1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /********************************************************************
4 * COPYRIGHT:
5 * Copyright (c) 1997-2010, International Business Machines Corporation and
6 * others. All Rights Reserved.
7 ********************************************************************/
8
9 #include "unicode/utypes.h"
10 #include "unicode/uchar.h"
11 #include "unicode/normlzr.h"
12 #include "unicode/uniset.h"
13 #include "unicode/usetiter.h"
14 #include "unicode/schriter.h"
15 #include "tstnorm.h"
16
17 #if !UCONFIG_NO_NORMALIZATION
18
19 static UErrorCode status = U_ZERO_ERROR;
20
21 // test APIs that are not otherwise used - improve test coverage
22 void
TestNormalizerAPI()23 BasicNormalizerTest::TestNormalizerAPI() {
24 // instantiate a Normalizer from a CharacterIterator
25 UnicodeString s=UnicodeString("a\\u0308\\uac00\\U0002f800", "").unescape();
26 s.append(s); // make s a bit longer and more interesting
27 StringCharacterIterator iter(s);
28 Normalizer norm(iter, UNORM_NFC);
29 if(norm.next()!=0xe4) {
30 dataerrln("error in Normalizer(CharacterIterator).next()");
31 }
32
33 // test copy constructor
34 Normalizer copy(norm);
35 if(copy.next()!=0xac00) {
36 dataerrln("error in Normalizer(Normalizer(CharacterIterator)).next()");
37 }
38
39 // test clone(), ==, and hashCode()
40 Normalizer *clone=copy.clone();
41 if(*clone!=copy) {
42 errln("error in Normalizer(Normalizer(CharacterIterator)).clone()!=copy");
43 }
44 // clone must have the same hashCode()
45 if(clone->hashCode()!=copy.hashCode()) {
46 errln("error in Normalizer(Normalizer(CharacterIterator)).clone()->hashCode()!=copy.hashCode()");
47 }
48 if(clone->next()!=0x4e3d) {
49 dataerrln("error in Normalizer(Normalizer(CharacterIterator)).clone()->next()");
50 }
51 // position changed, must change hashCode()
52 if(clone->hashCode()==copy.hashCode()) {
53 errln("error in Normalizer(Normalizer(CharacterIterator)).clone()->next().hashCode()==copy.hashCode()");
54 }
55 delete clone;
56 clone=0;
57
58 // test compose() and decompose()
59 UnicodeString tel, nfkc, nfkd;
60 tel=UnicodeString(1, (UChar32)0x2121, 10);
61 tel.insert(1, (UChar)0x301);
62
63 UErrorCode errorCode=U_ZERO_ERROR;
64 Normalizer::compose(tel, true, 0, nfkc, errorCode);
65 Normalizer::decompose(tel, true, 0, nfkd, errorCode);
66 if(U_FAILURE(errorCode)) {
67 dataerrln("error in Normalizer::(de)compose(): %s", u_errorName(errorCode));
68 } else if(
69 nfkc!=UnicodeString("TE\\u0139TELTELTELTELTELTELTELTELTEL", "").unescape() ||
70 nfkd!=UnicodeString("TEL\\u0301TELTELTELTELTELTELTELTELTEL", "").unescape()
71 ) {
72 errln("error in Normalizer::(de)compose(): wrong result(s)");
73 }
74
75 // test setIndex()
76 norm.setIndexOnly(3);
77 if(norm.current()!=0x4e3d) {
78 dataerrln("error in Normalizer(CharacterIterator).setIndex(3)");
79 }
80
81 // test setText(CharacterIterator) and getText()
82 UnicodeString out, out2;
83 errorCode=U_ZERO_ERROR;
84 copy.setText(iter, errorCode);
85 if(U_FAILURE(errorCode)) {
86 errln("error Normalizer::setText() failed: %s", u_errorName(errorCode));
87 } else {
88 copy.getText(out);
89 iter.getText(out2);
90 if( out!=out2 ||
91 copy.startIndex()!=iter.startIndex() ||
92 copy.endIndex()!=iter.endIndex()
93 ) {
94 errln("error in Normalizer::setText() or Normalizer::getText()");
95 }
96 }
97
98 // test setText(UChar *), getUMode() and setMode()
99 errorCode=U_ZERO_ERROR;
100 copy.setText(s.getBuffer()+1, s.length()-1, errorCode);
101 copy.setMode(UNORM_NFD);
102 if(copy.getUMode()!=UNORM_NFD) {
103 errln("error in Normalizer::setMode() or Normalizer::getUMode()");
104 }
105 if(copy.next()!=0x308 || copy.next()!=0x1100) {
106 dataerrln("error in Normalizer::setText(UChar *) or Normalizer::setMode()");
107 }
108
109 // test setText(UChar *, length=-1)
110 errorCode=U_ZERO_ERROR;
111
112 // NUL-terminate s
113 s.append((UChar)0); // append NUL
114 s.truncate(s.length()-1); // undo length change
115
116 copy.setText(s.getBuffer()+1, -1, errorCode);
117 if(copy.endIndex()!=s.length()-1) {
118 errln("error in Normalizer::setText(UChar *, -1)");
119 }
120
121 // test setOption() and getOption()
122 copy.setOption(0xaa0000, true);
123 copy.setOption(0x20000, false);
124 if(!copy.getOption(0x880000) || copy.getOption(0x20000)) {
125 errln("error in Normalizer::setOption() or Normalizer::getOption()");
126 }
127
128 // test last()/previous() with an internal buffer overflow
129 errorCode=U_ZERO_ERROR;
130 copy.setText(UnicodeString(1000, (UChar32)0x308, 1000), errorCode);
131 if(copy.last()!=0x308) {
132 errln("error in Normalizer(1000*U+0308).last()");
133 }
134
135 // test UNORM_NONE
136 norm.setMode(UNORM_NONE);
137 if(norm.first()!=0x61 || norm.next()!=0x308 || norm.last()!=0x2f800) {
138 errln("error in Normalizer(UNORM_NONE).first()/next()/last()");
139 }
140 Normalizer::normalize(s, UNORM_NONE, 0, out, status);
141 if(out!=s) {
142 errln("error in Normalizer::normalize(UNORM_NONE)");
143 }
144
145 // test that the same string can be used as source and destination
146 s.setTo((UChar)0xe4);
147 Normalizer::normalize(s, UNORM_NFD, 0, s, status);
148 if(s.charAt(1)!=0x308) {
149 dataerrln("error in Normalizer::normalize(UNORM_NFD, self)");
150 }
151 Normalizer::normalize(s, UNORM_NFC, 0, s, status);
152 if(s.charAt(0)!=0xe4) {
153 dataerrln("error in Normalizer::normalize(UNORM_NFC, self)");
154 }
155 Normalizer::decompose(s, false, 0, s, status);
156 if(s.charAt(1)!=0x308) {
157 dataerrln("error in Normalizer::decompose(self)");
158 }
159 Normalizer::compose(s, false, 0, s, status);
160 if(s.charAt(0)!=0xe4) {
161 dataerrln("error in Normalizer::compose(self)");
162 }
163 Normalizer::concatenate(s, s, s, UNORM_NFC, 0, status);
164 if(s.charAt(1)!=0xe4) {
165 dataerrln("error in Normalizer::decompose(self)");
166 }
167 }
168
169 #endif
170