001/*
002 * Licensed to the Apache Software Foundation (ASF) under one
003 * or more contributor license agreements.  See the NOTICE file
004 * distributed with this work for additional information
005 * regarding copyright ownership.  The ASF licenses this file
006 * to you under the Apache License, Version 2.0 (the
007 * "License"); you may not use this file except in compliance
008 * with the License.  You may obtain a copy of the License at
009 *
010 *     http://www.apache.org/licenses/LICENSE-2.0
011 *
012 * Unless required by applicable law or agreed to in writing, software
013 * distributed under the License is distributed on an "AS IS" BASIS,
014 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
015 * See the License for the specific language governing permissions and
016 * limitations under the License.
017 */
018package org.apache.hadoop.hbase.mapreduce;
019
020import java.io.IOException;
021import java.util.Arrays;
022import java.util.List;
023import org.apache.hadoop.conf.Configuration;
024import org.apache.hadoop.fs.Path;
025import org.apache.hadoop.hbase.CompareOperator;
026import org.apache.hadoop.hbase.HConstants;
027import org.apache.hadoop.hbase.TableName;
028import org.apache.hadoop.hbase.client.Scan;
029import org.apache.hadoop.hbase.filter.Filter;
030import org.apache.hadoop.hbase.filter.IncompatibleFilterException;
031import org.apache.hadoop.hbase.filter.PrefixFilter;
032import org.apache.hadoop.hbase.filter.RegexStringComparator;
033import org.apache.hadoop.hbase.filter.RowFilter;
034import org.apache.hadoop.hbase.security.visibility.Authorizations;
035import org.apache.hadoop.hbase.util.Bytes;
036import org.apache.hadoop.hbase.util.Triple;
037import org.apache.hadoop.mapreduce.lib.output.FileOutputFormat;
038import org.apache.yetus.audience.InterfaceAudience;
039import org.slf4j.Logger;
040import org.slf4j.LoggerFactory;
041
042/**
043 * Some helper methods are used by {@link org.apache.hadoop.hbase.mapreduce.Export} and
044 * org.apache.hadoop.hbase.coprocessor.Export (in hbase-endpooint).
045 */
046@InterfaceAudience.Private
047public final class ExportUtils {
048  private static final Logger LOG = LoggerFactory.getLogger(ExportUtils.class);
049  public static final String RAW_SCAN = "hbase.mapreduce.include.deleted.rows";
050  public static final String EXPORT_BATCHING = "hbase.export.scanner.batch";
051  public static final String EXPORT_CACHING = "hbase.export.scanner.caching";
052  public static final String EXPORT_VISIBILITY_LABELS = "hbase.export.visibility.labels";
053
054  /**
055   * Common usage for other export tools.
056   * @param errorMsg Error message. Can be null.
057   */
058  public static void usage(final String errorMsg) {
059    if (errorMsg != null && errorMsg.length() > 0) {
060      System.err.println("ERROR: " + errorMsg);
061    }
062    System.err.println("Usage: Export [-D <property=value>]* <tablename> <outputdir> [<versions> "
063      + "[<starttime> [<endtime>]] [^[regex pattern] or [Prefix] to filter]]\n");
064    System.err.println("  Note: -D properties will be applied to the conf used. ");
065    System.err.println("  For example: ");
066    System.err.println("   -D " + FileOutputFormat.COMPRESS + "=true");
067    System.err.println(
068      "   -D " + FileOutputFormat.COMPRESS_CODEC + "=org.apache.hadoop.io.compress.GzipCodec");
069    System.err.println("   -D " + FileOutputFormat.COMPRESS_TYPE + "=BLOCK");
070    System.err.println("  Additionally, the following SCAN properties can be specified");
071    System.err.println("  to control/limit what is exported..");
072    System.err
073      .println("   -D " + TableInputFormat.SCAN_COLUMN_FAMILY + "=<family1>,<family2>, ...");
074    System.err.println("   -D " + RAW_SCAN + "=true");
075    System.err.println("   -D " + TableInputFormat.SCAN_ROW_START + "=<ROWSTART>");
076    System.err.println("   -D " + TableInputFormat.SCAN_ROW_STOP + "=<ROWSTOP>");
077    System.err.println("   -D " + HConstants.HBASE_CLIENT_SCANNER_CACHING + "=100");
078    System.err.println("   -D " + EXPORT_VISIBILITY_LABELS + "=<labels>");
079    System.err.println("For tables with very wide rows consider setting the batch size as below:\n"
080      + "   -D " + EXPORT_BATCHING + "=10\n" + "   -D " + EXPORT_CACHING + "=100");
081  }
082
083  private static Filter getExportFilter(String[] args) {
084    Filter exportFilter;
085    String filterCriteria = (args.length > 5) ? args[5] : null;
086    if (filterCriteria == null) return null;
087    if (filterCriteria.startsWith("^")) {
088      String regexPattern = filterCriteria.substring(1, filterCriteria.length());
089      exportFilter = new RowFilter(CompareOperator.EQUAL, new RegexStringComparator(regexPattern));
090    } else {
091      exportFilter = new PrefixFilter(Bytes.toBytesBinary(filterCriteria));
092    }
093    return exportFilter;
094  }
095
096  public static boolean isValidArguements(String[] args) {
097    return args != null && args.length >= 2;
098  }
099
100  public static Triple<TableName, Scan, Path> getArgumentsFromCommandLine(Configuration conf,
101    String[] args) throws IOException {
102    if (!isValidArguements(args)) {
103      return null;
104    }
105    return new Triple<>(TableName.valueOf(args[0]), getScanFromCommandLine(conf, args),
106      new Path(args[1]));
107  }
108
109  static Scan getScanFromCommandLine(Configuration conf, String[] args) throws IOException {
110    Scan s = new Scan();
111    // Optional arguments.
112    // Set Scan Versions
113    int versions = args.length > 2 ? Integer.parseInt(args[2]) : 1;
114    s.readVersions(versions);
115    // Set Scan Range
116    long startTime = args.length > 3 ? Long.parseLong(args[3]) : 0L;
117    long endTime = args.length > 4 ? Long.parseLong(args[4]) : Long.MAX_VALUE;
118    s.setTimeRange(startTime, endTime);
119    // Set cache blocks
120    s.setCacheBlocks(false);
121    // set Start and Stop row
122    if (conf.get(TableInputFormat.SCAN_ROW_START) != null) {
123      s.withStartRow(Bytes.toBytesBinary(conf.get(TableInputFormat.SCAN_ROW_START)));
124    }
125    if (conf.get(TableInputFormat.SCAN_ROW_STOP) != null) {
126      s.withStopRow(Bytes.toBytesBinary(conf.get(TableInputFormat.SCAN_ROW_STOP)));
127    }
128    // Set Scan Column Family
129    boolean raw = Boolean.parseBoolean(conf.get(RAW_SCAN));
130    if (raw) {
131      s.setRaw(raw);
132    }
133    for (String columnFamily : conf.getTrimmedStrings(TableInputFormat.SCAN_COLUMN_FAMILY)) {
134      s.addFamily(Bytes.toBytes(columnFamily));
135    }
136    // Set RowFilter or Prefix Filter if applicable.
137    Filter exportFilter = getExportFilter(args);
138    if (exportFilter != null) {
139      LOG.info("Setting Scan Filter for Export.");
140      s.setFilter(exportFilter);
141    }
142    List<String> labels = null;
143    if (conf.get(EXPORT_VISIBILITY_LABELS) != null) {
144      labels = Arrays.asList(conf.getStrings(EXPORT_VISIBILITY_LABELS));
145      if (!labels.isEmpty()) {
146        s.setAuthorizations(new Authorizations(labels));
147      }
148    }
149
150    int batching = conf.getInt(EXPORT_BATCHING, -1);
151    if (batching != -1) {
152      try {
153        s.setBatch(batching);
154      } catch (IncompatibleFilterException e) {
155        LOG.error("Batching could not be set", e);
156      }
157    }
158
159    int caching = conf.getInt(EXPORT_CACHING, 100);
160    if (caching != -1) {
161      try {
162        s.setCaching(caching);
163      } catch (IncompatibleFilterException e) {
164        LOG.error("Caching could not be set", e);
165      }
166    }
167    LOG.info("versions=" + versions + ", starttime=" + startTime + ", endtime=" + endTime
168      + ", keepDeletedCells=" + raw + ", visibility labels=" + labels);
169    return s;
170  }
171
172  private ExportUtils() {
173  }
174}