]> git.saurik.com Git - apple/icu.git/blob - icuSources/test/intltest/ustrtest.cpp
ICU-461.18.tar.gz
[apple/icu.git] / icuSources / test / intltest / ustrtest.cpp
1 /********************************************************************
2 * COPYRIGHT:
3 * Copyright (c) 1997-2010, International Business Machines Corporation and
4 * others. All Rights Reserved.
5 ********************************************************************/
6
7 #include "ustrtest.h"
8 #include "unicode/std_string.h"
9 #include "unicode/unistr.h"
10 #include "unicode/uchar.h"
11 #include "unicode/ustring.h"
12 #include "unicode/locid.h"
13 #include "unicode/ucnv.h"
14 #include "unicode/uenum.h"
15 #include "cmemory.h"
16 #include "charstr.h"
17
18 #if 0
19 #include "unicode/ustream.h"
20
21 #if U_IOSTREAM_SOURCE >= 199711
22 #include <iostream>
23 using namespace std;
24 #elif U_IOSTREAM_SOURCE >= 198506
25 #include <iostream.h>
26 #endif
27
28 #endif
29
30 #define LENGTHOF(array) (int32_t)((sizeof(array)/sizeof((array)[0])))
31
32 UnicodeStringTest::~UnicodeStringTest() {}
33
34 void UnicodeStringTest::runIndexedTest( int32_t index, UBool exec, const char* &name, char *par)
35 {
36 if (exec) logln("TestSuite UnicodeStringTest: ");
37 switch (index) {
38 case 0:
39 name = "StringCaseTest";
40 if (exec) {
41 logln("StringCaseTest---"); logln("");
42 StringCaseTest test;
43 callTest(test, par);
44 }
45 break;
46 case 1: name = "TestBasicManipulation"; if (exec) TestBasicManipulation(); break;
47 case 2: name = "TestCompare"; if (exec) TestCompare(); break;
48 case 3: name = "TestExtract"; if (exec) TestExtract(); break;
49 case 4: name = "TestRemoveReplace"; if (exec) TestRemoveReplace(); break;
50 case 5: name = "TestSearching"; if (exec) TestSearching(); break;
51 case 6: name = "TestSpacePadding"; if (exec) TestSpacePadding(); break;
52 case 7: name = "TestPrefixAndSuffix"; if (exec) TestPrefixAndSuffix(); break;
53 case 8: name = "TestFindAndReplace"; if (exec) TestFindAndReplace(); break;
54 case 9: name = "TestBogus"; if (exec) TestBogus(); break;
55 case 10: name = "TestReverse"; if (exec) TestReverse(); break;
56 case 11: name = "TestMiscellaneous"; if (exec) TestMiscellaneous(); break;
57 case 12: name = "TestStackAllocation"; if (exec) TestStackAllocation(); break;
58 case 13: name = "TestUnescape"; if (exec) TestUnescape(); break;
59 case 14: name = "TestCountChar32"; if (exec) TestCountChar32(); break;
60 case 15: name = "TestStringEnumeration"; if (exec) TestStringEnumeration(); break;
61 case 16: name = "TestNameSpace"; if (exec) TestNameSpace(); break;
62 case 17: name = "TestUTF32"; if (exec) TestUTF32(); break;
63 case 18: name = "TestUTF8"; if (exec) TestUTF8(); break;
64 case 19: name = "TestReadOnlyAlias"; if (exec) TestReadOnlyAlias(); break;
65
66 default: name = ""; break; //needed to end loop
67 }
68 }
69
70 void
71 UnicodeStringTest::TestBasicManipulation()
72 {
73 UnicodeString test1("Now is the time for all men to come swiftly to the aid of the party.\n");
74 UnicodeString expectedValue;
75 UnicodeString *c;
76
77 c=(UnicodeString *)test1.clone();
78 test1.insert(24, "good ");
79 expectedValue = "Now is the time for all good men to come swiftly to the aid of the party.\n";
80 if (test1 != expectedValue)
81 errln("insert() failed: expected \"" + expectedValue + "\"\n,got \"" + test1 + "\"");
82
83 c->insert(24, "good ");
84 if(*c != expectedValue) {
85 errln("clone()->insert() failed: expected \"" + expectedValue + "\"\n,got \"" + *c + "\"");
86 }
87 delete c;
88
89 test1.remove(41, 8);
90 expectedValue = "Now is the time for all good men to come to the aid of the party.\n";
91 if (test1 != expectedValue)
92 errln("remove() failed: expected \"" + expectedValue + "\"\n,got \"" + test1 + "\"");
93
94 test1.replace(58, 6, "ir country");
95 expectedValue = "Now is the time for all good men to come to the aid of their country.\n";
96 if (test1 != expectedValue)
97 errln("replace() failed: expected \"" + expectedValue + "\"\n,got \"" + test1 + "\"");
98
99 UChar temp[80];
100 test1.extract(0, 15, temp);
101
102 UnicodeString test2(temp, 15);
103
104 expectedValue = "Now is the time";
105 if (test2 != expectedValue)
106 errln("extract() failed: expected \"" + expectedValue + "\"\n,got \"" + test2 + "\"");
107
108 test2 += " for me to go!\n";
109 expectedValue = "Now is the time for me to go!\n";
110 if (test2 != expectedValue)
111 errln("operator+=() failed: expected \"" + expectedValue + "\"\n,got \"" + test2 + "\"");
112
113 if (test1.length() != 70)
114 errln("length() failed: expected 70, got " + test1.length());
115 if (test2.length() != 30)
116 errln("length() failed: expected 30, got " + test2.length());
117
118 UnicodeString test3;
119 test3.append((UChar32)0x20402);
120 if(test3 != CharsToUnicodeString("\\uD841\\uDC02")){
121 errln((UnicodeString)"append failed for UChar32, expected \"\\\\ud841\\\\udc02\", got " + prettify(test3));
122 }
123 if(test3.length() != 2){
124 errln("append or length failed for UChar32, expected 2, got " + test3.length());
125 }
126 test3.append((UChar32)0x0074);
127 if(test3 != CharsToUnicodeString("\\uD841\\uDC02t")){
128 errln((UnicodeString)"append failed for UChar32, expected \"\\\\uD841\\\\uDC02t\", got " + prettify(test3));
129 }
130 if(test3.length() != 3){
131 errln((UnicodeString)"append or length failed for UChar32, expected 2, got " + test3.length());
132 }
133
134 // test some UChar32 overloads
135 if( test3.setTo((UChar32)0x10330).length() != 2 ||
136 test3.insert(0, (UChar32)0x20100).length() != 4 ||
137 test3.replace(2, 2, (UChar32)0xe0061).length() != 4 ||
138 (test3 = (UChar32)0x14001).length() != 2
139 ) {
140 errln((UnicodeString)"simple UChar32 overloads for replace, insert, setTo or = failed");
141 }
142
143 {
144 // test moveIndex32()
145 UnicodeString s=UNICODE_STRING("\\U0002f999\\U0001d15f\\u00c4\\u1ed0", 32).unescape();
146
147 if(
148 s.moveIndex32(2, -1)!=0 ||
149 s.moveIndex32(2, 1)!=4 ||
150 s.moveIndex32(2, 2)!=5 ||
151 s.moveIndex32(5, -2)!=2 ||
152 s.moveIndex32(0, -1)!=0 ||
153 s.moveIndex32(6, 1)!=6
154 ) {
155 errln("UnicodeString::moveIndex32() failed");
156 }
157
158 if(s.getChar32Start(1)!=0 || s.getChar32Start(2)!=2) {
159 errln("UnicodeString::getChar32Start() failed");
160 }
161
162 if(s.getChar32Limit(1)!=2 || s.getChar32Limit(2)!=2) {
163 errln("UnicodeString::getChar32Limit() failed");
164 }
165 }
166
167 {
168 // test new 2.2 constructors and setTo function that parallel Java's substring function.
169 UnicodeString src("Hello folks how are you?");
170 UnicodeString target1("how are you?");
171 if (target1 != UnicodeString(src, 12)) {
172 errln("UnicodeString(const UnicodeString&, int32_t) failed");
173 }
174 UnicodeString target2("folks");
175 if (target2 != UnicodeString(src, 6, 5)) {
176 errln("UnicodeString(const UnicodeString&, int32_t, int32_t) failed");
177 }
178 if (target1 != target2.setTo(src, 12)) {
179 errln("UnicodeString::setTo(const UnicodeString&, int32_t) failed");
180 }
181 }
182
183 {
184 // op+ is new in ICU 2.8
185 UnicodeString s=UnicodeString("abc", "")+UnicodeString("def", "")+UnicodeString("ghi", "");
186 if(s!=UnicodeString("abcdefghi", "")) {
187 errln("operator+(UniStr, UniStr) failed");
188 }
189 }
190
191 {
192 // tests for Jitterbug 2360
193 // verify that APIs with source pointer + length accept length == -1
194 // mostly test only where modified, only few functions did not already do this
195 if(UnicodeString("abc", -1, "")!=UnicodeString("abc", "")) {
196 errln("UnicodeString(codepageData, dataLength, codepage) does not work with dataLength==-1");
197 }
198
199 UChar buffer[10]={ 0x61, 0x62, 0x20ac, 0xd900, 0xdc05, 0, 0x62, 0xffff, 0xdbff, 0xdfff };
200 UnicodeString s, t(buffer, -1, LENGTHOF(buffer));
201
202 if(s.setTo(buffer, -1, LENGTHOF(buffer)).length()!=u_strlen(buffer)) {
203 errln("UnicodeString.setTo(buffer, length, capacity) does not work with length==-1");
204 }
205 if(t.length()!=u_strlen(buffer)) {
206 errln("UnicodeString(buffer, length, capacity) does not work with length==-1");
207 }
208
209 if(0!=s.caseCompare(buffer, -1, U_FOLD_CASE_DEFAULT)) {
210 errln("UnicodeString.caseCompare(const UChar *, length, options) does not work with length==-1");
211 }
212 if(0!=s.caseCompare(0, s.length(), buffer, U_FOLD_CASE_DEFAULT)) {
213 errln("UnicodeString.caseCompare(start, _length, const UChar *, options) does not work");
214 }
215
216 buffer[u_strlen(buffer)]=0xe4;
217 UnicodeString u(buffer, -1, LENGTHOF(buffer));
218 if(s.setTo(buffer, -1, LENGTHOF(buffer)).length()!=LENGTHOF(buffer)) {
219 errln("UnicodeString.setTo(buffer without NUL, length, capacity) does not work with length==-1");
220 }
221 if(u.length()!=LENGTHOF(buffer)) {
222 errln("UnicodeString(buffer without NUL, length, capacity) does not work with length==-1");
223 }
224
225 static const char cs[]={ 0x61, (char)0xe4, (char)0x85, 0 };
226 UConverter *cnv;
227 UErrorCode errorCode=U_ZERO_ERROR;
228
229 cnv=ucnv_open("ISO-8859-1", &errorCode);
230 UnicodeString v(cs, -1, cnv, errorCode);
231 ucnv_close(cnv);
232 if(v!=CharsToUnicodeString("a\\xe4\\x85")) {
233 errln("UnicodeString(const char *, length, cnv, errorCode) does not work with length==-1");
234 }
235 }
236
237 #if U_CHARSET_IS_UTF8
238 {
239 // Test the hardcoded-UTF-8 UnicodeString optimizations.
240 static const uint8_t utf8[]={ 0x61, 0xC3, 0xA4, 0xC3, 0x9F, 0xE4, 0xB8, 0x80, 0 };
241 static const UChar utf16[]={ 0x61, 0xE4, 0xDF, 0x4E00 };
242 UnicodeString from8a = UnicodeString((const char *)utf8);
243 UnicodeString from8b = UnicodeString((const char *)utf8, (int32_t)sizeof(utf8)-1);
244 UnicodeString from16(FALSE, utf16, LENGTHOF(utf16));
245 if(from8a != from16 || from8b != from16) {
246 errln("UnicodeString(const char * U_CHARSET_IS_UTF8) failed");
247 }
248 char buffer[16];
249 int32_t length8=from16.extract(0, 0x7fffffff, buffer, (uint32_t)sizeof(buffer));
250 if(length8!=((int32_t)sizeof(utf8)-1) || 0!=uprv_memcmp(buffer, utf8, sizeof(utf8))) {
251 errln("UnicodeString::extract(char * U_CHARSET_IS_UTF8) failed");
252 }
253 length8=from16.extract(1, 2, buffer, (uint32_t)sizeof(buffer));
254 if(length8!=4 || buffer[length8]!=0 || 0!=uprv_memcmp(buffer, utf8+1, length8)) {
255 errln("UnicodeString::extract(substring to char * U_CHARSET_IS_UTF8) failed");
256 }
257 }
258 #endif
259 }
260
261 void
262 UnicodeStringTest::TestCompare()
263 {
264 UnicodeString test1("this is a test");
265 UnicodeString test2("this is a test");
266 UnicodeString test3("this is a test of the emergency broadcast system");
267 UnicodeString test4("never say, \"this is a test\"!!");
268
269 UnicodeString test5((UChar)0x5000);
270 UnicodeString test6((UChar)0x5100);
271
272 UChar uniChars[] = { 0x74, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73,
273 0x20, 0x61, 0x20, 0x74, 0x65, 0x73, 0x74, 0 };
274 char chars[] = "this is a test";
275
276 // test operator== and operator!=
277 if (test1 != test2 || test1 == test3 || test1 == test4)
278 errln("operator== or operator!= failed");
279
280 // test operator> and operator<
281 if (test1 > test2 || test1 < test2 || !(test1 < test3) || !(test1 > test4) ||
282 !(test5 < test6)
283 ) {
284 errln("operator> or operator< failed");
285 }
286
287 // test operator>= and operator<=
288 if (!(test1 >= test2) || !(test1 <= test2) || !(test1 <= test3) || !(test1 >= test4))
289 errln("operator>= or operator<= failed");
290
291 // test compare(UnicodeString)
292 if (test1.compare(test2) != 0 || test1.compare(test3) >= 0 || test1.compare(test4) <= 0)
293 errln("compare(UnicodeString) failed");
294
295 //test compare(offset, length, UnicodeString)
296 if(test1.compare(0, 14, test2) != 0 ||
297 test3.compare(0, 14, test2) != 0 ||
298 test4.compare(12, 14, test2) != 0 ||
299 test3.compare(0, 18, test1) <=0 )
300 errln("compare(offset, length, UnicodeString) failes");
301
302 // test compare(UChar*)
303 if (test2.compare(uniChars) != 0 || test3.compare(uniChars) <= 0 || test4.compare(uniChars) >= 0)
304 errln("compare(UChar*) failed");
305
306 // test compare(char*)
307 if (test2.compare(chars) != 0 || test3.compare(chars) <= 0 || test4.compare(chars) >= 0)
308 errln("compare(char*) failed");
309
310 // test compare(UChar*, length)
311 if (test1.compare(uniChars, 4) <= 0 || test1.compare(uniChars, 4) <= 0)
312 errln("compare(UChar*, length) failed");
313
314 // test compare(thisOffset, thisLength, that, thatOffset, thatLength)
315 if (test1.compare(0, 14, test2, 0, 14) != 0
316 || test1.compare(0, 14, test3, 0, 14) != 0
317 || test1.compare(0, 14, test4, 12, 14) != 0)
318 errln("1. compare(thisOffset, thisLength, that, thatOffset, thatLength) failed");
319
320 if (test1.compare(10, 4, test2, 0, 4) >= 0
321 || test1.compare(10, 4, test3, 22, 9) <= 0
322 || test1.compare(10, 4, test4, 22, 4) != 0)
323 errln("2. compare(thisOffset, thisLength, that, thatOffset, thatLength) failed");
324
325 // test compareBetween
326 if (test1.compareBetween(0, 14, test2, 0, 14) != 0 || test1.compareBetween(0, 14, test3, 0, 14) != 0
327 || test1.compareBetween(0, 14, test4, 12, 26) != 0)
328 errln("compareBetween failed");
329
330 if (test1.compareBetween(10, 14, test2, 0, 4) >= 0 || test1.compareBetween(10, 14, test3, 22, 31) <= 0
331 || test1.compareBetween(10, 14, test4, 22, 26) != 0)
332 errln("compareBetween failed");
333
334 // test compare() etc. with strings that share a buffer but are not equal
335 test2=test1; // share the buffer, length() too large for the stackBuffer
336 test2.truncate(1); // change only the length, not the buffer
337 if( test1==test2 || test1<=test2 ||
338 test1.compare(test2)<=0 ||
339 test1.compareCodePointOrder(test2)<=0 ||
340 test1.compareCodePointOrder(0, INT32_MAX, test2)<=0 ||
341 test1.compareCodePointOrder(0, INT32_MAX, test2, 0, INT32_MAX)<=0 ||
342 test1.compareCodePointOrderBetween(0, INT32_MAX, test2, 0, INT32_MAX)<=0 ||
343 test1.caseCompare(test2, U_FOLD_CASE_DEFAULT)<=0
344 ) {
345 errln("UnicodeStrings that share a buffer but have different lengths compare as equal");
346 }
347
348 /* test compareCodePointOrder() */
349 {
350 /* these strings are in ascending order */
351 static const UChar strings[][4]={
352 { 0x61, 0 }, /* U+0061 */
353 { 0x20ac, 0xd801, 0 }, /* U+20ac U+d801 */
354 { 0x20ac, 0xd800, 0xdc00, 0 }, /* U+20ac U+10000 */
355 { 0xd800, 0 }, /* U+d800 */
356 { 0xd800, 0xff61, 0 }, /* U+d800 U+ff61 */
357 { 0xdfff, 0 }, /* U+dfff */
358 { 0xff61, 0xdfff, 0 }, /* U+ff61 U+dfff */
359 { 0xff61, 0xd800, 0xdc02, 0 }, /* U+ff61 U+10002 */
360 { 0xd800, 0xdc02, 0 }, /* U+10002 */
361 { 0xd84d, 0xdc56, 0 } /* U+23456 */
362 };
363 UnicodeString u[20]; // must be at least as long as strings[]
364 int32_t i;
365
366 for(i=0; i<(int32_t)(sizeof(strings)/sizeof(strings[0])); ++i) {
367 u[i]=UnicodeString(TRUE, strings[i], -1);
368 }
369
370 for(i=0; i<(int32_t)(sizeof(strings)/sizeof(strings[0])-1); ++i) {
371 if(u[i].compareCodePointOrder(u[i+1])>=0 || u[i].compareCodePointOrder(0, INT32_MAX, u[i+1].getBuffer())>=0) {
372 errln("error: UnicodeString::compareCodePointOrder() fails for string %d and the following one\n", i);
373 }
374 }
375 }
376
377 /* test caseCompare() */
378 {
379 static const UChar
380 _mixed[]= { 0x61, 0x42, 0x131, 0x3a3, 0xdf, 0x130, 0x49, 0xfb03, 0xd93f, 0xdfff, 0 },
381 _otherDefault[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x69, 0x307, 0x69, 0x46, 0x66, 0x49, 0xd93f, 0xdfff, 0 },
382 _otherExcludeSpecialI[]={ 0x41, 0x62, 0x131, 0x3c3, 0x53, 0x73, 0x69, 0x131, 0x66, 0x46, 0x69, 0xd93f, 0xdfff, 0 },
383 _different[]= { 0x41, 0x62, 0x131, 0x3c3, 0x73, 0x53, 0x130, 0x49, 0x46, 0x66, 0x49, 0xd93f, 0xdffd, 0 };
384
385 UnicodeString
386 mixed(TRUE, _mixed, -1),
387 otherDefault(TRUE, _otherDefault, -1),
388 otherExcludeSpecialI(TRUE, _otherExcludeSpecialI, -1),
389 different(TRUE, _different, -1);
390
391 int8_t result;
392
393 /* test caseCompare() */
394 result=mixed.caseCompare(otherDefault, U_FOLD_CASE_DEFAULT);
395 if(result!=0 || 0!=mixed.caseCompareBetween(0, INT32_MAX, otherDefault, 0, INT32_MAX, U_FOLD_CASE_DEFAULT)) {
396 errln("error: mixed.caseCompare(other, default)=%ld instead of 0\n", result);
397 }
398 result=mixed.caseCompare(otherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I);
399 if(result!=0) {
400 errln("error: mixed.caseCompare(otherExcludeSpecialI, U_FOLD_CASE_EXCLUDE_SPECIAL_I)=%ld instead of 0\n", result);
401 }
402 result=mixed.caseCompare(otherDefault, U_FOLD_CASE_EXCLUDE_SPECIAL_I);
403 if(result==0 || 0==mixed.caseCompareBetween(0, INT32_MAX, otherDefault, 0, INT32_MAX, U_FOLD_CASE_EXCLUDE_SPECIAL_I)) {
404 errln("error: mixed.caseCompare(other, U_FOLD_CASE_EXCLUDE_SPECIAL_I)=0 instead of !=0\n");
405 }
406
407 /* test caseCompare() */
408 result=mixed.caseCompare(different, U_FOLD_CASE_DEFAULT);
409 if(result<=0) {
410 errln("error: mixed.caseCompare(different, default)=%ld instead of positive\n", result);
411 }
412
413 /* test caseCompare() - include the folded sharp s (U+00df) with different lengths */
414 result=mixed.caseCompare(1, 4, different, 1, 5, U_FOLD_CASE_DEFAULT);
415 if(result!=0 || 0!=mixed.caseCompareBetween(1, 5, different, 1, 6, U_FOLD_CASE_DEFAULT)) {
416 errln("error: mixed.caseCompare(mixed, 1, 4, different, 1, 5, default)=%ld instead of 0\n", result);
417 }
418
419 /* test caseCompare() - stop in the middle of the sharp s (U+00df) */
420 result=mixed.caseCompare(1, 4, different, 1, 4, U_FOLD_CASE_DEFAULT);
421 if(result<=0) {
422 errln("error: mixed.caseCompare(1, 4, different, 1, 4, default)=%ld instead of positive\n", result);
423 }
424 }
425
426 // test that srcLength=-1 is handled in functions that
427 // take input const UChar */int32_t srcLength (j785)
428 {
429 static const UChar u[]={ 0x61, 0x308, 0x62, 0 };
430 UnicodeString s=UNICODE_STRING("a\\u0308b", 8).unescape();
431
432 if(s.compare(u, -1)!=0 || s.compare(0, 999, u, 0, -1)!=0) {
433 errln("error UnicodeString::compare(..., const UChar *, srcLength=-1) does not work");
434 }
435
436 if(s.compareCodePointOrder(u, -1)!=0 || s.compareCodePointOrder(0, 999, u, 0, -1)!=0) {
437 errln("error UnicodeString::compareCodePointOrder(..., const UChar *, srcLength=-1, ...) does not work");
438 }
439
440 if(s.caseCompare(u, -1, U_FOLD_CASE_DEFAULT)!=0 || s.caseCompare(0, 999, u, 0, -1, U_FOLD_CASE_DEFAULT)!=0) {
441 errln("error UnicodeString::caseCompare(..., const UChar *, srcLength=-1, ...) does not work");
442 }
443
444 if(s.indexOf(u, 1, -1, 0, 999)!=1 || s.indexOf(u+1, -1, 0, 999)!=1 || s.indexOf(u+1, -1, 0)!=1) {
445 errln("error UnicodeString::indexOf(const UChar *, srcLength=-1, ...) does not work");
446 }
447
448 if(s.lastIndexOf(u, 1, -1, 0, 999)!=1 || s.lastIndexOf(u+1, -1, 0, 999)!=1 || s.lastIndexOf(u+1, -1, 0)!=1) {
449 errln("error UnicodeString::lastIndexOf(const UChar *, srcLength=-1, ...) does not work");
450 }
451
452 UnicodeString s2, s3;
453 s2.replace(0, 0, u+1, -1);
454 s3.replace(0, 0, u, 1, -1);
455 if(s.compare(1, 999, s2)!=0 || s2!=s3) {
456 errln("error UnicodeString::replace(..., const UChar *, srcLength=-1, ...) does not work");
457 }
458 }
459 }
460
461 void
462 UnicodeStringTest::TestExtract()
463 {
464 UnicodeString test1("Now is the time for all good men to come to the aid of their country.", "");
465 UnicodeString test2;
466 UChar test3[13] = {1, 2, 3, 4, 5, 6, 7, 8, 8, 10, 11, 12, 13};
467 char test4[13] = {1, 2, 3, 4, 5, 6, 7, 8, 8, 10, 11, 12, 13};
468 UnicodeString test5;
469 char test6[13] = {1, 2, 3, 4, 5, 6, 7, 8, 8, 10, 11, 12, 13};
470
471 test1.extract(11, 12, test2);
472 test1.extract(11, 12, test3);
473 if (test1.extract(11, 12, test4) != 12 || test4[12] != 0) {
474 errln("UnicodeString.extract(char *) failed to return the correct size of destination buffer.");
475 }
476
477 // test proper pinning in extractBetween()
478 test1.extractBetween(-3, 7, test5);
479 if(test5!=UNICODE_STRING("Now is ", 7)) {
480 errln("UnicodeString.extractBetween(-3, 7) did not pin properly.");
481 }
482
483 test1.extractBetween(11, 23, test5);
484 if (test1.extract(60, 71, test6) != 9) {
485 errln("UnicodeString.extract() failed to return the correct size of destination buffer for end of buffer.");
486 }
487 if (test1.extract(11, 12, test6) != 12) {
488 errln("UnicodeString.extract() failed to return the correct size of destination buffer.");
489 }
490
491 // convert test4 back to Unicode for comparison
492 UnicodeString test4b(test4, 12);
493
494 if (test1.extract(11, 12, (char *)NULL) != 12) {
495 errln("UnicodeString.extract(NULL) failed to return the correct size of destination buffer.");
496 }
497 if (test1.extract(11, -1, test6) != 0) {
498 errln("UnicodeString.extract(-1) failed to stop reading the string.");
499 }
500
501 for (int32_t i = 0; i < 12; i++) {
502 if (test1.charAt((int32_t)(11 + i)) != test2.charAt(i)) {
503 errln(UnicodeString("extracting into a UnicodeString failed at position ") + i);
504 break;
505 }
506 if (test1.charAt((int32_t)(11 + i)) != test3[i]) {
507 errln(UnicodeString("extracting into an array of UChar failed at position ") + i);
508 break;
509 }
510 if (((char)test1.charAt((int32_t)(11 + i))) != test4b.charAt(i)) {
511 errln(UnicodeString("extracting into an array of char failed at position ") + i);
512 break;
513 }
514 if (test1.charAt((int32_t)(11 + i)) != test5.charAt(i)) {
515 errln(UnicodeString("extracting with extractBetween failed at position ") + i);
516 break;
517 }
518 }
519
520 // test preflighting and overflows with invariant conversion
521 if (test1.extract(0, 10, (char *)NULL, "") != 10) {
522 errln("UnicodeString.extract(0, 10, (char *)NULL, \"\") != 10");
523 }
524
525 test4[2] = (char)0xff;
526 if (test1.extract(0, 10, test4, 2, "") != 10) {
527 errln("UnicodeString.extract(0, 10, test4, 2, \"\") != 10");
528 }
529 if (test4[2] != (char)0xff) {
530 errln("UnicodeString.extract(0, 10, test4, 2, \"\") overwrote test4[2]");
531 }
532
533 {
534 // test new, NUL-terminating extract() function
535 UnicodeString s("terminate", "");
536 UChar dest[20]={
537 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5,
538 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5, 0xa5
539 };
540 UErrorCode errorCode;
541 int32_t length;
542
543 errorCode=U_ZERO_ERROR;
544 length=s.extract((UChar *)NULL, 0, errorCode);
545 if(errorCode!=U_BUFFER_OVERFLOW_ERROR || length!=s.length()) {
546 errln("UnicodeString.extract(NULL, 0)==%d (%s) expected %d (U_BUFFER_OVERFLOW_ERROR)", length, s.length(), u_errorName(errorCode));
547 }
548
549 errorCode=U_ZERO_ERROR;
550 length=s.extract(dest, s.length()-1, errorCode);
551 if(errorCode!=U_BUFFER_OVERFLOW_ERROR || length!=s.length()) {
552 errln("UnicodeString.extract(dest too short)==%d (%s) expected %d (U_BUFFER_OVERFLOW_ERROR)",
553 length, u_errorName(errorCode), s.length());
554 }
555
556 errorCode=U_ZERO_ERROR;
557 length=s.extract(dest, s.length(), errorCode);
558 if(errorCode!=U_STRING_NOT_TERMINATED_WARNING || length!=s.length()) {
559 errln("UnicodeString.extract(dest just right without NUL)==%d (%s) expected %d (U_STRING_NOT_TERMINATED_WARNING)",
560 length, u_errorName(errorCode), s.length());
561 }
562 if(dest[length-1]!=s[length-1] || dest[length]!=0xa5) {
563 errln("UnicodeString.extract(dest just right without NUL) did not extract the string correctly");
564 }
565
566 errorCode=U_ZERO_ERROR;
567 length=s.extract(dest, s.length()+1, errorCode);
568 if(errorCode!=U_ZERO_ERROR || length!=s.length()) {
569 errln("UnicodeString.extract(dest large enough)==%d (%s) expected %d (U_ZERO_ERROR)",
570 length, u_errorName(errorCode), s.length());
571 }
572 if(dest[length-1]!=s[length-1] || dest[length]!=0 || dest[length+1]!=0xa5) {
573 errln("UnicodeString.extract(dest large enough) did not extract the string correctly");
574 }
575 }
576
577 {
578 // test new UConverter extract() and constructor
579 UnicodeString s=UNICODE_STRING("\\U0002f999\\U0001d15f\\u00c4\\u1ed0", 32).unescape();
580 char buffer[32];
581 static const char expect[]={
582 (char)0xf0, (char)0xaf, (char)0xa6, (char)0x99,
583 (char)0xf0, (char)0x9d, (char)0x85, (char)0x9f,
584 (char)0xc3, (char)0x84,
585 (char)0xe1, (char)0xbb, (char)0x90
586 };
587 UErrorCode errorCode=U_ZERO_ERROR;
588 UConverter *cnv=ucnv_open("UTF-8", &errorCode);
589 int32_t length;
590
591 if(U_SUCCESS(errorCode)) {
592 // test preflighting
593 if( (length=s.extract(NULL, 0, cnv, errorCode))!=13 ||
594 errorCode!=U_BUFFER_OVERFLOW_ERROR
595 ) {
596 errln("UnicodeString::extract(NULL, UConverter) preflighting failed (length=%ld, %s)",
597 length, u_errorName(errorCode));
598 }
599 errorCode=U_ZERO_ERROR;
600 if( (length=s.extract(buffer, 2, cnv, errorCode))!=13 ||
601 errorCode!=U_BUFFER_OVERFLOW_ERROR
602 ) {
603 errln("UnicodeString::extract(too small, UConverter) preflighting failed (length=%ld, %s)",
604 length, u_errorName(errorCode));
605 }
606
607 // try error cases
608 errorCode=U_ZERO_ERROR;
609 if( s.extract(NULL, 2, cnv, errorCode)==13 || U_SUCCESS(errorCode)) {
610 errln("UnicodeString::extract(UConverter) succeeded with an illegal destination");
611 }
612 errorCode=U_ILLEGAL_ARGUMENT_ERROR;
613 if( s.extract(NULL, 0, cnv, errorCode)==13 || U_SUCCESS(errorCode)) {
614 errln("UnicodeString::extract(UConverter) succeeded with a previous error code");
615 }
616 errorCode=U_ZERO_ERROR;
617
618 // extract for real
619 if( (length=s.extract(buffer, sizeof(buffer), cnv, errorCode))!=13 ||
620 uprv_memcmp(buffer, expect, 13)!=0 ||
621 buffer[13]!=0 ||
622 U_FAILURE(errorCode)
623 ) {
624 errln("UnicodeString::extract(UConverter) conversion failed (length=%ld, %s)",
625 length, u_errorName(errorCode));
626 }
627 // Test again with just the converter name.
628 if( (length=s.extract(0, s.length(), buffer, sizeof(buffer), "UTF-8"))!=13 ||
629 uprv_memcmp(buffer, expect, 13)!=0 ||
630 buffer[13]!=0 ||
631 U_FAILURE(errorCode)
632 ) {
633 errln("UnicodeString::extract(\"UTF-8\") conversion failed (length=%ld, %s)",
634 length, u_errorName(errorCode));
635 }
636
637 // try the constructor
638 UnicodeString t(expect, sizeof(expect), cnv, errorCode);
639 if(U_FAILURE(errorCode) || s!=t) {
640 errln("UnicodeString(UConverter) conversion failed (%s)",
641 u_errorName(errorCode));
642 }
643
644 ucnv_close(cnv);
645 }
646 }
647 }
648
649 void
650 UnicodeStringTest::TestRemoveReplace()
651 {
652 UnicodeString test1("The rain in Spain stays mainly on the plain");
653 UnicodeString test2("eat SPAMburgers!");
654 UChar test3[] = { 0x53, 0x50, 0x41, 0x4d, 0x4d, 0 };
655 char test4[] = "SPAM";
656 UnicodeString& test5 = test1;
657
658 test1.replace(4, 4, test2, 4, 4);
659 test1.replace(12, 5, test3, 4);
660 test3[4] = 0;
661 test1.replace(17, 4, test3);
662 test1.replace(23, 4, test4);
663 test1.replaceBetween(37, 42, test2, 4, 8);
664
665 if (test1 != "The SPAM in SPAM SPAMs SPAMly on the SPAM")
666 errln("One of the replace methods failed:\n"
667 " expected \"The SPAM in SPAM SPAMs SPAMly on the SPAM\",\n"
668 " got \"" + test1 + "\"");
669
670 test1.remove(21, 1);
671 test1.removeBetween(26, 28);
672
673 if (test1 != "The SPAM in SPAM SPAM SPAM on the SPAM")
674 errln("One of the remove methods failed:\n"
675 " expected \"The SPAM in SPAM SPAM SPAM on the SPAM\",\n"
676 " got \"" + test1 + "\"");
677
678 for (int32_t i = 0; i < test1.length(); i++) {
679 if (test5[i] != 0x53 && test5[i] != 0x50 && test5[i] != 0x41 && test5[i] != 0x4d && test5[i] != 0x20) {
680 test1.setCharAt(i, 0x78);
681 }
682 }
683
684 if (test1 != "xxx SPAM xx SPAM SPAM SPAM xx xxx SPAM")
685 errln("One of the remove methods failed:\n"
686 " expected \"xxx SPAM xx SPAM SPAM SPAM xx xxx SPAM\",\n"
687 " got \"" + test1 + "\"");
688
689 test1.remove();
690 if (test1.length() != 0)
691 errln("Remove() failed: expected empty string, got \"" + test1 + "\"");
692 }
693
694 void
695 UnicodeStringTest::TestSearching()
696 {
697 UnicodeString test1("test test ttest tetest testesteststt");
698 UnicodeString test2("test");
699 UChar testChar = 0x74;
700
701 UChar32 testChar32 = 0x20402;
702 UChar testData[]={
703 // 0 1 2 3 4 5 6 7
704 0xd841, 0xdc02, 0x0071, 0xdc02, 0xd841, 0x0071, 0xd841, 0xdc02,
705
706 // 8 9 10 11 12 13 14 15
707 0x0071, 0x0072, 0xd841, 0xdc02, 0x0071, 0xd841, 0xdc02, 0x0071,
708
709 // 16 17 18 19
710 0xdc02, 0xd841, 0x0073, 0x0000
711 };
712 UnicodeString test3(testData);
713 UnicodeString test4(testChar32);
714
715 uint16_t occurrences = 0;
716 int32_t startPos = 0;
717 for ( ;
718 startPos != -1 && startPos < test1.length();
719 (startPos = test1.indexOf(test2, startPos)) != -1 ? (++occurrences, startPos += 4) : 0)
720 ;
721 if (occurrences != 6)
722 errln("indexOf failed: expected to find 6 occurrences, found " + occurrences);
723
724 for ( occurrences = 0, startPos = 10;
725 startPos != -1 && startPos < test1.length();
726 (startPos = test1.indexOf(test2, startPos)) != -1 ? (++occurrences, startPos += 4) : 0)
727 ;
728 if (occurrences != 4)
729 errln("indexOf with starting offset failed: expected to find 4 occurrences, found " + occurrences);
730
731 int32_t endPos = 28;
732 for ( occurrences = 0, startPos = 5;
733 startPos != -1 && startPos < test1.length();
734 (startPos = test1.indexOf(test2, startPos, endPos - startPos)) != -1 ? (++occurrences, startPos += 4) : 0)
735 ;
736 if (occurrences != 4)
737 errln("indexOf with starting and ending offsets failed: expected to find 4 occurrences, found " + occurrences);
738
739 //using UChar32 string
740 for ( startPos=0, occurrences=0;
741 startPos != -1 && startPos < test3.length();
742 (startPos = test3.indexOf(test4, startPos)) != -1 ? (++occurrences, startPos += 2) : 0)
743 ;
744 if (occurrences != 4)
745 errln((UnicodeString)"indexOf failed: expected to find 4 occurrences, found " + occurrences);
746
747 for ( startPos=10, occurrences=0;
748 startPos != -1 && startPos < test3.length();
749 (startPos = test3.indexOf(test4, startPos)) != -1 ? (++occurrences, startPos += 2) : 0)
750 ;
751 if (occurrences != 2)
752 errln("indexOf failed: expected to find 2 occurrences, found " + occurrences);
753 //---
754
755 for ( occurrences = 0, startPos = 0;
756 startPos != -1 && startPos < test1.length();
757 (startPos = test1.indexOf(testChar, startPos)) != -1 ? (++occurrences, startPos += 1) : 0)
758 ;
759 if (occurrences != 16)
760 errln("indexOf with character failed: expected to find 16 occurrences, found " + occurrences);
761
762 for ( occurrences = 0, startPos = 10;
763 startPos != -1 && startPos < test1.length();
764 (startPos = test1.indexOf(testChar, startPos)) != -1 ? (++occurrences, startPos += 1) : 0)
765 ;
766 if (occurrences != 12)
767 errln("indexOf with character & start offset failed: expected to find 12 occurrences, found " + occurrences);
768
769 for ( occurrences = 0, startPos = 5, endPos = 28;
770 startPos != -1 && startPos < test1.length();
771 (startPos = test1.indexOf(testChar, startPos, endPos - startPos)) != -1 ? (++occurrences, startPos += 1) : 0)
772 ;
773 if (occurrences != 10)
774 errln("indexOf with character & start & end offsets failed: expected to find 10 occurrences, found " + occurrences);
775
776 //testing for UChar32
777 UnicodeString subString;
778 for( occurrences =0, startPos=0; startPos < test3.length(); startPos +=1){
779 subString.append(test3, startPos, test3.length());
780 if(subString.indexOf(testChar32) != -1 ){
781 ++occurrences;
782 }
783 subString.remove();
784 }
785 if (occurrences != 14)
786 errln((UnicodeString)"indexOf failed: expected to find 14 occurrences, found " + occurrences);
787
788 for ( occurrences = 0, startPos = 0;
789 startPos != -1 && startPos < test3.length();
790 (startPos = test3.indexOf(testChar32, startPos)) != -1 ? (++occurrences, startPos += 1) : 0)
791 ;
792 if (occurrences != 4)
793 errln((UnicodeString)"indexOf failed: expected to find 4 occurrences, found " + occurrences);
794
795 endPos=test3.length();
796 for ( occurrences = 0, startPos = 5;
797 startPos != -1 && startPos < test3.length();
798 (startPos = test3.indexOf(testChar32, startPos, endPos - startPos)) != -1 ? (++occurrences, startPos += 1) : 0)
799 ;
800 if (occurrences != 3)
801 errln((UnicodeString)"indexOf with character & start & end offsets failed: expected to find 2 occurrences, found " + occurrences);
802 //---
803
804 if(test1.lastIndexOf(test2)!=29) {
805 errln("test1.lastIndexOf(test2)!=29");
806 }
807
808 if(test1.lastIndexOf(test2, 15)!=29 || test1.lastIndexOf(test2, 29)!=29 || test1.lastIndexOf(test2, 30)!=-1) {
809 errln("test1.lastIndexOf(test2, start) failed");
810 }
811
812 for ( occurrences = 0, startPos = 32;
813 startPos != -1;
814 (startPos = test1.lastIndexOf(test2, 5, startPos - 5)) != -1 ? ++occurrences : 0)
815 ;
816 if (occurrences != 4)
817 errln("lastIndexOf with starting and ending offsets failed: expected to find 4 occurrences, found " + occurrences);
818
819 for ( occurrences = 0, startPos = 32;
820 startPos != -1;
821 (startPos = test1.lastIndexOf(testChar, 5, startPos - 5)) != -1 ? ++occurrences : 0)
822 ;
823 if (occurrences != 11)
824 errln("lastIndexOf with character & start & end offsets failed: expected to find 11 occurrences, found " + occurrences);
825
826 //testing UChar32
827 startPos=test3.length();
828 for ( occurrences = 0;
829 startPos != -1;
830 (startPos = test3.lastIndexOf(testChar32, 5, startPos - 5)) != -1 ? ++occurrences : 0)
831 ;
832 if (occurrences != 3)
833 errln((UnicodeString)"lastIndexOf with character & start & end offsets failed: expected to find 3 occurrences, found " + occurrences);
834
835
836 for ( occurrences = 0, endPos = test3.length(); endPos > 0; endPos -= 1){
837 subString.remove();
838 subString.append(test3, 0, endPos);
839 if(subString.lastIndexOf(testChar32) != -1 ){
840 ++occurrences;
841 }
842 }
843 if (occurrences != 18)
844 errln((UnicodeString)"indexOf failed: expected to find 18 occurrences, found " + occurrences);
845 //---
846
847 // test that indexOf(UChar32) and lastIndexOf(UChar32)
848 // do not find surrogate code points when they are part of matched pairs
849 // (= part of supplementary code points)
850 // Jitterbug 1542
851 if(test3.indexOf((UChar32)0xd841) != 4 || test3.indexOf((UChar32)0xdc02) != 3) {
852 errln("error: UnicodeString::indexOf(UChar32 surrogate) finds a partial supplementary code point");
853 }
854 if( UnicodeString(test3, 0, 17).lastIndexOf((UChar)0xd841, 0) != 4 ||
855 UnicodeString(test3, 0, 17).lastIndexOf((UChar32)0xd841, 2) != 4 ||
856 test3.lastIndexOf((UChar32)0xd841, 0, 17) != 4 || test3.lastIndexOf((UChar32)0xdc02, 0, 17) != 16
857 ) {
858 errln("error: UnicodeString::lastIndexOf(UChar32 surrogate) finds a partial supplementary code point");
859 }
860 }
861
862 void
863 UnicodeStringTest::TestSpacePadding()
864 {
865 UnicodeString test1("hello");
866 UnicodeString test2(" there");
867 UnicodeString test3("Hi! How ya doin'? Beautiful day, isn't it?");
868 UnicodeString test4;
869 UBool returnVal;
870 UnicodeString expectedValue;
871
872 returnVal = test1.padLeading(15);
873 expectedValue = " hello";
874 if (returnVal == FALSE || test1 != expectedValue)
875 errln("padLeading() failed: expected \"" + expectedValue + "\", got \"" + test1 + "\".");
876
877 returnVal = test2.padTrailing(15);
878 expectedValue = " there ";
879 if (returnVal == FALSE || test2 != expectedValue)
880 errln("padTrailing() failed: expected \"" + expectedValue + "\", got \"" + test2 + "\".");
881
882 expectedValue = test3;
883 returnVal = test3.padTrailing(15);
884 if (returnVal == TRUE || test3 != expectedValue)
885 errln("padTrailing() failed: expected \"" + expectedValue + "\", got \"" + test3 + "\".");
886
887 expectedValue = "hello";
888 test4.setTo(test1).trim();
889
890 if (test4 != expectedValue || test1 == expectedValue || test4 != expectedValue)
891 errln("trim(UnicodeString&) failed");
892
893 test1.trim();
894 if (test1 != expectedValue)
895 errln("trim() failed: expected \"" + expectedValue + "\", got \"" + test1 + "\".");
896
897 test2.trim();
898 expectedValue = "there";
899 if (test2 != expectedValue)
900 errln("trim() failed: expected \"" + expectedValue + "\", got \"" + test2 + "\".");
901
902 test3.trim();
903 expectedValue = "Hi! How ya doin'? Beautiful day, isn't it?";
904 if (test3 != expectedValue)
905 errln("trim() failed: expected \"" + expectedValue + "\", got \"" + test3 + "\".");
906
907 returnVal = test1.truncate(15);
908 expectedValue = "hello";
909 if (returnVal == TRUE || test1 != expectedValue)
910 errln("truncate() failed: expected \"" + expectedValue + "\", got \"" + test1 + "\".");
911
912 returnVal = test2.truncate(15);
913 expectedValue = "there";
914 if (returnVal == TRUE || test2 != expectedValue)
915 errln("truncate() failed: expected \"" + expectedValue + "\", got \"" + test2 + "\".");
916
917 returnVal = test3.truncate(15);
918 expectedValue = "Hi! How ya doi";
919 if (returnVal == FALSE || test3 != expectedValue)
920 errln("truncate() failed: expected \"" + expectedValue + "\", got \"" + test3 + "\".");
921 }
922
923 void
924 UnicodeStringTest::TestPrefixAndSuffix()
925 {
926 UnicodeString test1("Now is the time for all good men to come to the aid of their country.");
927 UnicodeString test2("Now");
928 UnicodeString test3("country.");
929 UnicodeString test4("count");
930
931 if (!test1.startsWith(test2) || !test1.startsWith(test2, 0, test2.length())) {
932 errln("startsWith() failed: \"" + test2 + "\" should be a prefix of \"" + test1 + "\".");
933 }
934
935 if (test1.startsWith(test3) ||
936 test1.startsWith(test3.getBuffer(), test3.length()) ||
937 test1.startsWith(test3.getTerminatedBuffer(), 0, -1)
938 ) {
939 errln("startsWith() failed: \"" + test3 + "\" shouldn't be a prefix of \"" + test1 + "\".");
940 }
941
942 if (test1.endsWith(test2)) {
943 errln("endsWith() failed: \"" + test2 + "\" shouldn't be a suffix of \"" + test1 + "\".");
944 }
945
946 if (!test1.endsWith(test3)) {
947 errln("endsWith(test3) failed: \"" + test3 + "\" should be a suffix of \"" + test1 + "\".");
948 }
949 if (!test1.endsWith(test3, 0, INT32_MAX)) {
950 errln("endsWith(test3, 0, INT32_MAX) failed: \"" + test3 + "\" should be a suffix of \"" + test1 + "\".");
951 }
952
953 if(!test1.endsWith(test3.getBuffer(), test3.length())) {
954 errln("endsWith(test3.getBuffer(), test3.length()) failed: \"" + test3 + "\" should be a suffix of \"" + test1 + "\".");
955 }
956 if(!test1.endsWith(test3.getTerminatedBuffer(), 0, -1)) {
957 errln("endsWith(test3.getTerminatedBuffer(), 0, -1) failed: \"" + test3 + "\" should be a suffix of \"" + test1 + "\".");
958 }
959
960 if (!test3.startsWith(test4)) {
961 errln("endsWith(test4) failed: \"" + test4 + "\" should be a prefix of \"" + test3 + "\".");
962 }
963
964 if (test4.startsWith(test3)) {
965 errln("startsWith(test3) failed: \"" + test3 + "\" shouldn't be a prefix of \"" + test4 + "\".");
966 }
967 }
968
969 void
970 UnicodeStringTest::TestFindAndReplace()
971 {
972 UnicodeString test1("One potato, two potato, three potato, four\n");
973 UnicodeString test2("potato");
974 UnicodeString test3("MISSISSIPPI");
975
976 UnicodeString expectedValue;
977
978 test1.findAndReplace(test2, test3);
979 expectedValue = "One MISSISSIPPI, two MISSISSIPPI, three MISSISSIPPI, four\n";
980 if (test1 != expectedValue)
981 errln("findAndReplace failed: expected \"" + expectedValue + "\", got \"" + test1 + "\".");
982 test1.findAndReplace(2, 32, test3, test2);
983 expectedValue = "One potato, two potato, three MISSISSIPPI, four\n";
984 if (test1 != expectedValue)
985 errln("findAndReplace failed: expected \"" + expectedValue + "\", got \"" + test1 + "\".");
986 }
987
988 void
989 UnicodeStringTest::TestReverse()
990 {
991 UnicodeString test("backwards words say to used I");
992
993 test.reverse();
994 test.reverse(2, 4);
995 test.reverse(7, 2);
996 test.reverse(10, 3);
997 test.reverse(14, 5);
998 test.reverse(20, 9);
999
1000 if (test != "I used to say words backwards")
1001 errln("reverse() failed: Expected \"I used to say words backwards\",\n got \""
1002 + test + "\"");
1003
1004 test=UNICODE_STRING("\\U0002f999\\U0001d15f\\u00c4\\u1ed0", 32).unescape();
1005 test.reverse();
1006 if(test.char32At(0)!=0x1ed0 || test.char32At(1)!=0xc4 || test.char32At(2)!=0x1d15f || test.char32At(4)!=0x2f999) {
1007 errln("reverse() failed with supplementary characters");
1008 }
1009
1010 // Test case for ticket #8091:
1011 // UnicodeString::reverse() failed to see a lead surrogate in the middle of
1012 // an odd-length string that contains no other lead surrogates.
1013 test=UNICODE_STRING_SIMPLE("ab\\U0001F4A9e").unescape();
1014 UnicodeString expected=UNICODE_STRING_SIMPLE("e\\U0001F4A9ba").unescape();
1015 test.reverse();
1016 if(test!=expected) {
1017 errln("reverse() failed with only lead surrogate in the middle");
1018 }
1019 }
1020
1021 void
1022 UnicodeStringTest::TestMiscellaneous()
1023 {
1024 UnicodeString test1("This is a test");
1025 UnicodeString test2("This is a test");
1026 UnicodeString test3("Me too!");
1027
1028 // test getBuffer(minCapacity) and releaseBuffer()
1029 test1=UnicodeString(); // make sure that it starts with its stackBuffer
1030 UChar *p=test1.getBuffer(20);
1031 if(test1.getCapacity()<20) {
1032 errln("UnicodeString::getBuffer(20).getCapacity()<20");
1033 }
1034
1035 test1.append((UChar)7); // must not be able to modify the string here
1036 test1.setCharAt(3, 7);
1037 test1.reverse();
1038 if( test1.length()!=0 ||
1039 test1.charAt(0)!=0xffff || test1.charAt(3)!=0xffff ||
1040 test1.getBuffer(10)!=0 || test1.getBuffer()!=0
1041 ) {
1042 errln("UnicodeString::getBuffer(minCapacity) allows read or write access to the UnicodeString");
1043 }
1044
1045 p[0]=1;
1046 p[1]=2;
1047 p[2]=3;
1048 test1.releaseBuffer(3);
1049 test1.append((UChar)4);
1050
1051 if(test1.length()!=4 || test1.charAt(0)!=1 || test1.charAt(1)!=2 || test1.charAt(2)!=3 || test1.charAt(3)!=4) {
1052 errln("UnicodeString::releaseBuffer(newLength) does not properly reallow access to the UnicodeString");
1053 }
1054
1055 // test releaseBuffer() without getBuffer(minCapacity) - must not have any effect
1056 test1.releaseBuffer(1);
1057 if(test1.length()!=4 || test1.charAt(0)!=1 || test1.charAt(1)!=2 || test1.charAt(2)!=3 || test1.charAt(3)!=4) {
1058 errln("UnicodeString::releaseBuffer(newLength) without getBuffer(minCapacity) changed the UnicodeString");
1059 }
1060
1061 // test getBuffer(const)
1062 const UChar *q=test1.getBuffer(), *r=test1.getBuffer();
1063 if( test1.length()!=4 ||
1064 q[0]!=1 || q[1]!=2 || q[2]!=3 || q[3]!=4 ||
1065 r[0]!=1 || r[1]!=2 || r[2]!=3 || r[3]!=4
1066 ) {
1067 errln("UnicodeString::getBuffer(const) does not return a usable buffer pointer");
1068 }
1069
1070 // test releaseBuffer() with a NUL-terminated buffer
1071 test1.getBuffer(20)[2]=0;
1072 test1.releaseBuffer(); // implicit -1
1073 if(test1.length()!=2 || test1.charAt(0)!=1 || test1.charAt(1) !=2) {
1074 errln("UnicodeString::releaseBuffer(-1) does not properly set the length of the UnicodeString");
1075 }
1076
1077 // test releaseBuffer() with a non-NUL-terminated buffer
1078 p=test1.getBuffer(256);
1079 for(int32_t i=0; i<test1.getCapacity(); ++i) {
1080 p[i]=(UChar)1; // fill the buffer with all non-NUL code units
1081 }
1082 test1.releaseBuffer(); // implicit -1
1083 if(test1.length()!=test1.getCapacity() || test1.charAt(1)!=1 || test1.charAt(100)!=1 || test1.charAt(test1.getCapacity()-1)!=1) {
1084 errln("UnicodeString::releaseBuffer(-1 but no NUL) does not properly set the length of the UnicodeString");
1085 }
1086
1087 // test getTerminatedBuffer()
1088 test1=UnicodeString("This is another test.", "");
1089 test2=UnicodeString("This is another test.", "");
1090 q=test1.getTerminatedBuffer();
1091 if(q[test1.length()]!=0 || test1!=test2 || test2.compare(q, -1)!=0) {
1092 errln("getTerminatedBuffer()[length]!=0");
1093 }
1094
1095 const UChar u[]={ 5, 6, 7, 8, 0 };
1096 test1.setTo(FALSE, u, 3);
1097 q=test1.getTerminatedBuffer();
1098 if(q==u || q[0]!=5 || q[1]!=6 || q[2]!=7 || q[3]!=0) {
1099 errln("UnicodeString(u[3]).getTerminatedBuffer() returns a bad buffer");
1100 }
1101
1102 test1.setTo(TRUE, u, -1);
1103 q=test1.getTerminatedBuffer();
1104 if(q!=u || test1.length()!=4 || q[3]!=8 || q[4]!=0) {
1105 errln("UnicodeString(u[-1]).getTerminatedBuffer() returns a bad buffer");
1106 }
1107
1108 test1=UNICODE_STRING("la", 2);
1109 test1.append(UNICODE_STRING(" lila", 5).getTerminatedBuffer(), 0, -1);
1110 if(test1!=UNICODE_STRING("la lila", 7)) {
1111 errln("UnicodeString::append(const UChar *, start, length) failed");
1112 }
1113
1114 test1.insert(3, UNICODE_STRING("dudum ", 6), 0, INT32_MAX);
1115 if(test1!=UNICODE_STRING("la dudum lila", 13)) {
1116 errln("UnicodeString::insert(start, const UniStr &, start, length) failed");
1117 }
1118
1119 static const UChar ucs[]={ 0x68, 0x6d, 0x20, 0 };
1120 test1.insert(9, ucs, -1);
1121 if(test1!=UNICODE_STRING("la dudum hm lila", 16)) {
1122 errln("UnicodeString::insert(start, const UChar *, length) failed");
1123 }
1124
1125 test1.replace(9, 2, (UChar)0x2b);
1126 if(test1!=UNICODE_STRING("la dudum + lila", 15)) {
1127 errln("UnicodeString::replace(start, length, UChar) failed");
1128 }
1129
1130 if(test1.hasMetaData() || UnicodeString().hasMetaData()) {
1131 errln("UnicodeString::hasMetaData() returns TRUE");
1132 }
1133
1134 // test getTerminatedBuffer() on a truncated, shared, heap-allocated string
1135 test1=UNICODE_STRING_SIMPLE("abcdefghijklmnopqrstuvwxyz0123456789.");
1136 test1.truncate(36); // ensure length()<getCapacity()
1137 test2=test1; // share the buffer
1138 test1.truncate(5);
1139 if(test1.length()!=5 || test1.getTerminatedBuffer()[5]!=0) {
1140 errln("UnicodeString(shared buffer).truncate() failed");
1141 }
1142 if(test2.length()!=36 || test2[5]!=0x66 || u_strlen(test2.getTerminatedBuffer())!=36) {
1143 errln("UnicodeString(shared buffer).truncate().getTerminatedBuffer() "
1144 "modified another copy of the string!");
1145 }
1146 test1=UNICODE_STRING_SIMPLE("abcdefghijklmnopqrstuvwxyz0123456789.");
1147 test1.truncate(36); // ensure length()<getCapacity()
1148 test2=test1; // share the buffer
1149 test1.remove();
1150 if(test1.length()!=0 || test1.getTerminatedBuffer()[0]!=0) {
1151 errln("UnicodeString(shared buffer).remove() failed");
1152 }
1153 if(test2.length()!=36 || test2[0]!=0x61 || u_strlen(test2.getTerminatedBuffer())!=36) {
1154 errln("UnicodeString(shared buffer).remove().getTerminatedBuffer() "
1155 "modified another copy of the string!");
1156 }
1157 }
1158
1159 void
1160 UnicodeStringTest::TestStackAllocation()
1161 {
1162 UChar testString[] ={
1163 0x54, 0x68, 0x69, 0x73, 0x20, 0x69, 0x73, 0x20, 0x61, 0x20, 0x63, 0x72, 0x61, 0x7a, 0x79, 0x20, 0x74, 0x65, 0x73, 0x74, 0x2e, 0 };
1164 UChar guardWord = 0x4DED;
1165 UnicodeString* test = 0;
1166
1167 test = new UnicodeString(testString);
1168 if (*test != "This is a crazy test.")
1169 errln("Test string failed to initialize properly.");
1170 if (guardWord != 0x04DED)
1171 errln("Test string initialization overwrote guard word!");
1172
1173 test->insert(8, "only ");
1174 test->remove(15, 6);
1175 if (*test != "This is only a test.")
1176 errln("Manipulation of test string failed to work right.");
1177 if (guardWord != 0x4DED)
1178 errln("Manipulation of test string overwrote guard word!");
1179
1180 // we have to deinitialize and release the backing store by calling the destructor
1181 // explicitly, since we can't overload operator delete
1182 delete test;
1183
1184 UChar workingBuffer[] = {
1185 0x4e, 0x6f, 0x77, 0x20, 0x69, 0x73, 0x20, 0x74, 0x68, 0x65, 0x20, 0x74, 0x69, 0x6d, 0x65, 0x20,
1186 0x66, 0x6f, 0x72, 0x20, 0x61, 0x6c, 0x6c, 0x20, 0x6d, 0x65, 0x6e, 0x20, 0x74, 0x6f, 0x20,
1187 0x63, 0x6f, 0x6d, 0x65, 0xffff, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1188 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
1189 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
1190 UChar guardWord2 = 0x4DED;
1191
1192 test = new UnicodeString(workingBuffer, 35, 100);
1193 if (*test != "Now is the time for all men to come")
1194 errln("Stack-allocated backing store failed to initialize correctly.");
1195 if (guardWord2 != 0x4DED)
1196 errln("Stack-allocated backing store overwrote guard word!");
1197
1198 test->insert(24, "good ");
1199 if (*test != "Now is the time for all good men to come")
1200 errln("insert() on stack-allocated UnicodeString didn't work right");
1201 if (guardWord2 != 0x4DED)
1202 errln("insert() on stack-allocated UnicodeString overwrote guard word!");
1203
1204 if (workingBuffer[24] != 0x67)
1205 errln("insert() on stack-allocated UnicodeString didn't affect backing store");
1206
1207 *test += " to the aid of their country.";
1208 if (*test != "Now is the time for all good men to come to the aid of their country.")
1209 errln("Stack-allocated UnicodeString overflow didn't work");
1210 if (guardWord2 != 0x4DED)
1211 errln("Stack-allocated UnicodeString overflow overwrote guard word!");
1212
1213 *test = "ha!";
1214 if (*test != "ha!")
1215 errln("Assignment to stack-allocated UnicodeString didn't work");
1216 if (workingBuffer[0] != 0x4e)
1217 errln("Change to UnicodeString after overflow are still affecting original buffer");
1218 if (guardWord2 != 0x4DED)
1219 errln("Change to UnicodeString after overflow overwrote guard word!");
1220
1221 // test read-only aliasing with setTo()
1222 workingBuffer[0] = 0x20ac;
1223 workingBuffer[1] = 0x125;
1224 workingBuffer[2] = 0;
1225 test->setTo(TRUE, workingBuffer, 2);
1226 if(test->length() != 2 || test->charAt(0) != 0x20ac || test->charAt(1) != 0x125) {
1227 errln("UnicodeString.setTo(readonly alias) does not alias correctly");
1228 }
1229
1230 UnicodeString *c=(UnicodeString *)test->clone();
1231
1232 workingBuffer[1] = 0x109;
1233 if(test->charAt(1) != 0x109) {
1234 errln("UnicodeString.setTo(readonly alias) made a copy: did not see change in buffer");
1235 }
1236
1237 if(c->length() != 2 || c->charAt(1) != 0x125) {
1238 errln("clone(alias) did not copy the buffer");
1239 }
1240 delete c;
1241
1242 test->setTo(TRUE, workingBuffer, -1);
1243 if(test->length() != 2 || test->charAt(0) != 0x20ac || test->charAt(1) != 0x109) {
1244 errln("UnicodeString.setTo(readonly alias, length -1) does not alias correctly");
1245 }
1246
1247 test->setTo(FALSE, workingBuffer, -1);
1248 if(!test->isBogus()) {
1249 errln("UnicodeString.setTo(unterminated readonly alias, length -1) does not result in isBogus()");
1250 }
1251
1252 delete test;
1253
1254 test=new UnicodeString();
1255 UChar buffer[]={0x0061, 0x0062, 0x20ac, 0x0043, 0x0042, 0x0000};
1256 test->setTo(buffer, 4, 10);
1257 if(test->length() !=4 || test->charAt(0) != 0x0061 || test->charAt(1) != 0x0062 ||
1258 test->charAt(2) != 0x20ac || test->charAt(3) != 0x0043){
1259 errln((UnicodeString)"UnicodeString.setTo(UChar*, length, capacity) does not work correctly\n" + prettify(*test));
1260 }
1261 delete test;
1262
1263
1264 // test the UChar32 constructor
1265 UnicodeString c32Test((UChar32)0x10ff2a);
1266 if( c32Test.length() != UTF_CHAR_LENGTH(0x10ff2a) ||
1267 c32Test.char32At(c32Test.length() - 1) != 0x10ff2a
1268 ) {
1269 errln("The UnicodeString(UChar32) constructor does not work with a 0x10ff2a filler");
1270 }
1271
1272 // test the (new) capacity constructor
1273 UnicodeString capTest(5, (UChar32)0x2a, 5);
1274 if( capTest.length() != 5 * UTF_CHAR_LENGTH(0x2a) ||
1275 capTest.char32At(0) != 0x2a ||
1276 capTest.char32At(4) != 0x2a
1277 ) {
1278 errln("The UnicodeString capacity constructor does not work with an ASCII filler");
1279 }
1280
1281 capTest = UnicodeString(5, (UChar32)0x10ff2a, 5);
1282 if( capTest.length() != 5 * UTF_CHAR_LENGTH(0x10ff2a) ||
1283 capTest.char32At(0) != 0x10ff2a ||
1284 capTest.char32At(4) != 0x10ff2a
1285 ) {
1286 errln("The UnicodeString capacity constructor does not work with a 0x10ff2a filler");
1287 }
1288
1289 capTest = UnicodeString(5, (UChar32)0, 0);
1290 if(capTest.length() != 0) {
1291 errln("The UnicodeString capacity constructor does not work with a 0x10ff2a filler");
1292 }
1293 }
1294
1295 /**
1296 * Test the unescape() function.
1297 */
1298 void UnicodeStringTest::TestUnescape(void) {
1299 UnicodeString IN("abc\\u4567 \\n\\r \\U00101234xyz\\x1\\x{5289}\\x1b", -1, US_INV);
1300 UnicodeString OUT("abc");
1301 OUT.append((UChar)0x4567);
1302 OUT.append(" ");
1303 OUT.append((UChar)0xA);
1304 OUT.append((UChar)0xD);
1305 OUT.append(" ");
1306 OUT.append((UChar32)0x00101234);
1307 OUT.append("xyz");
1308 OUT.append((UChar32)1).append((UChar32)0x5289).append((UChar)0x1b);
1309 UnicodeString result = IN.unescape();
1310 if (result != OUT) {
1311 errln("FAIL: " + prettify(IN) + ".unescape() -> " +
1312 prettify(result) + ", expected " +
1313 prettify(OUT));
1314 }
1315
1316 // test that an empty string is returned in case of an error
1317 if (!UNICODE_STRING("wrong \\u sequence", 17).unescape().isEmpty()) {
1318 errln("FAIL: unescaping of a string with an illegal escape sequence did not return an empty string");
1319 }
1320 }
1321
1322 /* test code point counting functions --------------------------------------- */
1323
1324 /* reference implementation of UnicodeString::hasMoreChar32Than() */
1325 static int32_t
1326 _refUnicodeStringHasMoreChar32Than(const UnicodeString &s, int32_t start, int32_t length, int32_t number) {
1327 int32_t count=s.countChar32(start, length);
1328 return count>number;
1329 }
1330
1331 /* compare the real function against the reference */
1332 void
1333 UnicodeStringTest::_testUnicodeStringHasMoreChar32Than(const UnicodeString &s, int32_t start, int32_t length, int32_t number) {
1334 if(s.hasMoreChar32Than(start, length, number)!=_refUnicodeStringHasMoreChar32Than(s, start, length, number)) {
1335 errln("hasMoreChar32Than(%d, %d, %d)=%hd is wrong\n",
1336 start, length, number, s.hasMoreChar32Than(start, length, number));
1337 }
1338 }
1339
1340 void
1341 UnicodeStringTest::TestCountChar32(void) {
1342 {
1343 UnicodeString s=UNICODE_STRING("\\U0002f999\\U0001d15f\\u00c4\\u1ed0", 32).unescape();
1344
1345 // test countChar32()
1346 // note that this also calls and tests u_countChar32(length>=0)
1347 if(
1348 s.countChar32()!=4 ||
1349 s.countChar32(1)!=4 ||
1350 s.countChar32(2)!=3 ||
1351 s.countChar32(2, 3)!=2 ||
1352 s.countChar32(2, 0)!=0
1353 ) {
1354 errln("UnicodeString::countChar32() failed");
1355 }
1356
1357 // NUL-terminate the string buffer and test u_countChar32(length=-1)
1358 const UChar *buffer=s.getTerminatedBuffer();
1359 if(
1360 u_countChar32(buffer, -1)!=4 ||
1361 u_countChar32(buffer+1, -1)!=4 ||
1362 u_countChar32(buffer+2, -1)!=3 ||
1363 u_countChar32(buffer+3, -1)!=3 ||
1364 u_countChar32(buffer+4, -1)!=2 ||
1365 u_countChar32(buffer+5, -1)!=1 ||
1366 u_countChar32(buffer+6, -1)!=0
1367 ) {
1368 errln("u_countChar32(length=-1) failed");
1369 }
1370
1371 // test u_countChar32() with bad input
1372 if(u_countChar32(NULL, 5)!=0 || u_countChar32(buffer, -2)!=0) {
1373 errln("u_countChar32(bad input) failed (returned non-zero counts)");
1374 }
1375 }
1376
1377 /* test data and variables for hasMoreChar32Than() */
1378 static const UChar str[]={
1379 0x61, 0x62, 0xd800, 0xdc00,
1380 0xd801, 0xdc01, 0x63, 0xd802,
1381 0x64, 0xdc03, 0x65, 0x66,
1382 0xd804, 0xdc04, 0xd805, 0xdc05,
1383 0x67
1384 };
1385 UnicodeString string(str, LENGTHOF(str));
1386 int32_t start, length, number;
1387
1388 /* test hasMoreChar32Than() */
1389 for(length=string.length(); length>=0; --length) {
1390 for(start=0; start<=length; ++start) {
1391 for(number=-1; number<=((length-start)+2); ++number) {
1392 _testUnicodeStringHasMoreChar32Than(string, start, length-start, number);
1393 }
1394 }
1395 }
1396
1397 /* test hasMoreChar32Than() with pinning */
1398 for(start=-1; start<=string.length()+1; ++start) {
1399 for(number=-1; number<=((string.length()-start)+2); ++number) {
1400 _testUnicodeStringHasMoreChar32Than(string, start, 0x7fffffff, number);
1401 }
1402 }
1403
1404 /* test hasMoreChar32Than() with a bogus string */
1405 string.setToBogus();
1406 for(length=-1; length<=1; ++length) {
1407 for(start=-1; start<=length; ++start) {
1408 for(number=-1; number<=((length-start)+2); ++number) {
1409 _testUnicodeStringHasMoreChar32Than(string, start, length-start, number);
1410 }
1411 }
1412 }
1413 }
1414
1415 void
1416 UnicodeStringTest::TestBogus() {
1417 UnicodeString test1("This is a test");
1418 UnicodeString test2("This is a test");
1419 UnicodeString test3("Me too!");
1420
1421 // test isBogus() and setToBogus()
1422 if (test1.isBogus() || test2.isBogus() || test3.isBogus()) {
1423 errln("A string returned TRUE for isBogus()!");
1424 }
1425
1426 // NULL pointers are treated like empty strings
1427 // use other illegal arguments to make a bogus string
1428 test3.setTo(FALSE, test1.getBuffer(), -2);
1429 if(!test3.isBogus()) {
1430 errln("A bogus string returned FALSE for isBogus()!");
1431 }
1432 if (test1.hashCode() != test2.hashCode() || test1.hashCode() == test3.hashCode()) {
1433 errln("hashCode() failed");
1434 }
1435 if(test3.getBuffer()!=0 || test3.getBuffer(20)!=0 || test3.getTerminatedBuffer()!=0) {
1436 errln("bogus.getBuffer()!=0");
1437 }
1438 if (test1.indexOf(test3) != -1) {
1439 errln("bogus.indexOf() != -1");
1440 }
1441 if (test1.lastIndexOf(test3) != -1) {
1442 errln("bogus.lastIndexOf() != -1");
1443 }
1444 if (test1.caseCompare(test3, U_FOLD_CASE_DEFAULT) != 1 || test3.caseCompare(test1, U_FOLD_CASE_DEFAULT) != -1) {
1445 errln("caseCompare() doesn't work with bogus strings");
1446 }
1447 if (test1.compareCodePointOrder(test3) != 1 || test3.compareCodePointOrder(test1) != -1) {
1448 errln("compareCodePointOrder() doesn't work with bogus strings");
1449 }
1450
1451 // verify that non-assignment modifications fail and do not revive a bogus string
1452 test3.setToBogus();
1453 test3.append((UChar)0x61);
1454 if(!test3.isBogus() || test3.getBuffer()!=0) {
1455 errln("bogus.append('a') worked but must not");
1456 }
1457
1458 test3.setToBogus();
1459 test3.findAndReplace(UnicodeString((UChar)0x61), test2);
1460 if(!test3.isBogus() || test3.getBuffer()!=0) {
1461 errln("bogus.findAndReplace() worked but must not");
1462 }
1463
1464 test3.setToBogus();
1465 test3.trim();
1466 if(!test3.isBogus() || test3.getBuffer()!=0) {
1467 errln("bogus.trim() revived bogus but must not");
1468 }
1469
1470 test3.setToBogus();
1471 test3.remove(1);
1472 if(!test3.isBogus() || test3.getBuffer()!=0) {
1473 errln("bogus.remove(1) revived bogus but must not");
1474 }
1475
1476 test3.setToBogus();
1477 if(!test3.setCharAt(0, 0x62).isBogus() || !test3.isEmpty()) {
1478 errln("bogus.setCharAt(0, 'b') worked but must not");
1479 }
1480
1481 test3.setToBogus();
1482 if(test3.truncate(1) || !test3.isBogus() || !test3.isEmpty()) {
1483 errln("bogus.truncate(1) revived bogus but must not");
1484 }
1485
1486 // verify that assignments revive a bogus string
1487 test3.setToBogus();
1488 if(!test3.isBogus() || (test3=test1).isBogus() || test3!=test1) {
1489 errln("bogus.operator=() failed");
1490 }
1491
1492 test3.setToBogus();
1493 if(!test3.isBogus() || test3.fastCopyFrom(test1).isBogus() || test3!=test1) {
1494 errln("bogus.fastCopyFrom() failed");
1495 }
1496
1497 test3.setToBogus();
1498 if(!test3.isBogus() || test3.setTo(test1).isBogus() || test3!=test1) {
1499 errln("bogus.setTo(UniStr) failed");
1500 }
1501
1502 test3.setToBogus();
1503 if(!test3.isBogus() || test3.setTo(test1, 0).isBogus() || test3!=test1) {
1504 errln("bogus.setTo(UniStr, 0) failed");
1505 }
1506
1507 test3.setToBogus();
1508 if(!test3.isBogus() || test3.setTo(test1, 0, 0x7fffffff).isBogus() || test3!=test1) {
1509 errln("bogus.setTo(UniStr, 0, len) failed");
1510 }
1511
1512 test3.setToBogus();
1513 if(!test3.isBogus() || test3.setTo(test1.getBuffer(), test1.length()).isBogus() || test3!=test1) {
1514 errln("bogus.setTo(const UChar *, len) failed");
1515 }
1516
1517 test3.setToBogus();
1518 if(!test3.isBogus() || test3.setTo((UChar)0x2028).isBogus() || test3!=UnicodeString((UChar)0x2028)) {
1519 errln("bogus.setTo(UChar) failed");
1520 }
1521
1522 test3.setToBogus();
1523 if(!test3.isBogus() || test3.setTo((UChar32)0x1d157).isBogus() || test3!=UnicodeString((UChar32)0x1d157)) {
1524 errln("bogus.setTo(UChar32) failed");
1525 }
1526
1527 test3.setToBogus();
1528 if(!test3.isBogus() || test3.setTo(FALSE, test1.getBuffer(), test1.length()).isBogus() || test3!=test1) {
1529 errln("bogus.setTo(readonly alias) failed");
1530 }
1531
1532 // writable alias to another string's buffer: very bad idea, just convenient for this test
1533 test3.setToBogus();
1534 if(!test3.isBogus() || test3.setTo((UChar *)test1.getBuffer(), test1.length(), test1.getCapacity()).isBogus() || test3!=test1) {
1535 errln("bogus.setTo(writable alias) failed");
1536 }
1537
1538 // verify simple, documented ways to turn a bogus string into an empty one
1539 test3.setToBogus();
1540 if(!test3.isBogus() || (test3=UnicodeString()).isBogus() || !test3.isEmpty()) {
1541 errln("bogus.operator=(UnicodeString()) failed");
1542 }
1543
1544 test3.setToBogus();
1545 if(!test3.isBogus() || test3.setTo(UnicodeString()).isBogus() || !test3.isEmpty()) {
1546 errln("bogus.setTo(UnicodeString()) failed");
1547 }
1548
1549 test3.setToBogus();
1550 if(test3.remove().isBogus() || test3.getBuffer()==0 || !test3.isEmpty()) {
1551 errln("bogus.remove() failed");
1552 }
1553
1554 test3.setToBogus();
1555 if(test3.remove(0, INT32_MAX).isBogus() || test3.getBuffer()==0 || !test3.isEmpty()) {
1556 errln("bogus.remove(0, INT32_MAX) failed");
1557 }
1558
1559 test3.setToBogus();
1560 if(test3.truncate(0) || test3.isBogus() || !test3.isEmpty()) {
1561 errln("bogus.truncate(0) failed");
1562 }
1563
1564 test3.setToBogus();
1565 if(!test3.isBogus() || test3.setTo((UChar32)-1).isBogus() || !test3.isEmpty()) {
1566 errln("bogus.setTo((UChar32)-1) failed");
1567 }
1568
1569 static const UChar nul=0;
1570
1571 test3.setToBogus();
1572 if(!test3.isBogus() || test3.setTo(&nul, 0).isBogus() || !test3.isEmpty()) {
1573 errln("bogus.setTo(&nul, 0) failed");
1574 }
1575
1576 test3.setToBogus();
1577 if(!test3.isBogus() || test3.getBuffer()!=0) {
1578 errln("setToBogus() failed to make a string bogus");
1579 }
1580
1581 test3.setToBogus();
1582 if(test1.isBogus() || !(test1=test3).isBogus()) {
1583 errln("normal=bogus failed to make the left string bogus");
1584 }
1585
1586 // test that NULL primitive input string values are treated like
1587 // empty strings, not errors (bogus)
1588 test2.setTo((UChar32)0x10005);
1589 if(test2.insert(1, NULL, 1).length()!=2) {
1590 errln("UniStr.insert(...NULL...) should not modify the string but does");
1591 }
1592
1593 UErrorCode errorCode=U_ZERO_ERROR;
1594 UnicodeString
1595 test4((const UChar *)NULL),
1596 test5(TRUE, (const UChar *)NULL, 1),
1597 test6((UChar *)NULL, 5, 5),
1598 test7((const char *)NULL, 3, NULL, errorCode);
1599 if(test4.isBogus() || test5.isBogus() || test6.isBogus() || test7.isBogus()) {
1600 errln("a constructor set to bogus for a NULL input string, should be empty");
1601 }
1602
1603 test4.setTo(NULL, 3);
1604 test5.setTo(TRUE, (const UChar *)NULL, 1);
1605 test6.setTo((UChar *)NULL, 5, 5);
1606 if(test4.isBogus() || test5.isBogus() || test6.isBogus()) {
1607 errln("a setTo() set to bogus for a NULL input string, should be empty");
1608 }
1609
1610 // test that bogus==bogus<any
1611 if(test1!=test3 || test1.compare(test3)!=0) {
1612 errln("bogus==bogus failed");
1613 }
1614
1615 test2.remove();
1616 if(test1>=test2 || !(test2>test1) || test1.compare(test2)>=0 || !(test2.compare(test1)>0)) {
1617 errln("bogus<empty failed");
1618 }
1619 }
1620
1621 // StringEnumeration ------------------------------------------------------- ***
1622 // most of StringEnumeration is tested elsewhere
1623 // this test improves code coverage
1624
1625 static const char *const
1626 testEnumStrings[]={
1627 "a",
1628 "b",
1629 "c",
1630 "this is a long string which helps us test some buffer limits",
1631 "eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee"
1632 };
1633
1634 class TestEnumeration : public StringEnumeration {
1635 public:
1636 TestEnumeration() : i(0) {}
1637
1638 virtual int32_t count(UErrorCode& /*status*/) const {
1639 return LENGTHOF(testEnumStrings);
1640 }
1641
1642 virtual const UnicodeString *snext(UErrorCode &status) {
1643 if(U_SUCCESS(status) && i<LENGTHOF(testEnumStrings)) {
1644 unistr=UnicodeString(testEnumStrings[i++], "");
1645 return &unistr;
1646 }
1647
1648 return NULL;
1649 }
1650
1651 virtual void reset(UErrorCode& /*status*/) {
1652 i=0;
1653 }
1654
1655 static inline UClassID getStaticClassID() {
1656 return (UClassID)&fgClassID;
1657 }
1658 virtual UClassID getDynamicClassID() const {
1659 return getStaticClassID();
1660 }
1661
1662 private:
1663 static const char fgClassID;
1664
1665 int32_t i, length;
1666 };
1667
1668 const char TestEnumeration::fgClassID=0;
1669
1670 void
1671 UnicodeStringTest::TestStringEnumeration() {
1672 UnicodeString s;
1673 TestEnumeration ten;
1674 int32_t i, length;
1675 UErrorCode status;
1676
1677 const UChar *pu;
1678 const char *pc;
1679
1680 // test the next() default implementation and ensureCharsCapacity()
1681 for(i=0; i<LENGTHOF(testEnumStrings); ++i) {
1682 status=U_ZERO_ERROR;
1683 pc=ten.next(&length, status);
1684 s=UnicodeString(testEnumStrings[i], "");
1685 if(U_FAILURE(status) || pc==NULL || length!=s.length() || UnicodeString(pc, length, "")!=s) {
1686 errln("StringEnumeration.next(%d) failed", i);
1687 }
1688 }
1689 status=U_ZERO_ERROR;
1690 if(ten.next(&length, status)!=NULL) {
1691 errln("StringEnumeration.next(done)!=NULL");
1692 }
1693
1694 // test the unext() default implementation
1695 ten.reset(status);
1696 for(i=0; i<LENGTHOF(testEnumStrings); ++i) {
1697 status=U_ZERO_ERROR;
1698 pu=ten.unext(&length, status);
1699 s=UnicodeString(testEnumStrings[i], "");
1700 if(U_FAILURE(status) || pu==NULL || length!=s.length() || UnicodeString(TRUE, pu, length)!=s) {
1701 errln("StringEnumeration.unext(%d) failed", i);
1702 }
1703 }
1704 status=U_ZERO_ERROR;
1705 if(ten.unext(&length, status)!=NULL) {
1706 errln("StringEnumeration.unext(done)!=NULL");
1707 }
1708
1709 // test that the default clone() implementation works, and returns NULL
1710 if(ten.clone()!=NULL) {
1711 errln("StringEnumeration.clone()!=NULL");
1712 }
1713
1714 // test that uenum_openFromStringEnumeration() works
1715 // Need a heap allocated string enumeration because it is adopted by the UEnumeration.
1716 StringEnumeration *newTen = new TestEnumeration;
1717 status=U_ZERO_ERROR;
1718 UEnumeration *uten = uenum_openFromStringEnumeration(newTen, &status);
1719 if (uten==NULL || U_FAILURE(status)) {
1720 errln("fail at file %s, line %d, UErrorCode is %s\n", __FILE__, __LINE__, u_errorName(status));
1721 return;
1722 }
1723
1724 // test uenum_next()
1725 for(i=0; i<LENGTHOF(testEnumStrings); ++i) {
1726 status=U_ZERO_ERROR;
1727 pc=uenum_next(uten, &length, &status);
1728 if(U_FAILURE(status) || pc==NULL || strcmp(pc, testEnumStrings[i]) != 0) {
1729 errln("File %s, line %d, StringEnumeration.next(%d) failed", __FILE__, __LINE__, i);
1730 }
1731 }
1732 status=U_ZERO_ERROR;
1733 if(uenum_next(uten, &length, &status)!=NULL) {
1734 errln("File %s, line %d, uenum_next(done)!=NULL");
1735 }
1736
1737 // test the uenum_unext()
1738 uenum_reset(uten, &status);
1739 for(i=0; i<LENGTHOF(testEnumStrings); ++i) {
1740 status=U_ZERO_ERROR;
1741 pu=uenum_unext(uten, &length, &status);
1742 s=UnicodeString(testEnumStrings[i], "");
1743 if(U_FAILURE(status) || pu==NULL || length!=s.length() || UnicodeString(TRUE, pu, length)!=s) {
1744 errln("File %s, Line %d, uenum_unext(%d) failed", __FILE__, __LINE__, i);
1745 }
1746 }
1747 status=U_ZERO_ERROR;
1748 if(uenum_unext(uten, &length, &status)!=NULL) {
1749 errln("File %s, Line %d, uenum_unext(done)!=NULL" __FILE__, __LINE__);
1750 }
1751
1752 uenum_close(uten);
1753 }
1754
1755 /*
1756 * Namespace test, to make sure that macros like UNICODE_STRING include the
1757 * namespace qualifier.
1758 *
1759 * Define a (bogus) UnicodeString class in another namespace and check for ambiguity.
1760 */
1761 #if U_HAVE_NAMESPACE
1762 namespace bogus {
1763 class UnicodeString {
1764 public:
1765 enum EInvariant { kInvariant };
1766 UnicodeString() : i(1) {}
1767 UnicodeString(UBool /*isTerminated*/, const UChar * /*text*/, int32_t textLength) : i(textLength) {}
1768 UnicodeString(const char * /*src*/, int32_t length, enum EInvariant /*inv*/
1769 ) : i(length) {}
1770 private:
1771 int32_t i;
1772 };
1773 }
1774 #endif
1775
1776 void
1777 UnicodeStringTest::TestNameSpace() {
1778 #if U_HAVE_NAMESPACE
1779 // Provoke name collision unless the UnicodeString macros properly
1780 // qualify the icu::UnicodeString class.
1781 using namespace bogus;
1782
1783 // Use all UnicodeString macros from unistr.h.
1784 icu::UnicodeString s1=icu::UnicodeString("abc", 3, US_INV);
1785 icu::UnicodeString s2=UNICODE_STRING("def", 3);
1786 icu::UnicodeString s3=UNICODE_STRING_SIMPLE("ghi");
1787
1788 // Make sure the compiler does not optimize away instantiation of s1, s2, s3.
1789 icu::UnicodeString s4=s1+s2+s3;
1790 if(s4.length()!=9) {
1791 errln("Something wrong with UnicodeString::operator+().");
1792 }
1793 #endif
1794 }
1795
1796 void
1797 UnicodeStringTest::TestUTF32() {
1798 // Input string length US_STACKBUF_SIZE to cause overflow of the
1799 // initially chosen fStackBuffer due to supplementary characters.
1800 static const UChar32 utf32[] = {
1801 0x41, 0xd900, 0x61, 0xdc00, -1, 0x110000, 0x5a, 0x50000, 0x7a,
1802 0x10000, 0x20000, 0xe0000, 0x10ffff
1803 };
1804 static const UChar expected_utf16[] = {
1805 0x41, 0xfffd, 0x61, 0xfffd, 0xfffd, 0xfffd, 0x5a, 0xd900, 0xdc00, 0x7a,
1806 0xd800, 0xdc00, 0xd840, 0xdc00, 0xdb40, 0xdc00, 0xdbff, 0xdfff
1807 };
1808 UnicodeString from32 = UnicodeString::fromUTF32(utf32, LENGTHOF(utf32));
1809 UnicodeString expected(FALSE, expected_utf16, LENGTHOF(expected_utf16));
1810 if(from32 != expected) {
1811 errln("UnicodeString::fromUTF32() did not create the expected string.");
1812 }
1813
1814 static const UChar utf16[] = {
1815 0x41, 0xd900, 0x61, 0xdc00, 0x5a, 0xd900, 0xdc00, 0x7a, 0xd800, 0xdc00, 0xdbff, 0xdfff
1816 };
1817 static const UChar32 expected_utf32[] = {
1818 0x41, 0xfffd, 0x61, 0xfffd, 0x5a, 0x50000, 0x7a, 0x10000, 0x10ffff
1819 };
1820 UChar32 result32[16];
1821 UErrorCode errorCode = U_ZERO_ERROR;
1822 int32_t length32 =
1823 UnicodeString(FALSE, utf16, LENGTHOF(utf16)).
1824 toUTF32(result32, LENGTHOF(result32), errorCode);
1825 if( length32 != LENGTHOF(expected_utf32) ||
1826 0 != uprv_memcmp(result32, expected_utf32, length32*4) ||
1827 result32[length32] != 0
1828 ) {
1829 errln("UnicodeString::toUTF32() did not create the expected string.");
1830 }
1831 }
1832
1833 class TestCheckedArrayByteSink : public CheckedArrayByteSink {
1834 public:
1835 TestCheckedArrayByteSink(char* outbuf, int32_t capacity)
1836 : CheckedArrayByteSink(outbuf, capacity), calledFlush(FALSE) {}
1837 virtual void Flush() { calledFlush = TRUE; }
1838 UBool calledFlush;
1839 };
1840
1841 void
1842 UnicodeStringTest::TestUTF8() {
1843 static const uint8_t utf8[] = {
1844 // Code points:
1845 // 0x41, 0xd900,
1846 // 0x61, 0xdc00,
1847 // 0x110000, 0x5a,
1848 // 0x50000, 0x7a,
1849 // 0x10000, 0x20000,
1850 // 0xe0000, 0x10ffff
1851 0x41, 0xed, 0xa4, 0x80,
1852 0x61, 0xed, 0xb0, 0x80,
1853 0xf4, 0x90, 0x80, 0x80, 0x5a,
1854 0xf1, 0x90, 0x80, 0x80, 0x7a,
1855 0xf0, 0x90, 0x80, 0x80, 0xf0, 0xa0, 0x80, 0x80,
1856 0xf3, 0xa0, 0x80, 0x80, 0xf4, 0x8f, 0xbf, 0xbf
1857 };
1858 static const UChar expected_utf16[] = {
1859 0x41, 0xfffd,
1860 0x61, 0xfffd,
1861 0xfffd, 0x5a,
1862 0xd900, 0xdc00, 0x7a,
1863 0xd800, 0xdc00, 0xd840, 0xdc00,
1864 0xdb40, 0xdc00, 0xdbff, 0xdfff
1865 };
1866 UnicodeString from8 = UnicodeString::fromUTF8(StringPiece((const char *)utf8, (int32_t)sizeof(utf8)));
1867 UnicodeString expected(FALSE, expected_utf16, LENGTHOF(expected_utf16));
1868
1869 if(from8 != expected) {
1870 errln("UnicodeString::fromUTF8(StringPiece) did not create the expected string.");
1871 }
1872 #if U_HAVE_STD_STRING
1873 U_STD_NSQ string utf8_string((const char *)utf8, sizeof(utf8));
1874 UnicodeString from8b = UnicodeString::fromUTF8(utf8_string);
1875 if(from8b != expected) {
1876 errln("UnicodeString::fromUTF8(std::string) did not create the expected string.");
1877 }
1878 #endif
1879
1880 static const UChar utf16[] = {
1881 0x41, 0xd900, 0x61, 0xdc00, 0x5a, 0xd900, 0xdc00, 0x7a, 0xd800, 0xdc00, 0xdbff, 0xdfff
1882 };
1883 static const uint8_t expected_utf8[] = {
1884 0x41, 0xef, 0xbf, 0xbd, 0x61, 0xef, 0xbf, 0xbd, 0x5a, 0xf1, 0x90, 0x80, 0x80, 0x7a,
1885 0xf0, 0x90, 0x80, 0x80, 0xf4, 0x8f, 0xbf, 0xbf
1886 };
1887 UnicodeString us(FALSE, utf16, LENGTHOF(utf16));
1888
1889 char buffer[64];
1890 TestCheckedArrayByteSink sink(buffer, (int32_t)sizeof(buffer));
1891 us.toUTF8(sink);
1892 if( sink.NumberOfBytesWritten() != (int32_t)sizeof(expected_utf8) ||
1893 0 != uprv_memcmp(buffer, expected_utf8, sizeof(expected_utf8))
1894 ) {
1895 errln("UnicodeString::toUTF8() did not create the expected string.");
1896 }
1897 if(!sink.calledFlush) {
1898 errln("UnicodeString::toUTF8(sink) did not sink.Flush().");
1899 }
1900 #if U_HAVE_STD_STRING
1901 // Initial contents for testing that toUTF8String() appends.
1902 U_STD_NSQ string result8 = "-->";
1903 U_STD_NSQ string expected8 = "-->" + U_STD_NSQ string((const char *)expected_utf8, sizeof(expected_utf8));
1904 // Use the return value just for testing.
1905 U_STD_NSQ string &result8r = us.toUTF8String(result8);
1906 if(result8r != expected8 || &result8r != &result8) {
1907 errln("UnicodeString::toUTF8String() did not create the expected string.");
1908 }
1909 #endif
1910 }
1911
1912 // Test if this compiler supports Return Value Optimization of unnamed temporary objects.
1913 static UnicodeString wrapUChars(const UChar *uchars) {
1914 return UnicodeString(TRUE, uchars, -1);
1915 }
1916
1917 void
1918 UnicodeStringTest::TestReadOnlyAlias() {
1919 UChar uchars[]={ 0x61, 0x62, 0 };
1920 UnicodeString alias(TRUE, uchars, 2);
1921 if(alias.length()!=2 || alias.getBuffer()!=uchars || alias.getTerminatedBuffer()!=uchars) {
1922 errln("UnicodeString read-only-aliasing constructor does not behave as expected.");
1923 return;
1924 }
1925 alias.truncate(1);
1926 if(alias.length()!=1 || alias.getBuffer()!=uchars) {
1927 errln("UnicodeString(read-only-alias).truncate() did not preserve aliasing as expected.");
1928 }
1929 if(alias.getTerminatedBuffer()==uchars) {
1930 errln("UnicodeString(read-only-alias).truncate().getTerminatedBuffer() "
1931 "did not allocate and copy as expected.");
1932 }
1933 if(uchars[1]!=0x62) {
1934 errln("UnicodeString(read-only-alias).truncate().getTerminatedBuffer() "
1935 "modified the original buffer.");
1936 }
1937 if(1!=u_strlen(alias.getTerminatedBuffer())) {
1938 errln("UnicodeString(read-only-alias).truncate().getTerminatedBuffer() "
1939 "does not return a buffer terminated at the proper length.");
1940 }
1941
1942 alias.setTo(TRUE, uchars, 2);
1943 if(alias.length()!=2 || alias.getBuffer()!=uchars || alias.getTerminatedBuffer()!=uchars) {
1944 errln("UnicodeString read-only-aliasing setTo() does not behave as expected.");
1945 return;
1946 }
1947 alias.remove();
1948 if(alias.length()!=0) {
1949 errln("UnicodeString(read-only-alias).remove() did not work.");
1950 }
1951 if(alias.getTerminatedBuffer()==uchars) {
1952 errln("UnicodeString(read-only-alias).remove().getTerminatedBuffer() "
1953 "did not un-alias as expected.");
1954 }
1955 if(uchars[0]!=0x61) {
1956 errln("UnicodeString(read-only-alias).remove().getTerminatedBuffer() "
1957 "modified the original buffer.");
1958 }
1959 if(0!=u_strlen(alias.getTerminatedBuffer())) {
1960 errln("UnicodeString.setTo(read-only-alias).remove().getTerminatedBuffer() "
1961 "does not return a buffer terminated at length 0.");
1962 }
1963
1964 UnicodeString longString=UNICODE_STRING_SIMPLE("abcdefghijklmnopqrstuvwxyz0123456789");
1965 alias.setTo(FALSE, longString.getBuffer(), longString.length());
1966 alias.remove(0, 10);
1967 if(longString.compare(10, INT32_MAX, alias)!=0 || alias.getBuffer()!=longString.getBuffer()+10) {
1968 errln("UnicodeString.setTo(read-only-alias).remove(0, 10) did not preserve aliasing as expected.");
1969 }
1970 alias.setTo(FALSE, longString.getBuffer(), longString.length());
1971 alias.remove(27, 99);
1972 if(longString.compare(0, 27, alias)!=0 || alias.getBuffer()!=longString.getBuffer()) {
1973 errln("UnicodeString.setTo(read-only-alias).remove(27, 99) did not preserve aliasing as expected.");
1974 }
1975 alias.setTo(FALSE, longString.getBuffer(), longString.length());
1976 alias.retainBetween(6, 30);
1977 if(longString.compare(6, 24, alias)!=0 || alias.getBuffer()!=longString.getBuffer()+6) {
1978 errln("UnicodeString.setTo(read-only-alias).retainBetween(6, 30) did not preserve aliasing as expected.");
1979 }
1980
1981 UChar abc[]={ 0x61, 0x62, 0x63, 0 };
1982 UBool hasRVO= wrapUChars(abc).getBuffer()==abc;
1983
1984 UnicodeString temp;
1985 temp.fastCopyFrom(longString.tempSubString());
1986 if(temp!=longString || (hasRVO && temp.getBuffer()!=longString.getBuffer())) {
1987 errln("UnicodeString.tempSubString() failed");
1988 }
1989 temp.fastCopyFrom(longString.tempSubString(-3, 5));
1990 if(longString.compare(0, 5, temp)!=0 || (hasRVO && temp.getBuffer()!=longString.getBuffer())) {
1991 errln("UnicodeString.tempSubString(-3, 5) failed");
1992 }
1993 temp.fastCopyFrom(longString.tempSubString(17));
1994 if(longString.compare(17, INT32_MAX, temp)!=0 || (hasRVO && temp.getBuffer()!=longString.getBuffer()+17)) {
1995 errln("UnicodeString.tempSubString(17) failed");
1996 }
1997 temp.fastCopyFrom(longString.tempSubString(99));
1998 if(!temp.isEmpty()) {
1999 errln("UnicodeString.tempSubString(99) failed");
2000 }
2001 temp.fastCopyFrom(longString.tempSubStringBetween(6));
2002 if(longString.compare(6, INT32_MAX, temp)!=0 || (hasRVO && temp.getBuffer()!=longString.getBuffer()+6)) {
2003 errln("UnicodeString.tempSubStringBetween(6) failed");
2004 }
2005 temp.fastCopyFrom(longString.tempSubStringBetween(8, 18));
2006 if(longString.compare(8, 10, temp)!=0 || (hasRVO && temp.getBuffer()!=longString.getBuffer()+8)) {
2007 errln("UnicodeString.tempSubStringBetween(8, 18) failed");
2008 }
2009 UnicodeString bogusString;
2010 bogusString.setToBogus();
2011 temp.fastCopyFrom(bogusString.tempSubStringBetween(8, 18));
2012 if(!temp.isBogus()) {
2013 errln("UnicodeString.setToBogus().tempSubStringBetween(8, 18) failed");
2014 }
2015 }