1 | /** |
---|
2 | * Licensed to the Apache Software Foundation (ASF) under one |
---|
3 | * or more contributor license agreements. See the NOTICE file |
---|
4 | * distributed with this work for additional information |
---|
5 | * regarding copyright ownership. The ASF licenses this file |
---|
6 | * to you under the Apache License, Version 2.0 (the |
---|
7 | * "License"); you may not use this file except in compliance |
---|
8 | * with the License. You may obtain a copy of the License at |
---|
9 | * |
---|
10 | * http://www.apache.org/licenses/LICENSE-2.0 |
---|
11 | * |
---|
12 | * Unless required by applicable law or agreed to in writing, software |
---|
13 | * distributed under the License is distributed on an "AS IS" BASIS, |
---|
14 | * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. |
---|
15 | * See the License for the specific language governing permissions and |
---|
16 | * limitations under the License. |
---|
17 | */ |
---|
18 | |
---|
19 | package org.apache.hadoop.io; |
---|
20 | |
---|
21 | import junit.framework.TestCase; |
---|
22 | |
---|
23 | import java.nio.ByteBuffer; |
---|
24 | import java.nio.charset.CharacterCodingException; |
---|
25 | import java.util.Random; |
---|
26 | |
---|
27 | /** Unit tests for LargeUTF8. */ |
---|
28 | public class TestText extends TestCase { |
---|
29 | private static final int NUM_ITERATIONS = 100; |
---|
30 | public TestText(String name) { super(name); } |
---|
31 | |
---|
32 | private static final Random RANDOM = new Random(1); |
---|
33 | |
---|
34 | private static final int RAND_LEN = -1; |
---|
35 | |
---|
36 | // generate a valid java String |
---|
37 | private static String getTestString(int len) throws Exception { |
---|
38 | StringBuffer buffer = new StringBuffer(); |
---|
39 | int length = (len==RAND_LEN) ? RANDOM.nextInt(1000) : len; |
---|
40 | while (buffer.length()<length) { |
---|
41 | int codePoint = RANDOM.nextInt(Character.MAX_CODE_POINT); |
---|
42 | char tmpStr[] = new char[2]; |
---|
43 | if (Character.isDefined(codePoint)) { |
---|
44 | //unpaired surrogate |
---|
45 | if (codePoint < Character.MIN_SUPPLEMENTARY_CODE_POINT && |
---|
46 | !Character.isHighSurrogate((char)codePoint) && |
---|
47 | !Character.isLowSurrogate((char)codePoint)) { |
---|
48 | Character.toChars(codePoint, tmpStr, 0); |
---|
49 | buffer.append(tmpStr); |
---|
50 | } |
---|
51 | } |
---|
52 | } |
---|
53 | return buffer.toString(); |
---|
54 | } |
---|
55 | |
---|
56 | public static String getTestString() throws Exception { |
---|
57 | return getTestString(RAND_LEN); |
---|
58 | } |
---|
59 | |
---|
60 | public static String getLongString() throws Exception { |
---|
61 | String str = getTestString(); |
---|
62 | int length = Short.MAX_VALUE+str.length(); |
---|
63 | StringBuffer buffer = new StringBuffer(); |
---|
64 | while(buffer.length()<length) |
---|
65 | buffer.append(str); |
---|
66 | |
---|
67 | return buffer.toString(); |
---|
68 | } |
---|
69 | |
---|
70 | public void testWritable() throws Exception { |
---|
71 | for (int i = 0; i < NUM_ITERATIONS; i++) { |
---|
72 | String str; |
---|
73 | if (i == 0) |
---|
74 | str = getLongString(); |
---|
75 | else |
---|
76 | str = getTestString(); |
---|
77 | TestWritable.testWritable(new Text(str)); |
---|
78 | } |
---|
79 | } |
---|
80 | |
---|
81 | |
---|
82 | public void testCoding() throws Exception { |
---|
83 | String before = "Bad \t encoding \t testcase"; |
---|
84 | Text text = new Text(before); |
---|
85 | String after = text.toString(); |
---|
86 | assertTrue(before.equals(after)); |
---|
87 | |
---|
88 | for (int i = 0; i < NUM_ITERATIONS; i++) { |
---|
89 | // generate a random string |
---|
90 | if (i == 0) |
---|
91 | before = getLongString(); |
---|
92 | else |
---|
93 | before = getTestString(); |
---|
94 | |
---|
95 | // test string to utf8 |
---|
96 | ByteBuffer bb = Text.encode(before); |
---|
97 | |
---|
98 | byte[] utf8Text = bb.array(); |
---|
99 | byte[] utf8Java = before.getBytes("UTF-8"); |
---|
100 | assertEquals(0, WritableComparator.compareBytes( |
---|
101 | utf8Text, 0, bb.limit(), |
---|
102 | utf8Java, 0, utf8Java.length)); |
---|
103 | |
---|
104 | // test utf8 to string |
---|
105 | after = Text.decode(utf8Java); |
---|
106 | assertTrue(before.equals(after)); |
---|
107 | } |
---|
108 | } |
---|
109 | |
---|
110 | |
---|
111 | public void testIO() throws Exception { |
---|
112 | DataOutputBuffer out = new DataOutputBuffer(); |
---|
113 | DataInputBuffer in = new DataInputBuffer(); |
---|
114 | |
---|
115 | for (int i = 0; i < NUM_ITERATIONS; i++) { |
---|
116 | // generate a random string |
---|
117 | String before; |
---|
118 | if (i == 0) |
---|
119 | before = getLongString(); |
---|
120 | else |
---|
121 | before = getTestString(); |
---|
122 | |
---|
123 | // write it |
---|
124 | out.reset(); |
---|
125 | Text.writeString(out, before); |
---|
126 | |
---|
127 | // test that it reads correctly |
---|
128 | in.reset(out.getData(), out.getLength()); |
---|
129 | String after = Text.readString(in); |
---|
130 | assertTrue(before.equals(after)); |
---|
131 | |
---|
132 | // Test compatibility with Java's other decoder |
---|
133 | int strLenSize = WritableUtils.getVIntSize(Text.utf8Length(before)); |
---|
134 | String after2 = new String(out.getData(), strLenSize, |
---|
135 | out.getLength()-strLenSize, "UTF-8"); |
---|
136 | assertTrue(before.equals(after2)); |
---|
137 | } |
---|
138 | } |
---|
139 | |
---|
140 | public void testCompare() throws Exception { |
---|
141 | DataOutputBuffer out1 = new DataOutputBuffer(); |
---|
142 | DataOutputBuffer out2 = new DataOutputBuffer(); |
---|
143 | DataOutputBuffer out3 = new DataOutputBuffer(); |
---|
144 | Text.Comparator comparator = new Text.Comparator(); |
---|
145 | for (int i=0; i<NUM_ITERATIONS; i++) { |
---|
146 | // reset output buffer |
---|
147 | out1.reset(); |
---|
148 | out2.reset(); |
---|
149 | out3.reset(); |
---|
150 | |
---|
151 | // generate two random strings |
---|
152 | String str1 = getTestString(); |
---|
153 | String str2 = getTestString(); |
---|
154 | if (i == 0) { |
---|
155 | str1 = getLongString(); |
---|
156 | str2 = getLongString(); |
---|
157 | } else { |
---|
158 | str1 = getTestString(); |
---|
159 | str2 = getTestString(); |
---|
160 | } |
---|
161 | |
---|
162 | // convert to texts |
---|
163 | Text txt1 = new Text(str1); |
---|
164 | Text txt2 = new Text(str2); |
---|
165 | Text txt3 = new Text(str1); |
---|
166 | |
---|
167 | // serialize them |
---|
168 | txt1.write(out1); |
---|
169 | txt2.write(out2); |
---|
170 | txt3.write(out3); |
---|
171 | |
---|
172 | // compare two strings by looking at their binary formats |
---|
173 | int ret1 = comparator.compare(out1.getData(), 0, out1.getLength(), |
---|
174 | out2.getData(), 0, out2.getLength()); |
---|
175 | // compare two strings |
---|
176 | int ret2 = txt1.compareTo(txt2); |
---|
177 | |
---|
178 | assertEquals(ret1, ret2); |
---|
179 | |
---|
180 | // test equal |
---|
181 | assertEquals(txt1.compareTo(txt3), 0); |
---|
182 | assertEquals(comparator.compare(out1.getData(), 0, out3.getLength(), |
---|
183 | out3.getData(), 0, out3.getLength()), 0); |
---|
184 | } |
---|
185 | } |
---|
186 | |
---|
187 | public void testFind() throws Exception { |
---|
188 | Text text = new Text("abcd\u20acbdcd\u20ac"); |
---|
189 | assertTrue(text.find("abd")==-1); |
---|
190 | assertTrue(text.find("ac")==-1); |
---|
191 | assertTrue(text.find("\u20ac")==4); |
---|
192 | assertTrue(text.find("\u20ac", 5)==11); |
---|
193 | } |
---|
194 | |
---|
195 | public void testFindAfterUpdatingContents() throws Exception { |
---|
196 | Text text = new Text("abcd"); |
---|
197 | text.set("a".getBytes()); |
---|
198 | assertEquals(text.getLength(),1); |
---|
199 | assertEquals(text.find("a"), 0); |
---|
200 | assertEquals(text.find("b"), -1); |
---|
201 | } |
---|
202 | |
---|
203 | public void testValidate() throws Exception { |
---|
204 | Text text = new Text("abcd\u20acbdcd\u20ac"); |
---|
205 | byte [] utf8 = text.getBytes(); |
---|
206 | int length = text.getLength(); |
---|
207 | Text.validateUTF8(utf8, 0, length); |
---|
208 | } |
---|
209 | |
---|
210 | public void testTextText() throws CharacterCodingException { |
---|
211 | Text a=new Text("abc"); |
---|
212 | Text b=new Text("a"); |
---|
213 | b.set(a); |
---|
214 | assertEquals("abc", b.toString()); |
---|
215 | a.append("xdefgxxx".getBytes(), 1, 4); |
---|
216 | assertEquals("modified aliased string", "abc", b.toString()); |
---|
217 | assertEquals("appended string incorrectly", "abcdefg", a.toString()); |
---|
218 | } |
---|
219 | |
---|
220 | private class ConcurrentEncodeDecodeThread extends Thread { |
---|
221 | public ConcurrentEncodeDecodeThread(String name) { |
---|
222 | super(name); |
---|
223 | } |
---|
224 | |
---|
225 | public void run() { |
---|
226 | String name = this.getName(); |
---|
227 | DataOutputBuffer out = new DataOutputBuffer(); |
---|
228 | DataInputBuffer in = new DataInputBuffer(); |
---|
229 | for (int i=0; i < 1000; ++i) { |
---|
230 | try { |
---|
231 | out.reset(); |
---|
232 | WritableUtils.writeString(out, name); |
---|
233 | |
---|
234 | in.reset(out.getData(), out.getLength()); |
---|
235 | String s = WritableUtils.readString(in); |
---|
236 | |
---|
237 | assertEquals(name, s); |
---|
238 | } catch (Exception ioe) { |
---|
239 | throw new RuntimeException(ioe); |
---|
240 | } |
---|
241 | } |
---|
242 | } |
---|
243 | } |
---|
244 | |
---|
245 | public void testConcurrentEncodeDecode() throws Exception{ |
---|
246 | Thread thread1 = new ConcurrentEncodeDecodeThread("apache"); |
---|
247 | Thread thread2 = new ConcurrentEncodeDecodeThread("hadoop"); |
---|
248 | |
---|
249 | thread1.start(); |
---|
250 | thread2.start(); |
---|
251 | |
---|
252 | thread2.join(); |
---|
253 | thread2.join(); |
---|
254 | } |
---|
255 | |
---|
256 | public static void main(String[] args) throws Exception |
---|
257 | { |
---|
258 | TestText test = new TestText("main"); |
---|
259 | test.testIO(); |
---|
260 | test.testCompare(); |
---|
261 | test.testCoding(); |
---|
262 | test.testWritable(); |
---|
263 | test.testFind(); |
---|
264 | test.testValidate(); |
---|
265 | } |
---|
266 | } |
---|