan();

        if (null != idsIterator) {
            final List<MultiRowRangeFilter.RowRange> rowRanges = new ArrayList<>();
            final int maxEntriesForBatchScanner = store.getProperties().getMaxEntriesForBatchScanner();
            int count = 0;
            while (idsIterator.hasNext() && count < maxEntriesForBatchScanner) {
                count++;
                rowRanges.addAll(rowRangeFactory.getRowRange(idsIterator.next(), operation));
            }

            if (rowRanges.isEmpty()) {
                return new WrappedCloseableIterable<>(Collections.emptyList());
            }

            scan.setFilter(new MultiRowRangeFilter(rowRanges));
        }

        scan.setAuthorizations(authorisations);
        scan.setAttribute(HBaseStoreConstants.SCHEMA, store.getSchema().toCompactJson());
        scan.setAttribute(HBaseStoreConstants.INCLUDE_MATCHED_VERTEX, Bytes.toBytes(Boolean.toString(includeMatchedVertex)));
        scan.setAttribute(HBaseStoreConstants.VIEW, operation.getView().toCompactJson());
        if (null != operation.getDirectedType()) {
            scan.setAttribute(HBaseStoreConstants.DIRECTED_TYPE, Bytes.toBytes(operation.getDirectedType().name()));
        }
        if (null != extraProcessors) {
            scan.setAttribute(HBaseStoreConstants.EXTRA_PROCESSORS, extraProcessors);
        }
        scan.setMaxVersions();
        table = store.getTable();
        return new WrappedCloseableIterable<>(table.getScanner(scan));
    } catch (final IOException | StoreException e) {
        if (null != table) {
            CloseableUtil.close(table);
        }
        throw new RuntimeException(e);
    }
}
 

import org.apache.hadoop.hbase.filter.MultiRowRangeFilter; //导入方法依赖的package包/类
/**
 * Sets up the actual job.
 *
 * @param conf  The current configuration.
 * @param args  The command line parameters.
 * @return The newly created job.
 * @throws IOException When setting up the job fails.
 */
public static Job createSubmittableJob(Configuration conf, String[] args)
throws IOException {
  String tableName = args[0];
  List<MultiRowRangeFilter.RowRange> rowRangeList = null;
  long startTime = 0;
  long endTime = 0;

  StringBuilder sb = new StringBuilder();

  final String rangeSwitch = "--range=";
  final String startTimeArgKey = "--starttime=";
  final String endTimeArgKey = "--endtime=";
  final String expectedCountArg = "--expected-count=";

  // First argument is table name, starting from second
  for (int i = 1; i < args.length; i++) {
    if (args[i].startsWith(rangeSwitch)) {
      try {
        rowRangeList = parseRowRangeParameter(args[i], rangeSwitch);
      } catch (IllegalArgumentException e) {
        return null;
      }
      continue;
    }
    if (args[i].startsWith(startTimeArgKey)) {
      startTime = Long.parseLong(args[i].substring(startTimeArgKey.length()));
      continue;
    }
    if (args[i].startsWith(endTimeArgKey)) {
      endTime = Long.parseLong(args[i].substring(endTimeArgKey.length()));
      continue;
    }
    if (args[i].startsWith(expectedCountArg)) {
      conf.setLong(EXPECTED_COUNT_KEY,
          Long.parseLong(args[i].substring(expectedCountArg.length())));
      continue;
    }
    // if no switch, assume column names
    sb.append(args[i]);
    sb.append(" ");
  }
  if (endTime < startTime) {
    printUsage("--endtime=" + endTime + " needs to be greater than --starttime=" + startTime);
    return null;
  }

  Job job = Job.getInstance(conf, conf.get(JOB_NAME_CONF_KEY, NAME + "_" + tableName));
  job.setJarByClass(RowCounter.class);
  Scan scan = new Scan();
  scan.setCacheBlocks(false);
  setScanFilter(scan, rowRangeList);
  if (sb.length() > 0) {
    for (String columnName : sb.toString().trim().split(" ")) {
      String family = StringUtils.substringBefore(columnName, ":");
      String qualifier = StringUtils.substringAfter(columnName, ":");

      if (StringUtils.isBlank(qualifier)) {
        scan.addFamily(Bytes.toBytes(family));
      }
      else {
        scan.addColumn(Bytes.toBytes(family), Bytes.toBytes(qualifier));
      }
    }
  }
  scan.setTimeRange(startTime, endTime == 0 ? HConstants.LATEST_TIMESTAMP : endTime);
  job.setOutputFormatClass(NullOutputFormat.class);
  TableMapReduceUtil.initTableMapperJob(tableName, scan,
    RowCounterMapper.class, ImmutableBytesWritable.class, Result.class, job);
  job.setNumReduceTasks(0);
  return job;
}