| OLD | NEW |
| (Empty) |
| 1 /* | |
| 2 ******************************************************************************* | |
| 3 * | |
| 4 * Copyright (C) 2002-2014, International Business Machines | |
| 5 * Corporation and others. All Rights Reserved. | |
| 6 * | |
| 7 ******************************************************************************* | |
| 8 * file name: strcase.cpp | |
| 9 * encoding: US-ASCII | |
| 10 * tab size: 8 (not used) | |
| 11 * indentation:4 | |
| 12 * | |
| 13 * created on: 2002mar12 | |
| 14 * created by: Markus W. Scherer | |
| 15 * | |
| 16 * Test file for string casing C++ API functions. | |
| 17 */ | |
| 18 | |
| 19 #include "unicode/uchar.h" | |
| 20 #include "unicode/ures.h" | |
| 21 #include "unicode/uloc.h" | |
| 22 #include "unicode/locid.h" | |
| 23 #include "unicode/ubrk.h" | |
| 24 #include "unicode/unistr.h" | |
| 25 #include "unicode/ucasemap.h" | |
| 26 #include "ucase.h" | |
| 27 #include "ustrtest.h" | |
| 28 #include "unicode/tstdtmod.h" | |
| 29 #include "cmemory.h" | |
| 30 | |
| 31 StringCaseTest::~StringCaseTest() {} | |
| 32 | |
| 33 void | |
| 34 StringCaseTest::runIndexedTest(int32_t index, UBool exec, const char *&name, cha
r * /*par*/) { | |
| 35 if(exec) { | |
| 36 logln("TestSuite StringCaseTest: "); | |
| 37 } | |
| 38 TESTCASE_AUTO_BEGIN; | |
| 39 TESTCASE_AUTO(TestCaseConversion); | |
| 40 #if !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILE_IO && !UCONFIG_NO_LEGACY_CON
VERSION | |
| 41 TESTCASE_AUTO(TestCasing); | |
| 42 #endif | |
| 43 TESTCASE_AUTO(TestFullCaseFoldingIterator); | |
| 44 TESTCASE_AUTO_END; | |
| 45 } | |
| 46 | |
| 47 void | |
| 48 StringCaseTest::TestCaseConversion() | |
| 49 { | |
| 50 static const UChar uppercaseGreek[] = | |
| 51 { 0x399, 0x395, 0x3a3, 0x3a5, 0x3a3, 0x20, 0x03a7, 0x3a1, 0x399, 0x3a3,
0x3a4, | |
| 52 0x39f, 0x3a3, 0 }; | |
| 53 // "IESUS CHRISTOS" | |
| 54 | |
| 55 static const UChar lowercaseGreek[] = | |
| 56 { 0x3b9, 0x3b5, 0x3c3, 0x3c5, 0x3c2, 0x20, 0x03c7, 0x3c1, 0x3b9, 0x3c3,
0x3c4, | |
| 57 0x3bf, 0x3c2, 0 }; | |
| 58 // "iesus christos" | |
| 59 | |
| 60 static const UChar lowercaseTurkish[] = | |
| 61 { 0x69, 0x73, 0x74, 0x61, 0x6e, 0x62, 0x75, 0x6c, 0x2c, 0x20, 0x6e, 0x6f
, 0x74, 0x20, 0x63, 0x6f, | |
| 62 0x6e, 0x73, 0x74, 0x61, 0x6e, 0x74, 0x0131, 0x6e, 0x6f, 0x70, 0x6c, 0x65
, 0x21, 0 }; | |
| 63 | |
| 64 static const UChar uppercaseTurkish[] = | |
| 65 { 0x54, 0x4f, 0x50, 0x4b, 0x41, 0x50, 0x49, 0x20, 0x50, 0x41, 0x4c, 0x41
, 0x43, 0x45, 0x2c, 0x20, | |
| 66 0x0130, 0x53, 0x54, 0x41, 0x4e, 0x42, 0x55, 0x4c, 0 }; | |
| 67 | |
| 68 UnicodeString expectedResult; | |
| 69 UnicodeString test3; | |
| 70 | |
| 71 test3 += (UChar32)0x0130; | |
| 72 test3 += "STANBUL, NOT CONSTANTINOPLE!"; | |
| 73 | |
| 74 UnicodeString test4(test3); | |
| 75 test4.toLower(Locale("")); | |
| 76 expectedResult = UnicodeString("i\\u0307stanbul, not constantinople!", "").u
nescape(); | |
| 77 if (test4 != expectedResult) | |
| 78 errln("1. toLower failed: expected \"" + expectedResult + "\", got \"" +
test4 + "\"."); | |
| 79 | |
| 80 test4 = test3; | |
| 81 test4.toLower(Locale("tr", "TR")); | |
| 82 expectedResult = lowercaseTurkish; | |
| 83 if (test4 != expectedResult) | |
| 84 errln("2. toLower failed: expected \"" + expectedResult + "\", got \"" +
test4 + "\"."); | |
| 85 | |
| 86 test3 = "topkap"; | |
| 87 test3 += (UChar32)0x0131; | |
| 88 test3 += " palace, istanbul"; | |
| 89 test4 = test3; | |
| 90 | |
| 91 test4.toUpper(Locale("")); | |
| 92 expectedResult = "TOPKAPI PALACE, ISTANBUL"; | |
| 93 if (test4 != expectedResult) | |
| 94 errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + te
st4 + "\"."); | |
| 95 | |
| 96 test4 = test3; | |
| 97 test4.toUpper(Locale("tr", "TR")); | |
| 98 expectedResult = uppercaseTurkish; | |
| 99 if (test4 != expectedResult) | |
| 100 errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + te
st4 + "\"."); | |
| 101 | |
| 102 test3 = CharsToUnicodeString("S\\u00FC\\u00DFmayrstra\\u00DFe"); | |
| 103 | |
| 104 test3.toUpper(Locale("de", "DE")); | |
| 105 expectedResult = CharsToUnicodeString("S\\u00DCSSMAYRSTRASSE"); | |
| 106 if (test3 != expectedResult) | |
| 107 errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + te
st3 + "\"."); | |
| 108 | |
| 109 test4.replace(0, test4.length(), uppercaseGreek); | |
| 110 | |
| 111 test4.toLower(Locale("el", "GR")); | |
| 112 expectedResult = lowercaseGreek; | |
| 113 if (test4 != expectedResult) | |
| 114 errln("toLower failed: expected \"" + expectedResult + "\", got \"" + te
st4 + "\"."); | |
| 115 | |
| 116 test4.replace(0, test4.length(), lowercaseGreek); | |
| 117 | |
| 118 test4.toUpper(); | |
| 119 expectedResult = uppercaseGreek; | |
| 120 if (test4 != expectedResult) | |
| 121 errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + te
st4 + "\"."); | |
| 122 | |
| 123 // more string case mapping tests with the new implementation | |
| 124 { | |
| 125 static const UChar | |
| 126 | |
| 127 beforeLower[]= { 0x61, 0x42, 0x49, 0x3a3, 0xdf, 0x3a3, 0x2f, 0xd93f, 0x
dfff }, | |
| 128 lowerRoot[]= { 0x61, 0x62, 0x69, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0x
dfff }, | |
| 129 lowerTurkish[]={ 0x61, 0x62, 0x131, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0x
dfff }, | |
| 130 | |
| 131 beforeUpper[]= { 0x61, 0x42, 0x69, 0x3c2, 0xdf, 0x3c3, 0x2f, 0xfb
03, 0xfb03, 0xfb03, 0xd93f, 0xdfff }, | |
| 132 upperRoot[]= { 0x41, 0x42, 0x49, 0x3a3, 0x53, 0x53, 0x3a3, 0x2f, 0x46
, 0x46, 0x49, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0xd93f, 0xdfff }, | |
| 133 upperTurkish[]={ 0x41, 0x42, 0x130, 0x3a3, 0x53, 0x53, 0x3a3, 0x2f, 0x46
, 0x46, 0x49, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0xd93f, 0xdfff }, | |
| 134 | |
| 135 beforeMiniUpper[]= { 0xdf, 0x61 }, | |
| 136 miniUpper[]= { 0x53, 0x53, 0x41 }; | |
| 137 | |
| 138 UnicodeString s; | |
| 139 | |
| 140 /* lowercase with root locale */ | |
| 141 s=UnicodeString(FALSE, beforeLower, (int32_t)(sizeof(beforeLower)/U_SIZE
OF_UCHAR)); | |
| 142 s.toLower(""); | |
| 143 if( s.length()!=(sizeof(lowerRoot)/U_SIZEOF_UCHAR) || | |
| 144 s!=UnicodeString(FALSE, lowerRoot, s.length()) | |
| 145 ) { | |
| 146 errln("error in toLower(root locale)=\"" + s + "\" expected \"" + Un
icodeString(FALSE, lowerRoot, (int32_t)(sizeof(lowerRoot)/U_SIZEOF_UCHAR)) + "\"
"); | |
| 147 } | |
| 148 | |
| 149 /* lowercase with turkish locale */ | |
| 150 s=UnicodeString(FALSE, beforeLower, (int32_t)(sizeof(beforeLower)/U_SIZE
OF_UCHAR)); | |
| 151 s.setCharAt(0, beforeLower[0]).toLower(Locale("tr")); | |
| 152 if( s.length()!=(sizeof(lowerTurkish)/U_SIZEOF_UCHAR) || | |
| 153 s!=UnicodeString(FALSE, lowerTurkish, s.length()) | |
| 154 ) { | |
| 155 errln("error in toLower(turkish locale)=\"" + s + "\" expected \"" +
UnicodeString(FALSE, lowerTurkish, (int32_t)(sizeof(lowerTurkish)/U_SIZEOF_UCHA
R)) + "\""); | |
| 156 } | |
| 157 | |
| 158 /* uppercase with root locale */ | |
| 159 s=UnicodeString(FALSE, beforeUpper, (int32_t)(sizeof(beforeUpper)/U_SIZE
OF_UCHAR)); | |
| 160 s.setCharAt(0, beforeUpper[0]).toUpper(Locale("")); | |
| 161 if( s.length()!=(sizeof(upperRoot)/U_SIZEOF_UCHAR) || | |
| 162 s!=UnicodeString(FALSE, upperRoot, s.length()) | |
| 163 ) { | |
| 164 errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + Un
icodeString(FALSE, upperRoot, (int32_t)(sizeof(upperRoot)/U_SIZEOF_UCHAR)) + "\"
"); | |
| 165 } | |
| 166 | |
| 167 /* uppercase with turkish locale */ | |
| 168 s=UnicodeString(FALSE, beforeUpper, (int32_t)(sizeof(beforeUpper)/U_SIZE
OF_UCHAR)); | |
| 169 s.toUpper(Locale("tr")); | |
| 170 if( s.length()!=(sizeof(upperTurkish)/U_SIZEOF_UCHAR) || | |
| 171 s!=UnicodeString(FALSE, upperTurkish, s.length()) | |
| 172 ) { | |
| 173 errln("error in toUpper(turkish locale)=\"" + s + "\" expected \"" +
UnicodeString(FALSE, upperTurkish, (int32_t)(sizeof(upperTurkish)/U_SIZEOF_UCHA
R)) + "\""); | |
| 174 } | |
| 175 | |
| 176 /* uppercase a short string with root locale */ | |
| 177 s=UnicodeString(FALSE, beforeMiniUpper, (int32_t)(sizeof(beforeMiniUpper
)/U_SIZEOF_UCHAR)); | |
| 178 s.setCharAt(0, beforeMiniUpper[0]).toUpper(""); | |
| 179 if( s.length()!=(sizeof(miniUpper)/U_SIZEOF_UCHAR) || | |
| 180 s!=UnicodeString(FALSE, miniUpper, s.length()) | |
| 181 ) { | |
| 182 errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + Un
icodeString(FALSE, miniUpper, (int32_t)(sizeof(miniUpper)/U_SIZEOF_UCHAR)) + "\"
"); | |
| 183 } | |
| 184 } | |
| 185 | |
| 186 // test some supplementary characters (>= Unicode 3.1) | |
| 187 { | |
| 188 UnicodeString t; | |
| 189 | |
| 190 UnicodeString | |
| 191 deseretInput=UnicodeString("\\U0001043C\\U00010414", "").unescape(), | |
| 192 deseretLower=UnicodeString("\\U0001043C\\U0001043C", "").unescape(), | |
| 193 deseretUpper=UnicodeString("\\U00010414\\U00010414", "").unescape(); | |
| 194 (t=deseretInput).toLower(); | |
| 195 if(t!=deseretLower) { | |
| 196 errln("error lowercasing Deseret (plane 1) characters"); | |
| 197 } | |
| 198 (t=deseretInput).toUpper(); | |
| 199 if(t!=deseretUpper) { | |
| 200 errln("error uppercasing Deseret (plane 1) characters"); | |
| 201 } | |
| 202 } | |
| 203 | |
| 204 // test some more cases that looked like problems | |
| 205 { | |
| 206 UnicodeString t; | |
| 207 | |
| 208 UnicodeString | |
| 209 ljInput=UnicodeString("ab'cD \\uFB00i\\u0131I\\u0130 \\u01C7\\u01C8\
\u01C9 \\U0001043C\\U00010414", "").unescape(), | |
| 210 ljLower=UnicodeString("ab'cd \\uFB00i\\u0131ii\\u0307 \\u01C9\\u01C9
\\u01C9 \\U0001043C\\U0001043C", "").unescape(), | |
| 211 ljUpper=UnicodeString("AB'CD FFIII\\u0130 \\u01C7\\u01C7\\u01C7 \\U0
0010414\\U00010414", "").unescape(); | |
| 212 (t=ljInput).toLower("en"); | |
| 213 if(t!=ljLower) { | |
| 214 errln("error lowercasing LJ characters"); | |
| 215 } | |
| 216 (t=ljInput).toUpper("en"); | |
| 217 if(t!=ljUpper) { | |
| 218 errln("error uppercasing LJ characters"); | |
| 219 } | |
| 220 } | |
| 221 | |
| 222 #if !UCONFIG_NO_NORMALIZATION | |
| 223 // some context-sensitive casing depends on normalization data being present | |
| 224 | |
| 225 // Unicode 3.1.1 SpecialCasing tests | |
| 226 { | |
| 227 UnicodeString t; | |
| 228 | |
| 229 // sigmas preceded and/or followed by cased letters | |
| 230 UnicodeString | |
| 231 sigmas=UnicodeString("i\\u0307\\u03a3\\u0308j \\u0307\\u03a3\\u0308j
i\\u00ad\\u03a3\\u0308 \\u0307\\u03a3\\u0308 ", "").unescape(), | |
| 232 sigmasLower=UnicodeString("i\\u0307\\u03c3\\u0308j \\u0307\\u03c3\\u
0308j i\\u00ad\\u03c2\\u0308 \\u0307\\u03c3\\u0308 ", "").unescape(), | |
| 233 sigmasUpper=UnicodeString("I\\u0307\\u03a3\\u0308J \\u0307\\u03a3\\u
0308J I\\u00ad\\u03a3\\u0308 \\u0307\\u03a3\\u0308 ", "").unescape(); | |
| 234 | |
| 235 (t=sigmas).toLower(); | |
| 236 if(t!=sigmasLower) { | |
| 237 errln("error in sigmas.toLower()=\"" + t + "\" expected \"" + sigmas
Lower + "\""); | |
| 238 } | |
| 239 | |
| 240 (t=sigmas).toUpper(Locale("")); | |
| 241 if(t!=sigmasUpper) { | |
| 242 errln("error in sigmas.toUpper()=\"" + t + "\" expected \"" + sigmas
Upper + "\""); | |
| 243 } | |
| 244 | |
| 245 // turkish & azerbaijani dotless i & dotted I | |
| 246 // remove dot above if there was a capital I before and there are no mor
e accents above | |
| 247 UnicodeString | |
| 248 dots=UnicodeString("I \\u0130 I\\u0307 I\\u0327\\u0307 I\\u0301\\u03
07 I\\u0327\\u0307\\u0301", "").unescape(), | |
| 249 dotsTurkish=UnicodeString("\\u0131 i i i\\u0327 \\u0131\\u0301\\u030
7 i\\u0327\\u0301", "").unescape(), | |
| 250 dotsDefault=UnicodeString("i i\\u0307 i\\u0307 i\\u0327\\u0307 i\\u0
301\\u0307 i\\u0327\\u0307\\u0301", "").unescape(); | |
| 251 | |
| 252 (t=dots).toLower("tr"); | |
| 253 if(t!=dotsTurkish) { | |
| 254 errln("error in dots.toLower(tr)=\"" + t + "\" expected \"" + dotsTu
rkish + "\""); | |
| 255 } | |
| 256 | |
| 257 (t=dots).toLower("de"); | |
| 258 if(t!=dotsDefault) { | |
| 259 errln("error in dots.toLower(de)=\"" + t + "\" expected \"" + dotsDe
fault + "\""); | |
| 260 } | |
| 261 } | |
| 262 | |
| 263 // more Unicode 3.1.1 tests | |
| 264 { | |
| 265 UnicodeString t; | |
| 266 | |
| 267 // lithuanian dot above in uppercasing | |
| 268 UnicodeString | |
| 269 dots=UnicodeString("a\\u0307 \\u0307 i\\u0307 j\\u0327\\u0307 j\\u03
01\\u0307", "").unescape(), | |
| 270 dotsLithuanian=UnicodeString("A\\u0307 \\u0307 I J\\u0327 J\\u0301\\
u0307", "").unescape(), | |
| 271 dotsDefault=UnicodeString("A\\u0307 \\u0307 I\\u0307 J\\u0327\\u0307
J\\u0301\\u0307", "").unescape(); | |
| 272 | |
| 273 (t=dots).toUpper("lt"); | |
| 274 if(t!=dotsLithuanian) { | |
| 275 errln("error in dots.toUpper(lt)=\"" + t + "\" expected \"" + dotsLi
thuanian + "\""); | |
| 276 } | |
| 277 | |
| 278 (t=dots).toUpper("de"); | |
| 279 if(t!=dotsDefault) { | |
| 280 errln("error in dots.toUpper(de)=\"" + t + "\" expected \"" + dotsDe
fault + "\""); | |
| 281 } | |
| 282 | |
| 283 // lithuanian adds dot above to i in lowercasing if there are more above
accents | |
| 284 UnicodeString | |
| 285 i=UnicodeString("I I\\u0301 J J\\u0301 \\u012e \\u012e\\u0301 \\u00c
c\\u00cd\\u0128", "").unescape(), | |
| 286 iLithuanian=UnicodeString("i i\\u0307\\u0301 j j\\u0307\\u0301 \\u01
2f \\u012f\\u0307\\u0301 i\\u0307\\u0300i\\u0307\\u0301i\\u0307\\u0303", "").une
scape(), | |
| 287 iDefault=UnicodeString("i i\\u0301 j j\\u0301 \\u012f \\u012f\\u0301
\\u00ec\\u00ed\\u0129", "").unescape(); | |
| 288 | |
| 289 (t=i).toLower("lt"); | |
| 290 if(t!=iLithuanian) { | |
| 291 errln("error in i.toLower(lt)=\"" + t + "\" expected \"" + iLithuani
an + "\""); | |
| 292 } | |
| 293 | |
| 294 (t=i).toLower("de"); | |
| 295 if(t!=iDefault) { | |
| 296 errln("error in i.toLower(de)=\"" + t + "\" expected \"" + iDefault
+ "\""); | |
| 297 } | |
| 298 } | |
| 299 | |
| 300 #endif | |
| 301 | |
| 302 // test case folding | |
| 303 { | |
| 304 UnicodeString | |
| 305 s=UnicodeString("A\\u00df\\u00b5\\ufb03\\U0001040c\\u0130\\u0131", "
").unescape(), | |
| 306 f=UnicodeString("ass\\u03bcffi\\U00010434i\\u0307\\u0131", "").unesc
ape(), | |
| 307 g=UnicodeString("ass\\u03bcffi\\U00010434i\\u0131", "").unescape(), | |
| 308 t; | |
| 309 | |
| 310 (t=s).foldCase(); | |
| 311 if(f!=t) { | |
| 312 errln("error in foldCase(\"" + s + "\", default)=\"" + t + "\" but e
xpected \"" + f + "\""); | |
| 313 } | |
| 314 | |
| 315 // alternate handling for dotted I/dotless i (U+0130, U+0131) | |
| 316 (t=s).foldCase(U_FOLD_CASE_EXCLUDE_SPECIAL_I); | |
| 317 if(g!=t) { | |
| 318 errln("error in foldCase(\"" + s + "\", U_FOLD_CASE_EXCLUDE_SPECIAL_
I)=\"" + t + "\" but expected \"" + g + "\""); | |
| 319 } | |
| 320 } | |
| 321 } | |
| 322 | |
| 323 // data-driven case mapping tests ------------------------------------------ *** | |
| 324 | |
| 325 enum { | |
| 326 TEST_LOWER, | |
| 327 TEST_UPPER, | |
| 328 TEST_TITLE, | |
| 329 TEST_FOLD, | |
| 330 TEST_COUNT | |
| 331 }; | |
| 332 | |
| 333 // names of TestData children in casing.txt | |
| 334 static const char *const dataNames[TEST_COUNT+1]={ | |
| 335 "lowercasing", | |
| 336 "uppercasing", | |
| 337 "titlecasing", | |
| 338 "casefolding", | |
| 339 "" | |
| 340 }; | |
| 341 | |
| 342 void | |
| 343 StringCaseTest::TestCasingImpl(const UnicodeString &input, | |
| 344 const UnicodeString &output, | |
| 345 int32_t whichCase, | |
| 346 void *iter, const char *localeID, uint32_t option
s) { | |
| 347 // UnicodeString | |
| 348 UnicodeString result; | |
| 349 const char *name; | |
| 350 Locale locale(localeID); | |
| 351 | |
| 352 result=input; | |
| 353 switch(whichCase) { | |
| 354 case TEST_LOWER: | |
| 355 name="toLower"; | |
| 356 result.toLower(locale); | |
| 357 break; | |
| 358 case TEST_UPPER: | |
| 359 name="toUpper"; | |
| 360 result.toUpper(locale); | |
| 361 break; | |
| 362 #if !UCONFIG_NO_BREAK_ITERATION | |
| 363 case TEST_TITLE: | |
| 364 name="toTitle"; | |
| 365 result.toTitle((BreakIterator *)iter, locale, options); | |
| 366 break; | |
| 367 #endif | |
| 368 case TEST_FOLD: | |
| 369 name="foldCase"; | |
| 370 result.foldCase(options); | |
| 371 break; | |
| 372 default: | |
| 373 name=""; | |
| 374 break; // won't happen | |
| 375 } | |
| 376 if(result!=output) { | |
| 377 dataerrln("error: UnicodeString.%s() got a wrong result for a test case
from casing.res", name); | |
| 378 } | |
| 379 #if !UCONFIG_NO_BREAK_ITERATION | |
| 380 if(whichCase==TEST_TITLE && options==0) { | |
| 381 result=input; | |
| 382 result.toTitle((BreakIterator *)iter, locale); | |
| 383 if(result!=output) { | |
| 384 dataerrln("error: UnicodeString.toTitle(options=0) got a wrong resul
t for a test case from casing.res"); | |
| 385 } | |
| 386 } | |
| 387 #endif | |
| 388 | |
| 389 // UTF-8 | |
| 390 char utf8In[100], utf8Out[100]; | |
| 391 int32_t utf8InLength, utf8OutLength, resultLength; | |
| 392 UChar *buffer; | |
| 393 | |
| 394 IcuTestErrorCode errorCode(*this, "TestCasingImpl"); | |
| 395 LocalUCaseMapPointer csm(ucasemap_open(localeID, options, errorCode)); | |
| 396 #if !UCONFIG_NO_BREAK_ITERATION | |
| 397 if(iter!=NULL) { | |
| 398 // Clone the break iterator so that the UCaseMap can safely adopt it. | |
| 399 UBreakIterator *clone=ubrk_safeClone((UBreakIterator *)iter, NULL, NULL,
errorCode); | |
| 400 ucasemap_setBreakIterator(csm.getAlias(), clone, errorCode); | |
| 401 } | |
| 402 #endif | |
| 403 | |
| 404 u_strToUTF8(utf8In, (int32_t)sizeof(utf8In), &utf8InLength, input.getBuffer(
), input.length(), errorCode); | |
| 405 switch(whichCase) { | |
| 406 case TEST_LOWER: | |
| 407 name="ucasemap_utf8ToLower"; | |
| 408 utf8OutLength=ucasemap_utf8ToLower(csm.getAlias(), | |
| 409 utf8Out, (int32_t)sizeof(utf8Out), | |
| 410 utf8In, utf8InLength, errorCode); | |
| 411 break; | |
| 412 case TEST_UPPER: | |
| 413 name="ucasemap_utf8ToUpper"; | |
| 414 utf8OutLength=ucasemap_utf8ToUpper(csm.getAlias(), | |
| 415 utf8Out, (int32_t)sizeof(utf8Out), | |
| 416 utf8In, utf8InLength, errorCode); | |
| 417 break; | |
| 418 #if !UCONFIG_NO_BREAK_ITERATION | |
| 419 case TEST_TITLE: | |
| 420 name="ucasemap_utf8ToTitle"; | |
| 421 utf8OutLength=ucasemap_utf8ToTitle(csm.getAlias(), | |
| 422 utf8Out, (int32_t)sizeof(utf8Out), | |
| 423 utf8In, utf8InLength, errorCode); | |
| 424 break; | |
| 425 #endif | |
| 426 case TEST_FOLD: | |
| 427 name="ucasemap_utf8FoldCase"; | |
| 428 utf8OutLength=ucasemap_utf8FoldCase(csm.getAlias(), | |
| 429 utf8Out, (int32_t)sizeof(utf8Out), | |
| 430 utf8In, utf8InLength, errorCode); | |
| 431 break; | |
| 432 default: | |
| 433 name=""; | |
| 434 utf8OutLength=0; | |
| 435 break; // won't happen | |
| 436 } | |
| 437 buffer=result.getBuffer(utf8OutLength); | |
| 438 u_strFromUTF8(buffer, result.getCapacity(), &resultLength, utf8Out, utf8OutL
ength, errorCode); | |
| 439 result.releaseBuffer(errorCode.isSuccess() ? resultLength : 0); | |
| 440 | |
| 441 if(errorCode.isFailure()) { | |
| 442 errcheckln(errorCode, "error: %s() got an error for a test case from cas
ing.res - %s", name, u_errorName(errorCode)); | |
| 443 errorCode.reset(); | |
| 444 } else if(result!=output) { | |
| 445 errln("error: %s() got a wrong result for a test case from casing.res",
name); | |
| 446 errln("expected \"" + output + "\" got \"" + result + "\"" ); | |
| 447 } | |
| 448 } | |
| 449 | |
| 450 void | |
| 451 StringCaseTest::TestCasing() { | |
| 452 UErrorCode status = U_ZERO_ERROR; | |
| 453 #if !UCONFIG_NO_BREAK_ITERATION | |
| 454 LocalUBreakIteratorPointer iter; | |
| 455 #endif | |
| 456 char cLocaleID[100]; | |
| 457 UnicodeString locale, input, output, optionsString, result; | |
| 458 uint32_t options; | |
| 459 int32_t whichCase, type; | |
| 460 LocalPointer<TestDataModule> driver(TestDataModule::getTestDataModule("casin
g", *this, status)); | |
| 461 if(U_SUCCESS(status)) { | |
| 462 for(whichCase=0; whichCase<TEST_COUNT; ++whichCase) { | |
| 463 #if UCONFIG_NO_BREAK_ITERATION | |
| 464 if(whichCase==TEST_TITLE) { | |
| 465 continue; | |
| 466 } | |
| 467 #endif | |
| 468 LocalPointer<TestData> casingTest(driver->createTestData(dataNames[w
hichCase], status)); | |
| 469 if(U_FAILURE(status)) { | |
| 470 errln("TestCasing failed to createTestData(%s) - %s", dataNames[
whichCase], u_errorName(status)); | |
| 471 break; | |
| 472 } | |
| 473 const DataMap *myCase = NULL; | |
| 474 while(casingTest->nextCase(myCase, status)) { | |
| 475 input = myCase->getString("Input", status); | |
| 476 output = myCase->getString("Output", status); | |
| 477 | |
| 478 if(whichCase!=TEST_FOLD) { | |
| 479 locale = myCase->getString("Locale", status); | |
| 480 } | |
| 481 locale.extract(0, 0x7fffffff, cLocaleID, sizeof(cLocaleID), ""); | |
| 482 | |
| 483 #if !UCONFIG_NO_BREAK_ITERATION | |
| 484 if(whichCase==TEST_TITLE) { | |
| 485 type = myCase->getInt("Type", status); | |
| 486 if(type>=0) { | |
| 487 iter.adoptInstead(ubrk_open((UBreakIteratorType)type, cL
ocaleID, NULL, 0, &status)); | |
| 488 } else if(type==-2) { | |
| 489 // Open a trivial break iterator that only delivers { 0,
length } | |
| 490 // or even just { 0 } as boundaries. | |
| 491 static const UChar rules[] = { 0x2e, 0x2a, 0x3b }; // "
.*;" | |
| 492 UParseError parseError; | |
| 493 iter.adoptInstead(ubrk_openRules(rules, UPRV_LENGTHOF(ru
les), NULL, 0, &parseError, &status)); | |
| 494 } | |
| 495 } | |
| 496 #endif | |
| 497 options = 0; | |
| 498 if(whichCase==TEST_TITLE || whichCase==TEST_FOLD) { | |
| 499 optionsString = myCase->getString("Options", status); | |
| 500 if(optionsString.indexOf((UChar)0x54)>=0) { // T | |
| 501 options|=U_FOLD_CASE_EXCLUDE_SPECIAL_I; | |
| 502 } | |
| 503 if(optionsString.indexOf((UChar)0x4c)>=0) { // L | |
| 504 options|=U_TITLECASE_NO_LOWERCASE; | |
| 505 } | |
| 506 if(optionsString.indexOf((UChar)0x41)>=0) { // A | |
| 507 options|=U_TITLECASE_NO_BREAK_ADJUSTMENT; | |
| 508 } | |
| 509 } | |
| 510 | |
| 511 if(U_FAILURE(status)) { | |
| 512 dataerrln("error: TestCasing() setup failed for %s test case
from casing.res: %s", dataNames[whichCase], u_errorName(status)); | |
| 513 status = U_ZERO_ERROR; | |
| 514 } else { | |
| 515 #if UCONFIG_NO_BREAK_ITERATION | |
| 516 LocalPointer<UMemory> iter; | |
| 517 #endif | |
| 518 TestCasingImpl(input, output, whichCase, iter.getAlias(), cL
ocaleID, options); | |
| 519 } | |
| 520 | |
| 521 #if !UCONFIG_NO_BREAK_ITERATION | |
| 522 iter.adoptInstead(NULL); | |
| 523 #endif | |
| 524 } | |
| 525 } | |
| 526 } | |
| 527 | |
| 528 #if !UCONFIG_NO_BREAK_ITERATION | |
| 529 // more tests for API coverage | |
| 530 status=U_ZERO_ERROR; | |
| 531 input=UNICODE_STRING_SIMPLE("sTrA\\u00dfE").unescape(); | |
| 532 (result=input).toTitle(NULL); | |
| 533 if(result!=UNICODE_STRING_SIMPLE("Stra\\u00dfe").unescape()) { | |
| 534 dataerrln("UnicodeString::toTitle(NULL) failed."); | |
| 535 } | |
| 536 #endif | |
| 537 } | |
| 538 | |
| 539 void | |
| 540 StringCaseTest::TestFullCaseFoldingIterator() { | |
| 541 UnicodeString ffi=UNICODE_STRING_SIMPLE("ffi"); | |
| 542 UnicodeString ss=UNICODE_STRING_SIMPLE("ss"); | |
| 543 FullCaseFoldingIterator iter; | |
| 544 int32_t count=0; | |
| 545 int32_t countSpecific=0; | |
| 546 UChar32 c; | |
| 547 UnicodeString full; | |
| 548 while((c=iter.next(full))>=0) { | |
| 549 ++count; | |
| 550 // Check that the full Case_Folding has more than 1 code point. | |
| 551 if(!full.hasMoreChar32Than(0, 0x7fffffff, 1)) { | |
| 552 errln("error: FullCaseFoldingIterator.next()=U+%04lX full Case_Foldi
ng has at most 1 code point", (long)c); | |
| 553 continue; | |
| 554 } | |
| 555 // Check that full == Case_Folding(c). | |
| 556 UnicodeString cf(c); | |
| 557 cf.foldCase(); | |
| 558 if(full!=cf) { | |
| 559 errln("error: FullCaseFoldingIterator.next()=U+%04lX full Case_Foldi
ng != cf(c)", (long)c); | |
| 560 continue; | |
| 561 } | |
| 562 // Spot-check a couple of specific cases. | |
| 563 if((full==ffi && c==0xfb03) || (full==ss && (c==0xdf || c==0x1e9e))) { | |
| 564 ++countSpecific; | |
| 565 } | |
| 566 } | |
| 567 if(countSpecific!=3) { | |
| 568 errln("error: FullCaseFoldingIterator did not yield exactly the expected
specific cases"); | |
| 569 } | |
| 570 if(count<70) { | |
| 571 errln("error: FullCaseFoldingIterator yielded only %d (cp, full) pairs",
(int)count); | |
| 572 } | |
| 573 } | |
| OLD | NEW |