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.IOException;
22 import java.util.NavigableSet;
23
24 import org.apache.hadoop.hbase.classification.InterfaceAudience;
25 import org.apache.hadoop.hbase.CellUtil;
26 import org.apache.hadoop.hbase.HConstants;
27 import org.apache.hadoop.hbase.regionserver.ScanQueryMatcher.MatchCode;
28 import org.apache.hadoop.hbase.util.Bytes;
29
30 /**
31 * This class is used for the tracking and enforcement of columns and numbers
32 * of versions during the course of a Get or Scan operation, when explicit
33 * column qualifiers have been asked for in the query.
34 *
35 * With a little magic (see {@link ScanQueryMatcher}), we can use this matcher
36 * for both scans and gets. The main difference is 'next' and 'done' collapse
37 * for the scan case (since we see all columns in order), and we only reset
38 * between rows.
39 *
40 * <p>
41 * This class is utilized by {@link ScanQueryMatcher} mainly through two methods:
42 * <ul><li>{@link #checkColumn} is called when a Put satisfies all other
43 * conditions of the query.
44 * <ul><li>{@link #getNextRowOrNextColumn} is called whenever ScanQueryMatcher
45 * believes that the current column should be skipped (by timestamp, filter etc.)
46 * <p>
47 * These two methods returns a
48 * {@link org.apache.hadoop.hbase.regionserver.ScanQueryMatcher.MatchCode}
49 * to define what action should be taken.
50 * <p>
51 * This class is NOT thread-safe as queries are never multi-threaded
52 */
53 @InterfaceAudience.Private
54 public class ExplicitColumnTracker implements ColumnTracker {
55
56 private final int maxVersions;
57 private final int minVersions;
58
59 /**
60 * Contains the list of columns that the ExplicitColumnTracker is tracking.
61 * Each ColumnCount instance also tracks how many versions of the requested
62 * column have been returned.
63 */
64 private final ColumnCount[] columns;
65 private int index;
66 private ColumnCount column;
67 /** Keeps track of the latest timestamp included for current column.
68 * Used to eliminate duplicates. */
69 private long latestTSOfCurrentColumn;
70 private long oldestStamp;
71
72 /**
73 * Default constructor.
74 * @param columns columns specified user in query
75 * @param minVersions minimum number of versions to keep
76 * @param maxVersions maximum versions to return per column
77 * @param oldestUnexpiredTS the oldest timestamp we are interested in,
78 * based on TTL
79 */
80 public ExplicitColumnTracker(NavigableSet<byte[]> columns, int minVersions,
81 int maxVersions, long oldestUnexpiredTS) {
82 this.maxVersions = maxVersions;
83 this.minVersions = minVersions;
84 this.oldestStamp = oldestUnexpiredTS;
85 this.columns = new ColumnCount[columns.size()];
86 int i=0;
87 for(byte [] column : columns) {
88 this.columns[i++] = new ColumnCount(column);
89 }
90 reset();
91 }
92
93 /**
94 * Done when there are no more columns to match against.
95 */
96 public boolean done() {
97 return this.index >= columns.length;
98 }
99
100 public ColumnCount getColumnHint() {
101 return this.column;
102 }
103
104 /**
105 * {@inheritDoc}
106 */
107 @Override
108 public ScanQueryMatcher.MatchCode checkColumn(byte [] bytes, int offset,
109 int length, byte type) {
110 // delete markers should never be passed to an
111 // *Explicit*ColumnTracker
112 assert !CellUtil.isDelete(type);
113 do {
114 // No more columns left, we are done with this query
115 if(done()) {
116 return ScanQueryMatcher.MatchCode.SEEK_NEXT_ROW; // done_row
117 }
118
119 // No more columns to match against, done with storefile
120 if(this.column == null) {
121 return ScanQueryMatcher.MatchCode.SEEK_NEXT_ROW; // done_row
122 }
123
124 // Compare specific column to current column
125 int ret = Bytes.compareTo(column.getBuffer(), column.getOffset(),
126 column.getLength(), bytes, offset, length);
127
128 // Column Matches. Return include code. The caller would call checkVersions
129 // to limit the number of versions.
130 if(ret == 0) {
131 return ScanQueryMatcher.MatchCode.INCLUDE;
132 }
133
134 resetTS();
135
136 if (ret > 0) {
137 // The current KV is smaller than the column the ExplicitColumnTracker
138 // is interested in, so seek to that column of interest.
139 return ScanQueryMatcher.MatchCode.SEEK_NEXT_COL;
140 }
141
142 // The current KV is bigger than the column the ExplicitColumnTracker
143 // is interested in. That means there is no more data for the column
144 // of interest. Advance the ExplicitColumnTracker state to next
145 // column of interest, and check again.
146 if (ret <= -1) {
147 ++this.index;
148 if (done()) {
149 // No more to match, do not include, done with this row.
150 return ScanQueryMatcher.MatchCode.SEEK_NEXT_ROW; // done_row
151 }
152 // This is the recursive case.
153 this.column = this.columns[this.index];
154 }
155 } while(true);
156 }
157
158 @Override
159 public ScanQueryMatcher.MatchCode checkVersions(byte[] bytes, int offset, int length,
160 long timestamp, byte type, boolean ignoreCount) throws IOException {
161 assert !CellUtil.isDelete(type);
162 if (ignoreCount) return ScanQueryMatcher.MatchCode.INCLUDE;
163 // Check if it is a duplicate timestamp
164 if (sameAsPreviousTS(timestamp)) {
165 // If duplicate, skip this Key
166 return ScanQueryMatcher.MatchCode.SKIP;
167 }
168 int count = this.column.increment();
169 if (count >= maxVersions || (count >= minVersions && isExpired(timestamp))) {
170 // Done with versions for this column
171 ++this.index;
172 resetTS();
173 if (done()) {
174 // We have served all the requested columns.
175 this.column = null;
176 return ScanQueryMatcher.MatchCode.INCLUDE_AND_SEEK_NEXT_ROW;
177 }
178 // We are done with current column; advance to next column
179 // of interest.
180 this.column = this.columns[this.index];
181 return ScanQueryMatcher.MatchCode.INCLUDE_AND_SEEK_NEXT_COL;
182 }
183 setTS(timestamp);
184 return ScanQueryMatcher.MatchCode.INCLUDE;
185 }
186
187 // Called between every row.
188 public void reset() {
189 this.index = 0;
190 this.column = this.columns[this.index];
191 for(ColumnCount col : this.columns) {
192 col.setCount(0);
193 }
194 resetTS();
195 }
196
197 private void resetTS() {
198 latestTSOfCurrentColumn = HConstants.LATEST_TIMESTAMP;
199 }
200
201 private void setTS(long timestamp) {
202 latestTSOfCurrentColumn = timestamp;
203 }
204
205 private boolean sameAsPreviousTS(long timestamp) {
206 return timestamp == latestTSOfCurrentColumn;
207 }
208
209 private boolean isExpired(long timestamp) {
210 return timestamp < oldestStamp;
211 }
212
213 /**
214 * This method is used to inform the column tracker that we are done with
215 * this column. We may get this information from external filters or
216 * timestamp range and we then need to indicate this information to
217 * tracker. It is required only in case of ExplicitColumnTracker.
218 * @param bytes
219 * @param offset
220 * @param length
221 */
222 public void doneWithColumn(byte [] bytes, int offset, int length) {
223 while (this.column != null) {
224 int compare = Bytes.compareTo(column.getBuffer(), column.getOffset(),
225 column.getLength(), bytes, offset, length);
226 resetTS();
227 if (compare <= 0) {
228 ++this.index;
229 if (done()) {
230 // Will not hit any more columns in this storefile
231 this.column = null;
232 } else {
233 this.column = this.columns[this.index];
234 }
235 if (compare <= -1)
236 continue;
237 }
238 return;
239 }
240 }
241
242 public MatchCode getNextRowOrNextColumn(byte[] bytes, int offset,
243 int qualLength) {
244 doneWithColumn(bytes, offset,qualLength);
245
246 if (getColumnHint() == null) {
247 return MatchCode.SEEK_NEXT_ROW;
248 } else {
249 return MatchCode.SEEK_NEXT_COL;
250 }
251 }
252
253 public boolean isDone(long timestamp) {
254 return minVersions <= 0 && isExpired(timestamp);
255 }
256 }