View Javadoc

1   /**
2    *
3    * Licensed to the Apache Software Foundation (ASF) under one
4    * or more contributor license agreements.  See the NOTICE file
5    * distributed with this work for additional information
6    * regarding copyright ownership.  The ASF licenses this file
7    * to you under the Apache License, Version 2.0 (the
8    * "License"); you may not use this file except in compliance
9    * with the License.  You may obtain a copy of the License at
10   *
11   *     http://www.apache.org/licenses/LICENSE-2.0
12   *
13   * Unless required by applicable law or agreed to in writing, software
14   * distributed under the License is distributed on an "AS IS" BASIS,
15   * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16   * See the License for the specific language governing permissions and
17   * limitations under the License.
18   */
19  package org.apache.hadoop.hbase.regionserver;
20  
21  import java.io.Closeable;
22  import java.io.IOException;
23  
24  import org.apache.hadoop.hbase.classification.InterfaceAudience;
25  import org.apache.hadoop.hbase.Cell;
26  import org.apache.hadoop.hbase.KeyValue;
27  import org.apache.hadoop.hbase.client.Scan;
28  
29  /**
30   * Scanner that returns the next KeyValue.
31   */
32  @InterfaceAudience.Private
33  // TODO: Change name from KeyValueScanner to CellScanner only we already have a simple CellScanner
34  // so this should be something else altogether, a decoration on our base CellScanner. TODO.
35  // This class shows in CPs so do it all in one swell swoop. HBase-2.0.0.
36  public interface KeyValueScanner extends Shipper, Closeable {
37    /**
38     * The byte array represents for NO_NEXT_INDEXED_KEY;
39     * The actual value is irrelevant because this is always compared by reference.
40     */
41    public static final Cell NO_NEXT_INDEXED_KEY = new KeyValue();
42  
43    /**
44     * Look at the next Cell in this scanner, but do not iterate scanner.
45     * @return the next Cell
46     */
47    Cell peek();
48  
49    /**
50     * Return the next Cell in this scanner, iterating the scanner
51     * @return the next Cell
52     */
53    Cell next() throws IOException;
54  
55    /**
56     * Seek the scanner at or after the specified KeyValue.
57     * @param key seek value
58     * @return true if scanner has values left, false if end of scanner
59     */
60    boolean seek(Cell key) throws IOException;
61  
62    /**
63     * Reseek the scanner at or after the specified KeyValue.
64     * This method is guaranteed to seek at or after the required key only if the
65     * key comes after the current position of the scanner. Should not be used
66     * to seek to a key which may come before the current position.
67     * @param key seek value (should be non-null)
68     * @return true if scanner has values left, false if end of scanner
69     */
70    boolean reseek(Cell key) throws IOException;
71  
72    /**
73     * Get the order of this KeyValueScanner. This is only relevant for StoreFileScanners and
74     * MemStoreScanners (other scanners simply return 0). This is required for comparing multiple
75     * files to find out which one has the latest data. StoreFileScanners are ordered from 0
76     * (oldest) to newest in increasing order. MemStoreScanner gets LONG.max since it always
77     * contains freshest data.
78     */
79    long getScannerOrder();
80  
81    /**
82     * Close the KeyValue scanner.
83     */
84    void close();
85  
86    /**
87     * Allows to filter out scanners (both StoreFile and memstore) that we don't
88     * want to use based on criteria such as Bloom filters and timestamp ranges.
89     * @param scan the scan that we are selecting scanners for
90     * @param store the store we are performing the scan on.
91     * @param oldestUnexpiredTS the oldest timestamp we are interested in for
92     *          this query, based on TTL
93     * @return true if the scanner should be included in the query
94     */
95    boolean shouldUseScanner(Scan scan, Store store, long oldestUnexpiredTS);
96  
97    // "Lazy scanner" optimizations
98  
99    /**
100    * Similar to {@link #seek} (or {@link #reseek} if forward is true) but only
101    * does a seek operation after checking that it is really necessary for the
102    * row/column combination specified by the kv parameter. This function was
103    * added to avoid unnecessary disk seeks by checking row-column Bloom filters
104    * before a seek on multi-column get/scan queries, and to optimize by looking
105    * up more recent files first.
106    * @param forward do a forward-only "reseek" instead of a random-access seek
107    * @param useBloom whether to enable multi-column Bloom filter optimization
108    */
109   boolean requestSeek(Cell kv, boolean forward, boolean useBloom)
110       throws IOException;
111 
112   /**
113    * We optimize our store scanners by checking the most recent store file
114    * first, so we sometimes pretend we have done a seek but delay it until the
115    * store scanner bubbles up to the top of the key-value heap. This method is
116    * then used to ensure the top store file scanner has done a seek operation.
117    */
118   boolean realSeekDone();
119 
120   /**
121    * Does the real seek operation in case it was skipped by
122    * seekToRowCol(KeyValue, boolean) (TODO: Whats this?). Note that this function should
123    * be never called on scanners that always do real seek operations (i.e. most
124    * of the scanners). The easiest way to achieve this is to call
125    * {@link #realSeekDone()} first.
126    */
127   void enforceSeek() throws IOException;
128 
129   /**
130    * @return true if this is a file scanner. Otherwise a memory scanner is
131    *         assumed.
132    */
133   boolean isFileScanner();
134 
135   // Support for "Reversed Scanner"
136   /**
137    * Seek the scanner at or before the row of specified Cell, it firstly
138    * tries to seek the scanner at or after the specified Cell, return if
139    * peek KeyValue of scanner has the same row with specified Cell,
140    * otherwise seek the scanner at the first Cell of the row which is the
141    * previous row of specified KeyValue
142    *
143    * @param key seek KeyValue
144    * @return true if the scanner is at the valid KeyValue, false if such
145    *         KeyValue does not exist
146    *
147    */
148   public boolean backwardSeek(Cell key) throws IOException;
149 
150   /**
151    * Seek the scanner at the first Cell of the row which is the previous row
152    * of specified key
153    * @param key seek value
154    * @return true if the scanner at the first valid Cell of previous row,
155    *         false if not existing such Cell
156    */
157   public boolean seekToPreviousRow(Cell key) throws IOException;
158 
159   /**
160    * Seek the scanner at the first KeyValue of last row
161    *
162    * @return true if scanner has values left, false if the underlying data is
163    *         empty
164    * @throws IOException
165    */
166   public boolean seekToLastRow() throws IOException;
167 
168   /**
169    * @return the next key in the index, usually the first key of next block OR a key that falls
170    * between last key of current block and first key of next block..
171    * see HFileWriterImpl#getMidpoint, or null if not known.
172    */
173   public Cell getNextIndexedKey();
174 }