View Javadoc

1   /**
2    * Copyright 2009 The Apache Software Foundation
3    *
4    * Licensed to the Apache Software Foundation (ASF) under one
5    * or more contributor license agreements.  See the NOTICE file
6    * distributed with this work for additional information
7    * regarding copyright ownership.  The ASF licenses this file
8    * to you under the Apache License, Version 2.0 (the
9    * "License"); you may not use this file except in compliance
10   * with the License.  You may obtain a copy of the License at
11   *
12   *     http://www.apache.org/licenses/LICENSE-2.0
13   *
14   * Unless required by applicable law or agreed to in writing, software
15   * distributed under the License is distributed on an "AS IS" BASIS,
16   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
17   * See the License for the specific language governing permissions and
18   * limitations under the License.
19   */
20  package org.apache.hadoop.hbase.regionserver;
21  
22  import java.io.IOException;
23  
24  import org.apache.hadoop.hbase.regionserver.ScanQueryMatcher.MatchCode;
25  
26  /**
27   * Implementing classes of this interface will be used for the tracking
28   * and enforcement of columns and numbers of versions and timeToLive during
29   * the course of a Get or Scan operation.
30   * <p>
31   * Currently there are two different types of Store/Family-level queries.
32   * <ul><li>{@link ExplicitColumnTracker} is used when the query specifies
33   * one or more column qualifiers to return in the family.
34   * <ul><li>{@link ScanWildcardColumnTracker} is used when no columns are
35   * explicitly specified.
36   * <p>
37   * This class is utilized by {@link ScanQueryMatcher} mainly through two methods:
38   * <ul><li>{@link #checkColumn} is called when a Put satisfies all other
39   * conditions of the query.
40   * <ul><li>{@link #getNextRowOrNextColumn} is called whenever ScanQueryMatcher
41   * believes that the current column should be skipped (by timestamp, filter etc.)
42   * <p>
43   * These two methods returns a 
44   * {@link org.apache.hadoop.hbase.regionserver.ScanQueryMatcher.MatchCode}
45   * to define what action should be taken.
46   * <p>
47   * This class is NOT thread-safe as queries are never multi-threaded
48   */
49  public interface ColumnTracker {
50  
51    /**
52     * Checks if the column is present in the list of requested columns by returning the match code
53     * instance. It does not check against the number of versions for the columns asked for. To do the
54     * version check, one has to call {@link #checkVersions(byte[], int, int, long, byte, boolean)}
55     * method based on the return type (INCLUDE) of this method. The values that can be returned by
56     * this method are {@link MatchCode#INCLUDE}, {@link MatchCode#SEEK_NEXT_COL} and
57     * {@link MatchCode#SEEK_NEXT_ROW}.
58     * @param bytes
59     * @param offset
60     * @param length
61     * @param type The type of the KeyValue
62     * @return The match code instance.
63     * @throws IOException in case there is an internal consistency problem caused by a data
64     *           corruption.
65     */
66    ScanQueryMatcher.MatchCode checkColumn(byte[] bytes, int offset, int length, byte type)
67        throws IOException;
68  
69    /**
70     * Keeps track of the number of versions for the columns asked for. It assumes that the user has
71     * already checked if the keyvalue needs to be included by calling the
72     * {@link #checkColumn(byte[], int, int, byte)} method. The enum values returned by this method
73     * are {@link MatchCode#SKIP}, {@link MatchCode#INCLUDE},
74     * {@link MatchCode#INCLUDE_AND_SEEK_NEXT_COL} and {@link MatchCode#INCLUDE_AND_SEEK_NEXT_ROW}.
75     * Implementations which include all the columns could just return {@link MatchCode#INCLUDE} in
76     * the {@link #checkColumn(byte[], int, int, byte)} method and perform all the operations in this
77     * checkVersions method.
78     * @param type the type of the key value (Put/Delete)
79     * @param ttl The timeToLive to enforce.
80     * @param ignoreCount indicates if the KV needs to be excluded while counting (used during
81     *          compactions. We only count KV's that are older than all the scanners' read points.)
82     * @return the scan query matcher match code instance
83     * @throws IOException in case there is an internal consistency problem caused by a data
84     *           corruption.
85     */
86    ScanQueryMatcher.MatchCode checkVersions(byte[] bytes, int offset, int length, long ttl,
87        byte type, boolean ignoreCount) throws IOException;
88  
89    /**
90     * Resets the Matcher
91     */
92    public void reset();
93  
94    /**
95     *
96     * @return <code>true</code> when done.
97     */
98    public boolean done();
99  
100   /**
101    * Used by matcher and scan/get to get a hint of the next column
102    * to seek to after checkColumn() returns SKIP.  Returns the next interesting
103    * column we want, or NULL there is none (wildcard scanner).
104    *
105    * Implementations aren't required to return anything useful unless the most recent
106    * call was to checkColumn() and the return code was SKIP.  This is pretty implementation
107    * detail-y, but optimizations are like that.
108    *
109    * @return null, or a ColumnCount that we should seek to
110    */
111   public ColumnCount getColumnHint();
112 
113   /**
114    * Retrieve the MatchCode for the next row or column
115    */
116   public MatchCode getNextRowOrNextColumn(byte[] bytes, int offset,
117       int qualLength);
118 
119   /**
120    * Give the tracker a chance to declare it's done based on only the timestamp
121    * to allow an early out.
122    *
123    * @param timestamp
124    * @return <code>true</code> to early out based on timestamp.
125    */
126   public boolean isDone(long timestamp);
127 }