1
package org.apache.lucene.index;
4
* Licensed to the Apache Software Foundation (ASF) under one or more
5
* contributor license agreements. See the NOTICE file distributed with
6
* this work for additional information regarding copyright ownership.
7
* The ASF licenses this file to You under the Apache License, Version 2.0
8
* (the "License"); you may not use this file except in compliance with
9
* the License. You may obtain a copy of the License at
11
* http://www.apache.org/licenses/LICENSE-2.0
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.
20
import java.io.IOException;
21
import java.util.HashSet;
22
import java.util.Iterator;
23
import java.util.Random;
26
import org.apache.lucene.analysis.MockAnalyzer;
27
import org.apache.lucene.document.Document;
28
import org.apache.lucene.document.Field;
29
import org.apache.lucene.store.Directory;
30
import org.apache.lucene.util.BytesRef;
31
import org.apache.lucene.util.LuceneTestCase;
32
import org.apache.lucene.util.UnicodeUtil;
34
public class TestIndexWriterUnicode extends LuceneTestCase {
36
final String[] utf8Data = new String[] {
37
// unpaired low surrogate
38
"ab\udc17cd", "ab\ufffdcd",
39
"\udc17abcd", "\ufffdabcd",
41
"ab\udc17\udc17cd", "ab\ufffd\ufffdcd",
42
"\udc17\udc17abcd", "\ufffd\ufffdabcd",
43
"\udc17\udc17", "\ufffd\ufffd",
45
// unpaired high surrogate
46
"ab\ud917cd", "ab\ufffdcd",
47
"\ud917abcd", "\ufffdabcd",
49
"ab\ud917\ud917cd", "ab\ufffd\ufffdcd",
50
"\ud917\ud917abcd", "\ufffd\ufffdabcd",
51
"\ud917\ud917", "\ufffd\ufffd",
53
// backwards surrogates
54
"ab\udc17\ud917cd", "ab\ufffd\ufffdcd",
55
"\udc17\ud917abcd", "\ufffd\ufffdabcd",
56
"\udc17\ud917", "\ufffd\ufffd",
57
"ab\udc17\ud917\udc17\ud917cd", "ab\ufffd\ud917\udc17\ufffdcd",
58
"\udc17\ud917\udc17\ud917abcd", "\ufffd\ud917\udc17\ufffdabcd",
59
"\udc17\ud917\udc17\ud917", "\ufffd\ud917\udc17\ufffd"
62
private int nextInt(int lim) {
63
return random.nextInt(lim);
66
private int nextInt(int start, int end) {
67
return start + nextInt(end-start);
70
private boolean fillUnicode(char[] buffer, char[] expected, int offset, int count) {
71
final int len = offset + count;
72
boolean hasIllegal = false;
74
if (offset > 0 && buffer[offset] >= 0xdc00 && buffer[offset] < 0xe000)
75
// Don't start in the middle of a valid surrogate pair
78
for(int i=offset;i<len;i++) {
80
if (0 == t && i < len-1) {
81
// Make a surrogate pair
83
expected[i] = buffer[i++] = (char) nextInt(0xd800, 0xdc00);
85
expected[i] = buffer[i] = (char) nextInt(0xdc00, 0xe000);
87
expected[i] = buffer[i] = (char) nextInt(0x80);
89
expected[i] = buffer[i] = (char) nextInt(0x80, 0x800);
91
expected[i] = buffer[i] = (char) nextInt(0x800, 0xd800);
93
expected[i] = buffer[i] = (char) nextInt(0xe000, 0xffff);
94
else if (5 == t && i < len-1) {
95
// Illegal unpaired surrogate
96
if (nextInt(10) == 7) {
97
if (random.nextBoolean())
98
buffer[i] = (char) nextInt(0xd800, 0xdc00);
100
buffer[i] = (char) nextInt(0xdc00, 0xe000);
101
expected[i++] = 0xfffd;
102
expected[i] = buffer[i] = (char) nextInt(0x800, 0xd800);
105
expected[i] = buffer[i] = (char) nextInt(0x800, 0xd800);
107
expected[i] = buffer[i] = ' ';
114
// both start & end are inclusive
115
private final int getInt(Random r, int start, int end) {
116
return start + r.nextInt(1+end-start);
119
private final String asUnicodeChar(char c) {
120
return "U+" + Integer.toHexString(c);
123
private final String termDesc(String s) {
125
assertTrue(s.length() <= 2);
126
if (s.length() == 1) {
127
s0 = asUnicodeChar(s.charAt(0));
129
s0 = asUnicodeChar(s.charAt(0)) + "," + asUnicodeChar(s.charAt(1));
135
public void testRandomUnicodeStrings() throws Throwable {
136
char[] buffer = new char[20];
137
char[] expected = new char[20];
139
UnicodeUtil.UTF8Result utf8 = new UnicodeUtil.UTF8Result();
140
UnicodeUtil.UTF16Result utf16 = new UnicodeUtil.UTF16Result();
142
int num = atLeast(100000);
143
for (int iter = 0; iter < num; iter++) {
144
boolean hasIllegal = fillUnicode(buffer, expected, 0, 20);
146
UnicodeUtil.UTF16toUTF8(buffer, 0, 20, utf8);
148
byte[] b = new String(buffer, 0, 20).getBytes("UTF-8");
149
assertEquals(b.length, utf8.length);
150
for(int i=0;i<b.length;i++)
151
assertEquals(b[i], utf8.result[i]);
154
UnicodeUtil.UTF8toUTF16(utf8.result, 0, utf8.length, utf16);
155
assertEquals(utf16.length, 20);
156
for(int i=0;i<20;i++)
157
assertEquals(expected[i], utf16.result[i]);
162
public void testAllUnicodeChars() throws Throwable {
164
UnicodeUtil.UTF8Result utf8 = new UnicodeUtil.UTF8Result();
165
UnicodeUtil.UTF16Result utf16 = new UnicodeUtil.UTF16Result();
166
char[] chars = new char[2];
167
for(int ch=0;ch<0x0010FFFF;ch++) {
170
// Skip invalid code points
175
chars[len++] = (char) ch;
177
chars[len++] = (char) (((ch-0x0010000) >> 10) + UnicodeUtil.UNI_SUR_HIGH_START);
178
chars[len++] = (char) (((ch-0x0010000) & 0x3FFL) + UnicodeUtil.UNI_SUR_LOW_START);
181
UnicodeUtil.UTF16toUTF8(chars, 0, len, utf8);
183
String s1 = new String(chars, 0, len);
184
String s2 = new String(utf8.result, 0, utf8.length, "UTF-8");
185
assertEquals("codepoint " + ch, s1, s2);
187
UnicodeUtil.UTF8toUTF16(utf8.result, 0, utf8.length, utf16);
188
assertEquals("codepoint " + ch, s1, new String(utf16.result, 0, utf16.length));
190
byte[] b = s1.getBytes("UTF-8");
191
assertEquals(utf8.length, b.length);
192
for(int j=0;j<utf8.length;j++)
193
assertEquals(utf8.result[j], b[j]);
197
public void testEmbeddedFFFF() throws Throwable {
199
Directory d = newDirectory();
200
IndexWriter w = new IndexWriter(d, newIndexWriterConfig( TEST_VERSION_CURRENT, new TestIndexWriter.StringSplitAnalyzer()));
201
Document doc = new Document();
202
doc.add(newField("field", "a a\uffffb", Field.Store.NO, Field.Index.ANALYZED));
204
doc = new Document();
205
doc.add(newField("field", "a", Field.Store.NO, Field.Index.ANALYZED));
213
public void testInvalidUTF16() throws Throwable {
214
Directory dir = newDirectory();
215
IndexWriter w = new IndexWriter(dir, newIndexWriterConfig( TEST_VERSION_CURRENT, new TestIndexWriter.StringSplitAnalyzer()));
216
Document doc = new Document();
218
final int count = utf8Data.length/2;
219
for(int i=0;i<count;i++)
220
doc.add(newField("f" + i, utf8Data[2*i], Field.Store.YES, Field.Index.ANALYZED));
224
IndexReader ir = IndexReader.open(dir, true);
225
Document doc2 = ir.document(0);
226
for(int i=0;i<count;i++) {
227
assertEquals("field " + i + " was not indexed correctly", 1, ir.docFreq(new Term("f"+i, utf8Data[2*i+1])));
228
assertEquals("field " + i + " is incorrect", utf8Data[2*i+1], doc2.getField("f"+i).stringValue());
235
public void testIncrementalUnicodeStrings() throws Throwable {
236
char[] buffer = new char[20];
237
char[] expected = new char[20];
239
UnicodeUtil.UTF8Result utf8 = new UnicodeUtil.UTF8Result();
240
UnicodeUtil.UTF16Result utf16 = new UnicodeUtil.UTF16Result();
241
UnicodeUtil.UTF16Result utf16a = new UnicodeUtil.UTF16Result();
243
boolean hasIllegal = false;
244
byte[] last = new byte[60];
246
int num = atLeast(100000);
247
for (int iter = 0; iter < num; iter++) {
251
if (iter == 0 || hasIllegal)
254
prefix = nextInt(20);
256
hasIllegal = fillUnicode(buffer, expected, prefix, 20-prefix);
258
UnicodeUtil.UTF16toUTF8(buffer, 0, 20, utf8);
260
byte[] b = new String(buffer, 0, 20).getBytes("UTF-8");
261
assertEquals(b.length, utf8.length);
262
for(int i=0;i<b.length;i++)
263
assertEquals(b[i], utf8.result[i]);
267
if (iter == 0 || hasIllegal)
270
for(int i=0;i<20;i++)
271
if (last[i] != utf8.result[i]) {
275
System.arraycopy(utf8.result, 0, last, 0, utf8.length);
277
UnicodeUtil.UTF8toUTF16(utf8.result, bytePrefix, utf8.length-bytePrefix, utf16);
278
assertEquals(20, utf16.length);
279
for(int i=0;i<20;i++)
280
assertEquals(expected[i], utf16.result[i]);
282
UnicodeUtil.UTF8toUTF16(utf8.result, 0, utf8.length, utf16a);
283
assertEquals(20, utf16a.length);
284
for(int i=0;i<20;i++)
285
assertEquals(expected[i], utf16a.result[i]);