Frank Tang | 3e05d9d | 2021-11-08 14:04:04 -0800 | [diff] [blame] | 1 | // © 2016 and later: Unicode, Inc. and others. |
| 2 | // License & terms of use: http://www.unicode.org/copyright.html |
| 3 | /* |
| 4 | ******************************************************************************* |
| 5 | * |
| 6 | * Copyright (C) 2002-2016, International Business Machines |
| 7 | * Corporation and others. All Rights Reserved. |
| 8 | * |
| 9 | ******************************************************************************* |
| 10 | * file name: strcase.cpp |
| 11 | * encoding: UTF-8 |
| 12 | * tab size: 8 (not used) |
| 13 | * indentation:4 |
| 14 | * |
| 15 | * created on: 2002mar12 |
| 16 | * created by: Markus W. Scherer |
| 17 | * |
| 18 | * Test file for string casing C++ API functions. |
| 19 | */ |
| 20 | |
| 21 | #include "unicode/std_string.h" |
| 22 | #include "unicode/brkiter.h" |
| 23 | #include "unicode/casemap.h" |
| 24 | #include "unicode/edits.h" |
| 25 | #include "unicode/uchar.h" |
| 26 | #include "unicode/ures.h" |
| 27 | #include "unicode/uloc.h" |
| 28 | #include "unicode/locid.h" |
| 29 | #include "unicode/ubrk.h" |
| 30 | #include "unicode/unistr.h" |
| 31 | #include "unicode/ucasemap.h" |
| 32 | #include "unicode/ustring.h" |
| 33 | #include "ucase.h" |
| 34 | #include "ustrtest.h" |
| 35 | #include "unicode/tstdtmod.h" |
| 36 | #include "cmemory.h" |
| 37 | #include "testutil.h" |
| 38 | |
| 39 | class StringCaseTest: public IntlTest { |
| 40 | public: |
| 41 | StringCaseTest(); |
| 42 | virtual ~StringCaseTest(); |
| 43 | |
| 44 | void runIndexedTest(int32_t index, UBool exec, const char *&name, char *par=0) override; |
| 45 | |
| 46 | void TestCaseConversion(); |
| 47 | |
| 48 | void TestCasingImpl(const UnicodeString &input, |
| 49 | const UnicodeString &output, |
| 50 | int32_t whichCase, |
| 51 | void *iter, const char *localeID, uint32_t options); |
| 52 | void TestCasing(); |
| 53 | void TestTitleOptions(); |
Frank Tang | d2858cb | 2022-04-08 20:34:12 -0700 | [diff] [blame] | 54 | void TestDutchTitle(); |
Frank Tang | 3e05d9d | 2021-11-08 14:04:04 -0800 | [diff] [blame] | 55 | void TestFullCaseFoldingIterator(); |
| 56 | void TestGreekUpper(); |
| 57 | void TestArmenian(); |
| 58 | void TestLongUpper(); |
| 59 | void TestMalformedUTF8(); |
| 60 | void TestBufferOverflow(); |
| 61 | void TestEdits(); |
| 62 | void TestCopyMoveEdits(); |
| 63 | void TestEditsFindFwdBwd(); |
| 64 | void TestMergeEdits(); |
| 65 | void TestCaseMapWithEdits(); |
| 66 | void TestCaseMapUTF8WithEdits(); |
| 67 | void TestCaseMapToString(); |
| 68 | void TestCaseMapUTF8ToString(); |
| 69 | void TestLongUnicodeString(); |
| 70 | void TestBug13127(); |
| 71 | void TestInPlaceTitle(); |
| 72 | void TestCaseMapEditsIteratorDocs(); |
| 73 | void TestCaseMapGreekExtended(); |
| 74 | |
| 75 | private: |
| 76 | void assertGreekUpper(const char16_t *s, const char16_t *expected); |
| 77 | |
| 78 | Locale GREEK_LOCALE_; |
| 79 | }; |
| 80 | |
| 81 | StringCaseTest::StringCaseTest() : GREEK_LOCALE_("el") {} |
| 82 | |
| 83 | StringCaseTest::~StringCaseTest() {} |
| 84 | |
| 85 | extern IntlTest *createStringCaseTest() { |
| 86 | return new StringCaseTest(); |
| 87 | } |
| 88 | |
| 89 | void |
| 90 | StringCaseTest::runIndexedTest(int32_t index, UBool exec, const char *&name, char * /*par*/) { |
| 91 | if(exec) { |
| 92 | logln("TestSuite StringCaseTest: "); |
| 93 | } |
| 94 | TESTCASE_AUTO_BEGIN; |
| 95 | TESTCASE_AUTO(TestCaseConversion); |
| 96 | #if !UCONFIG_NO_BREAK_ITERATION && !UCONFIG_NO_FILE_IO && !UCONFIG_NO_LEGACY_CONVERSION |
| 97 | TESTCASE_AUTO(TestCasing); |
| 98 | TESTCASE_AUTO(TestTitleOptions); |
Frank Tang | d2858cb | 2022-04-08 20:34:12 -0700 | [diff] [blame] | 99 | TESTCASE_AUTO(TestDutchTitle); |
Frank Tang | 3e05d9d | 2021-11-08 14:04:04 -0800 | [diff] [blame] | 100 | #endif |
| 101 | TESTCASE_AUTO(TestFullCaseFoldingIterator); |
| 102 | TESTCASE_AUTO(TestGreekUpper); |
| 103 | TESTCASE_AUTO(TestArmenian); |
| 104 | TESTCASE_AUTO(TestLongUpper); |
| 105 | TESTCASE_AUTO(TestMalformedUTF8); |
| 106 | TESTCASE_AUTO(TestBufferOverflow); |
| 107 | TESTCASE_AUTO(TestEdits); |
| 108 | TESTCASE_AUTO(TestCopyMoveEdits); |
| 109 | TESTCASE_AUTO(TestEditsFindFwdBwd); |
| 110 | TESTCASE_AUTO(TestMergeEdits); |
| 111 | TESTCASE_AUTO(TestCaseMapWithEdits); |
| 112 | TESTCASE_AUTO(TestCaseMapUTF8WithEdits); |
| 113 | TESTCASE_AUTO(TestCaseMapToString); |
| 114 | TESTCASE_AUTO(TestCaseMapUTF8ToString); |
| 115 | TESTCASE_AUTO(TestLongUnicodeString); |
| 116 | #if !UCONFIG_NO_BREAK_ITERATION |
| 117 | TESTCASE_AUTO(TestBug13127); |
| 118 | TESTCASE_AUTO(TestInPlaceTitle); |
| 119 | #endif |
| 120 | TESTCASE_AUTO(TestCaseMapEditsIteratorDocs); |
| 121 | TESTCASE_AUTO(TestCaseMapGreekExtended); |
| 122 | TESTCASE_AUTO_END; |
| 123 | } |
| 124 | |
| 125 | void |
| 126 | StringCaseTest::TestCaseConversion() |
| 127 | { |
| 128 | static const UChar uppercaseGreek[] = |
| 129 | { 0x399, 0x395, 0x3a3, 0x3a5, 0x3a3, 0x20, 0x03a7, 0x3a1, 0x399, 0x3a3, 0x3a4, |
| 130 | 0x39f, 0x3a3, 0 }; |
| 131 | // "IESUS CHRISTOS" |
| 132 | |
| 133 | static const UChar lowercaseGreek[] = |
| 134 | { 0x3b9, 0x3b5, 0x3c3, 0x3c5, 0x3c2, 0x20, 0x03c7, 0x3c1, 0x3b9, 0x3c3, 0x3c4, |
| 135 | 0x3bf, 0x3c2, 0 }; |
| 136 | // "iesus christos" |
| 137 | |
| 138 | static const UChar lowercaseTurkish[] = |
| 139 | { 0x69, 0x73, 0x74, 0x61, 0x6e, 0x62, 0x75, 0x6c, 0x2c, 0x20, 0x6e, 0x6f, 0x74, 0x20, 0x63, 0x6f, |
| 140 | 0x6e, 0x73, 0x74, 0x61, 0x6e, 0x74, 0x0131, 0x6e, 0x6f, 0x70, 0x6c, 0x65, 0x21, 0 }; |
| 141 | |
| 142 | static const UChar uppercaseTurkish[] = |
| 143 | { 0x54, 0x4f, 0x50, 0x4b, 0x41, 0x50, 0x49, 0x20, 0x50, 0x41, 0x4c, 0x41, 0x43, 0x45, 0x2c, 0x20, |
| 144 | 0x0130, 0x53, 0x54, 0x41, 0x4e, 0x42, 0x55, 0x4c, 0 }; |
| 145 | |
| 146 | UnicodeString expectedResult; |
| 147 | UnicodeString test3; |
| 148 | |
| 149 | test3 += (UChar32)0x0130; |
| 150 | test3 += "STANBUL, NOT CONSTANTINOPLE!"; |
| 151 | |
| 152 | UnicodeString test4(test3); |
| 153 | test4.toLower(Locale("")); |
| 154 | expectedResult = UnicodeString("i\\u0307stanbul, not constantinople!", "").unescape(); |
| 155 | if (test4 != expectedResult) |
| 156 | errln("1. toLower failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 157 | |
| 158 | test4 = test3; |
| 159 | test4.toLower(Locale("tr", "TR")); |
| 160 | expectedResult = lowercaseTurkish; |
| 161 | if (test4 != expectedResult) |
| 162 | errln("2. toLower failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 163 | |
| 164 | test3 = "topkap"; |
| 165 | test3 += (UChar32)0x0131; |
| 166 | test3 += " palace, istanbul"; |
| 167 | test4 = test3; |
| 168 | |
| 169 | test4.toUpper(Locale("")); |
| 170 | expectedResult = "TOPKAPI PALACE, ISTANBUL"; |
| 171 | if (test4 != expectedResult) |
| 172 | errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 173 | |
| 174 | test4 = test3; |
| 175 | test4.toUpper(Locale("tr", "TR")); |
| 176 | expectedResult = uppercaseTurkish; |
| 177 | if (test4 != expectedResult) |
| 178 | errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 179 | |
| 180 | test3 = CharsToUnicodeString("S\\u00FC\\u00DFmayrstra\\u00DFe"); |
| 181 | |
| 182 | test3.toUpper(Locale("de", "DE")); |
| 183 | expectedResult = CharsToUnicodeString("S\\u00DCSSMAYRSTRASSE"); |
| 184 | if (test3 != expectedResult) |
| 185 | errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + test3 + "\"."); |
| 186 | |
| 187 | test4.replace(0, test4.length(), uppercaseGreek); |
| 188 | |
| 189 | test4.toLower(Locale("el", "GR")); |
| 190 | expectedResult = lowercaseGreek; |
| 191 | if (test4 != expectedResult) |
| 192 | errln("toLower failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 193 | |
| 194 | test4.replace(0, test4.length(), lowercaseGreek); |
| 195 | |
| 196 | test4.toUpper(); |
| 197 | expectedResult = uppercaseGreek; |
| 198 | if (test4 != expectedResult) |
| 199 | errln("toUpper failed: expected \"" + expectedResult + "\", got \"" + test4 + "\"."); |
| 200 | |
| 201 | // more string case mapping tests with the new implementation |
| 202 | { |
| 203 | static const UChar |
| 204 | |
| 205 | beforeLower[]= { 0x61, 0x42, 0x49, 0x3a3, 0xdf, 0x3a3, 0x2f, 0xd93f, 0xdfff }, |
| 206 | lowerRoot[]= { 0x61, 0x62, 0x69, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0xdfff }, |
| 207 | lowerTurkish[]={ 0x61, 0x62, 0x131, 0x3c3, 0xdf, 0x3c2, 0x2f, 0xd93f, 0xdfff }, |
| 208 | |
| 209 | beforeUpper[]= { 0x61, 0x42, 0x69, 0x3c2, 0xdf, 0x3c3, 0x2f, 0xfb03, 0xfb03, 0xfb03, 0xd93f, 0xdfff }, |
| 210 | upperRoot[]= { 0x41, 0x42, 0x49, 0x3a3, 0x53, 0x53, 0x3a3, 0x2f, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0xd93f, 0xdfff }, |
| 211 | upperTurkish[]={ 0x41, 0x42, 0x130, 0x3a3, 0x53, 0x53, 0x3a3, 0x2f, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0x46, 0x46, 0x49, 0xd93f, 0xdfff }, |
| 212 | |
| 213 | beforeMiniUpper[]= { 0xdf, 0x61 }, |
| 214 | miniUpper[]= { 0x53, 0x53, 0x41 }; |
| 215 | |
| 216 | UnicodeString s; |
| 217 | |
| 218 | /* lowercase with root locale */ |
| 219 | s=UnicodeString(FALSE, beforeLower, UPRV_LENGTHOF(beforeLower)); |
| 220 | s.toLower(""); |
| 221 | if( s.length()!=UPRV_LENGTHOF(lowerRoot) || |
| 222 | s!=UnicodeString(FALSE, lowerRoot, s.length()) |
| 223 | ) { |
| 224 | errln("error in toLower(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerRoot, UPRV_LENGTHOF(lowerRoot)) + "\""); |
| 225 | } |
| 226 | |
| 227 | /* lowercase with turkish locale */ |
| 228 | s=UnicodeString(FALSE, beforeLower, UPRV_LENGTHOF(beforeLower)); |
| 229 | s.setCharAt(0, beforeLower[0]).toLower(Locale("tr")); |
| 230 | if( s.length()!=UPRV_LENGTHOF(lowerTurkish) || |
| 231 | s!=UnicodeString(FALSE, lowerTurkish, s.length()) |
| 232 | ) { |
| 233 | errln("error in toLower(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, lowerTurkish, UPRV_LENGTHOF(lowerTurkish)) + "\""); |
| 234 | } |
| 235 | |
| 236 | /* uppercase with root locale */ |
| 237 | s=UnicodeString(FALSE, beforeUpper, UPRV_LENGTHOF(beforeUpper)); |
| 238 | s.setCharAt(0, beforeUpper[0]).toUpper(Locale("")); |
| 239 | if( s.length()!=UPRV_LENGTHOF(upperRoot) || |
| 240 | s!=UnicodeString(FALSE, upperRoot, s.length()) |
| 241 | ) { |
| 242 | errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperRoot, UPRV_LENGTHOF(upperRoot)) + "\""); |
| 243 | } |
| 244 | |
| 245 | /* uppercase with turkish locale */ |
| 246 | s=UnicodeString(FALSE, beforeUpper, UPRV_LENGTHOF(beforeUpper)); |
| 247 | s.toUpper(Locale("tr")); |
| 248 | if( s.length()!=UPRV_LENGTHOF(upperTurkish) || |
| 249 | s!=UnicodeString(FALSE, upperTurkish, s.length()) |
| 250 | ) { |
| 251 | errln("error in toUpper(turkish locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, upperTurkish, UPRV_LENGTHOF(upperTurkish)) + "\""); |
| 252 | } |
| 253 | |
| 254 | /* uppercase a short string with root locale */ |
| 255 | s=UnicodeString(FALSE, beforeMiniUpper, UPRV_LENGTHOF(beforeMiniUpper)); |
| 256 | s.setCharAt(0, beforeMiniUpper[0]).toUpper(""); |
| 257 | if( s.length()!=UPRV_LENGTHOF(miniUpper) || |
| 258 | s!=UnicodeString(FALSE, miniUpper, s.length()) |
| 259 | ) { |
| 260 | errln("error in toUpper(root locale)=\"" + s + "\" expected \"" + UnicodeString(FALSE, miniUpper, UPRV_LENGTHOF(miniUpper)) + "\""); |
| 261 | } |
| 262 | } |
| 263 | |
| 264 | // test some supplementary characters (>= Unicode 3.1) |
| 265 | { |
| 266 | UnicodeString t; |
| 267 | |
| 268 | UnicodeString |
| 269 | deseretInput=UnicodeString("\\U0001043C\\U00010414", "").unescape(), |
| 270 | deseretLower=UnicodeString("\\U0001043C\\U0001043C", "").unescape(), |
| 271 | deseretUpper=UnicodeString("\\U00010414\\U00010414", "").unescape(); |
| 272 | (t=deseretInput).toLower(); |
| 273 | if(t!=deseretLower) { |
| 274 | errln("error lowercasing Deseret (plane 1) characters"); |
| 275 | } |
| 276 | (t=deseretInput).toUpper(); |
| 277 | if(t!=deseretUpper) { |
| 278 | errln("error uppercasing Deseret (plane 1) characters"); |
| 279 | } |
| 280 | } |
| 281 | |
| 282 | // test some more cases that looked like problems |
| 283 | { |
| 284 | UnicodeString t; |
| 285 | |
| 286 | UnicodeString |
| 287 | ljInput=UnicodeString("ab'cD \\uFB00i\\u0131I\\u0130 \\u01C7\\u01C8\\u01C9 \\U0001043C\\U00010414", "").unescape(), |
| 288 | ljLower=UnicodeString("ab'cd \\uFB00i\\u0131ii\\u0307 \\u01C9\\u01C9\\u01C9 \\U0001043C\\U0001043C", "").unescape(), |
| 289 | ljUpper=UnicodeString("AB'CD FFIII\\u0130 \\u01C7\\u01C7\\u01C7 \\U00010414\\U00010414", "").unescape(); |
| 290 | (t=ljInput).toLower("en"); |
| 291 | if(t!=ljLower) { |
| 292 | errln("error lowercasing LJ characters"); |
| 293 | } |
| 294 | (t=ljInput).toUpper("en"); |
| 295 | if(t!=ljUpper) { |
| 296 | errln("error uppercasing LJ characters"); |
| 297 | } |
| 298 | } |
| 299 | |
| 300 | #if !UCONFIG_NO_NORMALIZATION |
| 301 | // some context-sensitive casing depends on normalization data being present |
| 302 | |
| 303 | // Unicode 3.1.1 SpecialCasing tests |
| 304 | { |
| 305 | UnicodeString t; |
| 306 | |
| 307 | // sigmas preceded and/or followed by cased letters |
| 308 | UnicodeString |
| 309 | sigmas=UnicodeString("i\\u0307\\u03a3\\u0308j \\u0307\\u03a3\\u0308j i\\u00ad\\u03a3\\u0308 \\u0307\\u03a3\\u0308 ", "").unescape(), |
| 310 | sigmasLower=UnicodeString("i\\u0307\\u03c3\\u0308j \\u0307\\u03c3\\u0308j i\\u00ad\\u03c2\\u0308 \\u0307\\u03c3\\u0308 ", "").unescape(), |
| 311 | sigmasUpper=UnicodeString("I\\u0307\\u03a3\\u0308J \\u0307\\u03a3\\u0308J I\\u00ad\\u03a3\\u0308 \\u0307\\u03a3\\u0308 ", "").unescape(); |
| 312 | |
| 313 | (t=sigmas).toLower(); |
| 314 | if(t!=sigmasLower) { |
| 315 | errln("error in sigmas.toLower()=\"" + t + "\" expected \"" + sigmasLower + "\""); |
| 316 | } |
| 317 | |
| 318 | (t=sigmas).toUpper(Locale("")); |
| 319 | if(t!=sigmasUpper) { |
| 320 | errln("error in sigmas.toUpper()=\"" + t + "\" expected \"" + sigmasUpper + "\""); |
| 321 | } |
| 322 | |
| 323 | // turkish & azerbaijani dotless i & dotted I |
| 324 | // remove dot above if there was a capital I before and there are no more accents above |
| 325 | UnicodeString |
| 326 | dots=UnicodeString("I \\u0130 I\\u0307 I\\u0327\\u0307 I\\u0301\\u0307 I\\u0327\\u0307\\u0301", "").unescape(), |
| 327 | dotsTurkish=UnicodeString("\\u0131 i i i\\u0327 \\u0131\\u0301\\u0307 i\\u0327\\u0301", "").unescape(), |
| 328 | dotsDefault=UnicodeString("i i\\u0307 i\\u0307 i\\u0327\\u0307 i\\u0301\\u0307 i\\u0327\\u0307\\u0301", "").unescape(); |
| 329 | |
| 330 | (t=dots).toLower("tr"); |
| 331 | if(t!=dotsTurkish) { |
| 332 | errln("error in dots.toLower(tr)=\"" + t + "\" expected \"" + dotsTurkish + "\""); |
| 333 | } |
| 334 | |
| 335 | (t=dots).toLower("de"); |
| 336 | if(t!=dotsDefault) { |
| 337 | errln("error in dots.toLower(de)=\"" + t + "\" expected \"" + dotsDefault + "\""); |
| 338 | } |
| 339 | } |
| 340 | |
| 341 | // more Unicode 3.1.1 tests |
| 342 | { |
| 343 | UnicodeString t; |
| 344 | |
| 345 | // lithuanian dot above in uppercasing |
| 346 | UnicodeString |
| 347 | dots=UnicodeString("a\\u0307 \\u0307 i\\u0307 j\\u0327\\u0307 j\\u0301\\u0307", "").unescape(), |
| 348 | dotsLithuanian=UnicodeString("A\\u0307 \\u0307 I J\\u0327 J\\u0301\\u0307", "").unescape(), |
| 349 | dotsDefault=UnicodeString("A\\u0307 \\u0307 I\\u0307 J\\u0327\\u0307 J\\u0301\\u0307", "").unescape(); |
| 350 | |
| 351 | (t=dots).toUpper("lt"); |
| 352 | if(t!=dotsLithuanian) { |
| 353 | errln("error in dots.toUpper(lt)=\"" + t + "\" expected \"" + dotsLithuanian + "\""); |
| 354 | } |
| 355 | |
| 356 | (t=dots).toUpper("de"); |
| 357 | if(t!=dotsDefault) { |
| 358 | errln("error in dots.toUpper(de)=\"" + t + "\" expected \"" + dotsDefault + "\""); |
| 359 | } |
| 360 | |
| 361 | // lithuanian adds dot above to i in lowercasing if there are more above accents |
| 362 | UnicodeString |
| 363 | i=UnicodeString("I I\\u0301 J J\\u0301 \\u012e \\u012e\\u0301 \\u00cc\\u00cd\\u0128", "").unescape(), |
| 364 | iLithuanian=UnicodeString("i i\\u0307\\u0301 j j\\u0307\\u0301 \\u012f \\u012f\\u0307\\u0301 i\\u0307\\u0300i\\u0307\\u0301i\\u0307\\u0303", "").unescape(), |
| 365 | iDefault=UnicodeString("i i\\u0301 j j\\u0301 \\u012f \\u012f\\u0301 \\u00ec\\u00ed\\u0129", "").unescape(); |
| 366 | |
| 367 | (t=i).toLower("lt"); |
| 368 | if(t!=iLithuanian) { |
| 369 | errln("error in i.toLower(lt)=\"" + t + "\" expected \"" + iLithuanian + "\""); |
| 370 | } |
| 371 | |
| 372 | (t=i).toLower("de"); |
| 373 | if(t!=iDefault) { |
| 374 | errln("error in i.toLower(de)=\"" + t + "\" expected \"" + iDefault + "\""); |
| 375 | } |
| 376 | } |
| 377 | |
| 378 | #endif |
| 379 | |
| 380 | // test case folding |
| 381 | { |
| 382 | UnicodeString |
| 383 | s=UnicodeString("A\\u00df\\u00b5\\ufb03\\U0001040c\\u0130\\u0131", "").unescape(), |
| 384 | f=UnicodeString("ass\\u03bcffi\\U00010434i\\u0307\\u0131", "").unescape(), |
| 385 | g=UnicodeString("ass\\u03bcffi\\U00010434i\\u0131", "").unescape(), |
| 386 | t; |
| 387 | |
| 388 | (t=s).foldCase(); |
| 389 | if(f!=t) { |
| 390 | errln("error in foldCase(\"" + s + "\", default)=\"" + t + "\" but expected \"" + f + "\""); |
| 391 | } |
| 392 | |
| 393 | // alternate handling for dotted I/dotless i (U+0130, U+0131) |
| 394 | (t=s).foldCase(U_FOLD_CASE_EXCLUDE_SPECIAL_I); |
| 395 | if(g!=t) { |
| 396 | errln("error in foldCase(\"" + s + "\", U_FOLD_CASE_EXCLUDE_SPECIAL_I)=\"" + t + "\" but expected \"" + g + "\""); |
| 397 | } |
| 398 | } |
| 399 | } |
| 400 | |
| 401 | // data-driven case mapping tests ------------------------------------------ *** |
| 402 | |
| 403 | enum { |
| 404 | TEST_LOWER, |
| 405 | TEST_UPPER, |
| 406 | TEST_TITLE, |
| 407 | TEST_FOLD, |
| 408 | TEST_COUNT |
| 409 | }; |
| 410 | |
| 411 | // names of TestData children in casing.txt |
| 412 | static const char *const dataNames[TEST_COUNT+1]={ |
| 413 | "lowercasing", |
| 414 | "uppercasing", |
| 415 | "titlecasing", |
| 416 | "casefolding", |
| 417 | "" |
| 418 | }; |
| 419 | |
| 420 | void |
| 421 | StringCaseTest::TestCasingImpl(const UnicodeString &input, |
| 422 | const UnicodeString &output, |
| 423 | int32_t whichCase, |
| 424 | void *iter, const char *localeID, uint32_t options) { |
| 425 | // UnicodeString |
| 426 | UnicodeString result; |
| 427 | const char *name; |
| 428 | Locale locale(localeID); |
| 429 | |
| 430 | result=input; |
| 431 | switch(whichCase) { |
| 432 | case TEST_LOWER: |
| 433 | name="toLower"; |
| 434 | result.toLower(locale); |
| 435 | break; |
| 436 | case TEST_UPPER: |
| 437 | name="toUpper"; |
| 438 | result.toUpper(locale); |
| 439 | break; |
| 440 | #if !UCONFIG_NO_BREAK_ITERATION |
| 441 | case TEST_TITLE: |
| 442 | name="toTitle"; |
| 443 | result.toTitle((BreakIterator *)iter, locale, options); |
| 444 | break; |
| 445 | #endif |
| 446 | case TEST_FOLD: |
| 447 | name="foldCase"; |
| 448 | result.foldCase(options); |
| 449 | break; |
| 450 | default: |
| 451 | name=""; |
| 452 | break; // won't happen |
| 453 | } |
| 454 | if(result!=output) { |
| 455 | dataerrln("error: UnicodeString.%s() got a wrong result for a test case from casing.res", name); |
Frank Tang | d2858cb | 2022-04-08 20:34:12 -0700 | [diff] [blame] | 456 | dataerrln(UnicodeString("input = [") + input + "], expected = [" + output + "], actual = [" + result + "]"); |
Frank Tang | 3e05d9d | 2021-11-08 14:04:04 -0800 | [diff] [blame] | 457 | } |
| 458 | #if !UCONFIG_NO_BREAK_ITERATION |
| 459 | if(whichCase==TEST_TITLE && options==0) { |
| 460 | result=input; |
| 461 | result.toTitle((BreakIterator *)iter, locale); |
| 462 | if(result!=output) { |
| 463 | dataerrln("error: UnicodeString.toTitle(options=0) got a wrong result for a test case from casing.res"); |
| 464 | } |
| 465 | } |
| 466 | #endif |
| 467 | |
| 468 | // UTF-8 |
| 469 | char utf8In[100], utf8Out[100]; |
| 470 | int32_t utf8InLength, utf8OutLength, resultLength; |
| 471 | UChar *buffer; |
| 472 | |
| 473 | IcuTestErrorCode errorCode(*this, "TestCasingImpl"); |
| 474 | LocalUCaseMapPointer csm(ucasemap_open(localeID, options, errorCode)); |
| 475 | #if !UCONFIG_NO_BREAK_ITERATION |
| 476 | if(iter!=NULL) { |
| 477 | // Clone the break iterator so that the UCaseMap can safely adopt it. |
| 478 | UBreakIterator *clone=ubrk_safeClone((UBreakIterator *)iter, NULL, NULL, errorCode); |
| 479 | ucasemap_setBreakIterator(csm.getAlias(), clone, errorCode); |
| 480 | } |
| 481 | #endif |
| 482 | |
| 483 | u_strToUTF8(utf8In, (int32_t)sizeof(utf8In), &utf8InLength, input.getBuffer(), input.length(), errorCode); |
| 484 | switch(whichCase) { |
| 485 | case TEST_LOWER: |
| 486 | name="ucasemap_utf8ToLower"; |
| 487 | utf8OutLength=ucasemap_utf8ToLower(csm.getAlias(), |
| 488 | utf8Out, (int32_t)sizeof(utf8Out), |
| 489 | utf8In, utf8InLength, errorCode); |
| 490 | break; |
| 491 | case TEST_UPPER: |
| 492 | name="ucasemap_utf8ToUpper"; |
| 493 | utf8OutLength=ucasemap_utf8ToUpper(csm.getAlias(), |
| 494 | utf8Out, (int32_t)sizeof(utf8Out), |
| 495 | utf8In, utf8InLength, errorCode); |
| 496 | break; |
| 497 | #if !UCONFIG_NO_BREAK_ITERATION |
| 498 | case TEST_TITLE: |
| 499 | name="ucasemap_utf8ToTitle"; |
| 500 | utf8OutLength=ucasemap_utf8ToTitle(csm.getAlias(), |
| 501 | utf8Out, (int32_t)sizeof(utf8Out), |
| 502 | utf8In, utf8InLength, errorCode); |
| 503 | break; |
| 504 | #endif |
| 505 | case TEST_FOLD: |
| 506 | name="ucasemap_utf8FoldCase"; |
| 507 | utf8OutLength=ucasemap_utf8FoldCase(csm.getAlias(), |
| 508 | utf8Out, (int32_t)sizeof(utf8Out), |
| 509 | utf8In, utf8InLength, errorCode); |
| 510 | break; |
| 511 | default: |
| 512 | name=""; |
| 513 | utf8OutLength=0; |
| 514 | break; // won't happen |
| 515 | } |
| 516 | buffer=result.getBuffer(utf8OutLength); |
| 517 | u_strFromUTF8(buffer, result.getCapacity(), &resultLength, utf8Out, utf8OutLength, errorCode); |
| 518 | result.releaseBuffer(errorCode.isSuccess() ? resultLength : 0); |
| 519 | |
| 520 | if(errorCode.isFailure()) { |
| 521 | errcheckln(errorCode, "error: %s() got an error for a test case from casing.res - %s", name, u_errorName(errorCode)); |
| 522 | errorCode.reset(); |
| 523 | } else if(result!=output) { |
| 524 | errln("error: %s() got a wrong result for a test case from casing.res", name); |
| 525 | errln("expected \"" + output + "\" got \"" + result + "\"" ); |
| 526 | } |
| 527 | } |
| 528 | |
| 529 | void |
| 530 | StringCaseTest::TestCasing() { |
| 531 | UErrorCode status = U_ZERO_ERROR; |
| 532 | #if !UCONFIG_NO_BREAK_ITERATION |
| 533 | LocalUBreakIteratorPointer iter; |
| 534 | #endif |
| 535 | char cLocaleID[100]; |
| 536 | UnicodeString locale, input, output, optionsString, result; |
| 537 | uint32_t options; |
| 538 | int32_t whichCase, type; |
| 539 | LocalPointer<TestDataModule> driver(TestDataModule::getTestDataModule("casing", *this, status)); |
| 540 | if(U_SUCCESS(status)) { |
| 541 | for(whichCase=0; whichCase<TEST_COUNT; ++whichCase) { |
| 542 | #if UCONFIG_NO_BREAK_ITERATION |
| 543 | if(whichCase==TEST_TITLE) { |
| 544 | continue; |
| 545 | } |
| 546 | #endif |
| 547 | LocalPointer<TestData> casingTest(driver->createTestData(dataNames[whichCase], status)); |
| 548 | if(U_FAILURE(status)) { |
| 549 | errln("TestCasing failed to createTestData(%s) - %s", dataNames[whichCase], u_errorName(status)); |
| 550 | break; |
| 551 | } |
| 552 | const DataMap *myCase = NULL; |
| 553 | while(casingTest->nextCase(myCase, status)) { |
| 554 | input = myCase->getString("Input", status); |
| 555 | output = myCase->getString("Output", status); |
| 556 | |
| 557 | if(whichCase!=TEST_FOLD) { |
| 558 | locale = myCase->getString("Locale", status); |
| 559 | } |
| 560 | locale.extract(0, 0x7fffffff, cLocaleID, sizeof(cLocaleID), ""); |
| 561 | |
| 562 | #if !UCONFIG_NO_BREAK_ITERATION |
| 563 | if(whichCase==TEST_TITLE) { |
| 564 | type = myCase->getInt("Type", status); |
| 565 | if(type>=0) { |
| 566 | iter.adoptInstead(ubrk_open((UBreakIteratorType)type, cLocaleID, NULL, 0, &status)); |
| 567 | } else if(type==-2) { |
| 568 | // Open a trivial break iterator that only delivers { 0, length } |
| 569 | // or even just { 0 } as boundaries. |
| 570 | static const UChar rules[] = { 0x2e, 0x2a, 0x3b }; // ".*;" |
| 571 | UParseError parseError; |
| 572 | iter.adoptInstead(ubrk_openRules(rules, UPRV_LENGTHOF(rules), NULL, 0, &parseError, &status)); |
| 573 | } |
| 574 | } |
| 575 | #endif |
| 576 | options = 0; |
| 577 | if(whichCase==TEST_TITLE || whichCase==TEST_FOLD) { |
| 578 | optionsString = myCase->getString("Options", status); |
| 579 | if(optionsString.indexOf((UChar)0x54)>=0) { // T |
| 580 | options|=U_FOLD_CASE_EXCLUDE_SPECIAL_I; |
| 581 | } |
| 582 | if(optionsString.indexOf((UChar)0x4c)>=0) { // L |
| 583 | options|=U_TITLECASE_NO_LOWERCASE; |
| 584 | } |
| 585 | if(optionsString.indexOf((UChar)0x41)>=0) { // A |
| 586 | options|=U_TITLECASE_NO_BREAK_ADJUSTMENT; |
| 587 | } |
| 588 | } |
| 589 | |
| 590 | if(U_FAILURE(status)) { |
| 591 | dataerrln("error: TestCasing() setup failed for %s test case from casing.res: %s", dataNames[whichCase], u_errorName(status)); |
| 592 | status = U_ZERO_ERROR; |
| 593 | } else { |
| 594 | #if UCONFIG_NO_BREAK_ITERATION |
| 595 | LocalPointer<UMemory> iter; |
| 596 | #endif |
| 597 | TestCasingImpl(input, output, whichCase, iter.getAlias(), cLocaleID, options); |
| 598 | } |
| 599 | |
| 600 | #if !UCONFIG_NO_BREAK_ITERATION |
| 601 | iter.adoptInstead(NULL); |
| 602 | #endif |
| 603 | } |
| 604 | } |
| 605 | } |
| 606 | |
| 607 | #if !UCONFIG_NO_BREAK_ITERATION |
| 608 | // more tests for API coverage |
| 609 | status=U_ZERO_ERROR; |
| 610 | input=UNICODE_STRING_SIMPLE("sTrA\\u00dfE").unescape(); |
| 611 | (result=input).toTitle(NULL); |
| 612 | if(result!=UNICODE_STRING_SIMPLE("Stra\\u00dfe").unescape()) { |
| 613 | dataerrln("UnicodeString::toTitle(NULL) failed."); |
| 614 | } |
| 615 | #endif |
| 616 | } |
| 617 | |
| 618 | void |
| 619 | StringCaseTest::TestTitleOptions() { |
| 620 | // New options in ICU 60. |
| 621 | TestCasingImpl(u"ʻcAt! ʻeTc.", u"ʻCat! ʻetc.", TEST_TITLE, |
| 622 | nullptr, "", U_TITLECASE_WHOLE_STRING); |
| 623 | TestCasingImpl(u"a ʻCaT. A ʻdOg! ʻeTc.", u"A ʻCaT. A ʻdOg! ʻETc.", TEST_TITLE, |
| 624 | nullptr, "", U_TITLECASE_SENTENCES|U_TITLECASE_NO_LOWERCASE); |
| 625 | TestCasingImpl(u"49eRs", u"49ers", TEST_TITLE, |
| 626 | nullptr, "", U_TITLECASE_WHOLE_STRING); |
| 627 | TestCasingImpl(u"«丰(aBc)»", u"«丰(abc)»", TEST_TITLE, |
| 628 | nullptr, "", U_TITLECASE_WHOLE_STRING); |
| 629 | TestCasingImpl(u"49eRs", u"49Ers", TEST_TITLE, |
| 630 | nullptr, "", U_TITLECASE_WHOLE_STRING|U_TITLECASE_ADJUST_TO_CASED); |
| 631 | TestCasingImpl(u"«丰(aBc)»", u"«丰(Abc)»", TEST_TITLE, |
| 632 | nullptr, "", U_TITLECASE_WHOLE_STRING|U_TITLECASE_ADJUST_TO_CASED); |
| 633 | TestCasingImpl(u" john. Smith", u" John. Smith", TEST_TITLE, |
| 634 | nullptr, "", U_TITLECASE_WHOLE_STRING|U_TITLECASE_NO_LOWERCASE); |
| 635 | TestCasingImpl(u" john. Smith", u" john. smith", TEST_TITLE, |
| 636 | nullptr, "", U_TITLECASE_WHOLE_STRING|U_TITLECASE_NO_BREAK_ADJUSTMENT); |
| 637 | TestCasingImpl(u"«ijs»", u"«IJs»", TEST_TITLE, |
| 638 | nullptr, "nl-BE", U_TITLECASE_WHOLE_STRING); |
| 639 | TestCasingImpl(u"«ijs»", u"«İjs»", TEST_TITLE, |
| 640 | nullptr, "tr-DE", U_TITLECASE_WHOLE_STRING); |
| 641 | |
| 642 | #if !UCONFIG_NO_BREAK_ITERATION |
| 643 | // Test conflicting settings. |
| 644 | // If & when we add more options, then the ORed combinations may become |
| 645 | // indistinguishable from valid values. |
| 646 | IcuTestErrorCode errorCode(*this, "TestTitleOptions"); |
| 647 | CaseMap::toTitle("", U_TITLECASE_NO_BREAK_ADJUSTMENT|U_TITLECASE_ADJUST_TO_CASED, nullptr, |
| 648 | u"", 0, nullptr, 0, nullptr, errorCode); |
| 649 | if (errorCode.get() != U_ILLEGAL_ARGUMENT_ERROR) { |
| 650 | errln("CaseMap::toTitle(multiple adjustment options) -> %s not illegal argument", |
| 651 | errorCode.errorName()); |
| 652 | } |
| 653 | errorCode.reset(); |
| 654 | CaseMap::toTitle("", U_TITLECASE_WHOLE_STRING|U_TITLECASE_SENTENCES, nullptr, |
| 655 | u"", 0, nullptr, 0, nullptr, errorCode); |
| 656 | if (errorCode.get() != U_ILLEGAL_ARGUMENT_ERROR) { |
| 657 | errln("CaseMap::toTitle(multiple iterator options) -> %s not illegal argument", |
| 658 | errorCode.errorName()); |
| 659 | } |
| 660 | errorCode.reset(); |
| 661 | LocalPointer<BreakIterator> iter( |
| 662 | BreakIterator::createCharacterInstance(Locale::getRoot(), errorCode)); |
| 663 | CaseMap::toTitle("", U_TITLECASE_WHOLE_STRING, iter.getAlias(), |
| 664 | u"", 0, nullptr, 0, nullptr, errorCode); |
| 665 | if (errorCode.get() != U_ILLEGAL_ARGUMENT_ERROR) { |
| 666 | errln("CaseMap::toTitle(iterator option + iterator) -> %s not illegal argument", |
| 667 | errorCode.errorName()); |
| 668 | } |
| 669 | errorCode.reset(); |
| 670 | #endif |
| 671 | } |
| 672 | |
Frank Tang | d2858cb | 2022-04-08 20:34:12 -0700 | [diff] [blame] | 673 | #if !UCONFIG_NO_BREAK_ITERATION |
| 674 | void StringCaseTest::TestDutchTitle() { |
| 675 | IcuTestErrorCode errorCode(*this, "TestDutchTitle"); |
| 676 | |
| 677 | Locale nl("nl"); // Dutch |
| 678 | LocalPointer<BreakIterator> iter( |
| 679 | BreakIterator::createWordInstance(nl, errorCode)); |
| 680 | |
| 681 | // Dutch titlecase check in English |
| 682 | TestCasingImpl( |
| 683 | u"ijssel igloo IJMUIDEN", |
| 684 | u"Ijssel Igloo Ijmuiden", |
| 685 | TEST_TITLE, |
| 686 | nullptr, |
| 687 | "en", |
| 688 | 0); |
| 689 | |
| 690 | // Dutch titlecase check in Dutch |
| 691 | TestCasingImpl( |
| 692 | u"ijssel igloo IJMUIDEN", |
| 693 | u"IJssel Igloo IJmuiden", |
| 694 | TEST_TITLE, |
| 695 | nullptr, |
| 696 | "nl", |
| 697 | 0); |
| 698 | |
| 699 | // Dutch titlecase check in Dutch with nolowercase option |
| 700 | if (U_SUCCESS(errorCode)) { |
| 701 | iter->setText(u"ijssel igloo IjMUIdEN iPoD ijenough"); |
| 702 | TestCasingImpl( |
| 703 | u"ijssel igloo IjMUIdEN iPoD ijenough", |
| 704 | u"IJssel Igloo IJMUIdEN IPoD IJenough", |
| 705 | TEST_TITLE, |
| 706 | nullptr, |
| 707 | "nl", |
| 708 | U_TITLECASE_NO_LOWERCASE); |
| 709 | } |
| 710 | |
| 711 | errorCode.reset(); |
| 712 | |
| 713 | // Accented IJ testing |
| 714 | |
| 715 | struct dutchTitleTestCase { |
| 716 | const UnicodeString input; |
| 717 | const UnicodeString expectedFull; |
| 718 | const UnicodeString expectedOnlyChanged; |
| 719 | } dutchTitleTestCases[] = { |
| 720 | // input, expectedFull, expectedOnlyChanged |
| 721 | {u"ij", u"IJ", u"IJ"}, |
| 722 | {u"IJ", u"IJ", u""}, |
| 723 | {u"íj́", u"ÍJ́", u"ÍJ"}, |
| 724 | {u"ÍJ́", u"ÍJ́", u""}, |
| 725 | {u"íJ́", u"ÍJ́", u"Í"}, |
| 726 | {u"Ij́", u"Ij́", u""}, |
| 727 | {u"ij́", u"Ij́", u"I"}, |
| 728 | {u"ïj́", u"Ïj́", u"Ï"}, |
| 729 | {u"íj\u0308", u"Íj\u0308", u"Í"}, |
| 730 | {u"íj́\U0001D16E", u"Íj́\U0001D16E", u"Í"}, |
| 731 | {u"íj\u1ABE", u"Íj\u1ABE", u"Í"}, |
| 732 | |
| 733 | {u"ijabc", u"IJabc", u"IJ"}, |
| 734 | {u"IJabc", u"IJabc", u""}, |
| 735 | {u"íj́abc", u"ÍJ́abc", u"ÍJ"}, |
| 736 | {u"ÍJ́abc", u"ÍJ́abc", u""}, |
| 737 | {u"íJ́abc", u"ÍJ́abc", u"Í"}, |
| 738 | {u"Ij́abc", u"Ij́abc", u""}, |
| 739 | {u"ij́abc", u"Ij́abc", u"I"}, |
| 740 | {u"ïj́abc", u"Ïj́abc", u"Ï"}, |
| 741 | {u"íjabc\u0308", u"Íjabc\u0308", u"Í"}, |
| 742 | {u"íj́abc\U0001D16E", u"ÍJ́abc\U0001D16E", u"ÍJ"}, |
| 743 | {u"íjabc\u1ABE", u"Íjabc\u1ABE", u"Í"}, |
| 744 | |
| 745 | // Bug ICU-21919 |
| 746 | {u"Í", u"Í", u""}, |
| 747 | }; |
| 748 | |
| 749 | for (const auto& cas : dutchTitleTestCases) { |
| 750 | const UnicodeString &input = cas.input; |
| 751 | const UnicodeString &expectedFull = cas.expectedFull; |
| 752 | const UnicodeString &expectedOnlyChanged = cas.expectedOnlyChanged; |
| 753 | |
| 754 | for (const auto& isOnlyChanged : {true, false}) { |
| 755 | uint32_t testOptions = U_TITLECASE_NO_LOWERCASE |
| 756 | | (isOnlyChanged ? U_OMIT_UNCHANGED_TEXT : 0); |
| 757 | |
| 758 | const UnicodeString &expected = isOnlyChanged ? expectedOnlyChanged : expectedFull; |
| 759 | |
| 760 | TestCasingImpl( |
| 761 | input, |
| 762 | expected, |
| 763 | TEST_TITLE, |
| 764 | nullptr, |
| 765 | "nl", |
| 766 | testOptions |
| 767 | ); |
| 768 | } |
| 769 | } |
| 770 | } |
| 771 | #endif |
| 772 | |
Frank Tang | 3e05d9d | 2021-11-08 14:04:04 -0800 | [diff] [blame] | 773 | void |
| 774 | StringCaseTest::TestFullCaseFoldingIterator() { |
| 775 | UnicodeString ffi=UNICODE_STRING_SIMPLE("ffi"); |
| 776 | UnicodeString ss=UNICODE_STRING_SIMPLE("ss"); |
| 777 | FullCaseFoldingIterator iter; |
| 778 | int32_t count=0; |
| 779 | int32_t countSpecific=0; |
| 780 | UChar32 c; |
| 781 | UnicodeString full; |
| 782 | while((c=iter.next(full))>=0) { |
| 783 | ++count; |
| 784 | // Check that the full Case_Folding has more than 1 code point. |
| 785 | if(!full.hasMoreChar32Than(0, 0x7fffffff, 1)) { |
| 786 | errln("error: FullCaseFoldingIterator.next()=U+%04lX full Case_Folding has at most 1 code point", (long)c); |
| 787 | continue; |
| 788 | } |
| 789 | // Check that full == Case_Folding(c). |
| 790 | UnicodeString cf(c); |
| 791 | cf.foldCase(); |
| 792 | if(full!=cf) { |
| 793 | errln("error: FullCaseFoldingIterator.next()=U+%04lX full Case_Folding != cf(c)", (long)c); |
| 794 | continue; |
| 795 | } |
| 796 | // Spot-check a couple of specific cases. |
| 797 | if((full==ffi && c==0xfb03) || (full==ss && (c==0xdf || c==0x1e9e))) { |
| 798 | ++countSpecific; |
| 799 | } |
| 800 | } |
| 801 | if(countSpecific!=3) { |
| 802 | errln("error: FullCaseFoldingIterator did not yield exactly the expected specific cases"); |
| 803 | } |
| 804 | if(count<70) { |
| 805 | errln("error: FullCaseFoldingIterator yielded only %d (cp, full) pairs", (int)count); |
| 806 | } |
| 807 | } |
| 808 | |
| 809 | void |
| 810 | StringCaseTest::assertGreekUpper(const char16_t *s, const char16_t *expected) { |
| 811 | UnicodeString s16(s); |
| 812 | UnicodeString expected16(expected); |
| 813 | UnicodeString msg = UnicodeString("UnicodeString::toUpper/Greek(\"") + s16 + "\")"; |
| 814 | UnicodeString result16(s16); |
| 815 | result16.toUpper(GREEK_LOCALE_); |
| 816 | assertEquals(msg, expected16, result16); |
| 817 | |
| 818 | msg = UnicodeString("u_strToUpper/Greek(\"") + s16 + "\") cap="; |
| 819 | int32_t length = expected16.length(); |
| 820 | int32_t capacities[] = { |
| 821 | // Keep in sync with the UTF-8 capacities near the bottom of this function. |
| 822 | 0, length / 2, length - 1, length, length + 1 |
| 823 | }; |
| 824 | for (int32_t i = 0; i < UPRV_LENGTHOF(capacities); ++i) { |
| 825 | int32_t cap = capacities[i]; |
| 826 | UChar *dest16 = result16.getBuffer(expected16.length() + 1); |
| 827 | u_memset(dest16, 0x55AA, result16.getCapacity()); |
| 828 | UErrorCode errorCode = U_ZERO_ERROR; |
| 829 | length = u_strToUpper(dest16, cap, s16.getBuffer(), s16.length(), "el", &errorCode); |
| 830 | assertEquals(msg + cap, expected16.length(), length); |
| 831 | UErrorCode expectedErrorCode; |
| 832 | if (cap < expected16.length()) { |
| 833 | expectedErrorCode = U_BUFFER_OVERFLOW_ERROR; |
| 834 | } else if (cap == expected16.length()) { |
| 835 | expectedErrorCode = U_STRING_NOT_TERMINATED_WARNING; |
| 836 | } else { |
| 837 | expectedErrorCode = U_ZERO_ERROR; |
| 838 | assertEquals(msg + cap + " NUL", 0, dest16[length]); |
| 839 | } |
| 840 | assertEquals(msg + cap + " errorCode", expectedErrorCode, errorCode); |
| 841 | result16.releaseBuffer(length); |
| 842 | if (cap >= expected16.length()) { |
| 843 | assertEquals(msg + cap, expected16, result16); |
| 844 | } |
| 845 | } |
| 846 | |
| 847 | UErrorCode errorCode = U_ZERO_ERROR; |
| 848 | LocalUCaseMapPointer csm(ucasemap_open("el", 0, &errorCode)); |
| 849 | assertSuccess("ucasemap_open", errorCode); |
| 850 | std::string s8; |
| 851 | s16.toUTF8String(s8); |
| 852 | msg = UnicodeString("ucasemap_utf8ToUpper/Greek(\"") + s16 + "\")"; |
| 853 | char dest8[1000]; |
| 854 | length = ucasemap_utf8ToUpper(csm.getAlias(), dest8, UPRV_LENGTHOF(dest8), |
| 855 | s8.data(), static_cast<int32_t>(s8.length()), &errorCode); |
| 856 | assertSuccess("ucasemap_utf8ToUpper", errorCode); |
| 857 | StringPiece result8(dest8, length); |
| 858 | UnicodeString result16From8 = UnicodeString::fromUTF8(result8); |
| 859 | assertEquals(msg, expected16, result16From8); |
| 860 | |
| 861 | msg += " cap="; |
| 862 | capacities[1] = length / 2; |
| 863 | capacities[2] = length - 1; |
| 864 | capacities[3] = length; |
| 865 | capacities[4] = length + 1; |
| 866 | char dest8b[1000]; |
| 867 | int32_t expected8Length = length; // Assuming the previous call worked. |
| 868 | for (int32_t i = 0; i < UPRV_LENGTHOF(capacities); ++i) { |
| 869 | int32_t cap = capacities[i]; |
| 870 | memset(dest8b, 0x5A, UPRV_LENGTHOF(dest8b)); |
| 871 | UErrorCode errorCode = U_ZERO_ERROR; |
| 872 | length = ucasemap_utf8ToUpper(csm.getAlias(), dest8b, cap, |
| 873 | s8.data(), static_cast<int32_t>(s8.length()), &errorCode); |
| 874 | assertEquals(msg + cap, expected8Length, length); |
| 875 | UErrorCode expectedErrorCode; |
| 876 | if (cap < expected8Length) { |
| 877 | expectedErrorCode = U_BUFFER_OVERFLOW_ERROR; |
| 878 | } else if (cap == expected8Length) { |
| 879 | expectedErrorCode = U_STRING_NOT_TERMINATED_WARNING; |
| 880 | } else { |
| 881 | expectedErrorCode = U_ZERO_ERROR; |
| 882 | // Casts to int32_t to avoid matching UBool. |
| 883 | assertEquals(msg + cap + " NUL", (int32_t)0, (int32_t)dest8b[length]); |
| 884 | } |
| 885 | assertEquals(msg + cap + " errorCode", expectedErrorCode, errorCode); |
| 886 | if (cap >= expected8Length) { |
| 887 | assertEquals(msg + cap + " (memcmp)", 0, memcmp(dest8, dest8b, expected8Length)); |
| 888 | } |
| 889 | } |
| 890 | } |
| 891 | |
| 892 | void |
| 893 | StringCaseTest::TestGreekUpper() { |
| 894 | // https://unicode-org.atlassian.net/browse/ICU-5456 |
| 895 | assertGreekUpper(u"άδικος, κείμενο, ίριδα", u"ΑΔΙΚΟΣ, ΚΕΙΜΕΝΟ, ΙΡΙΔΑ"); |
| 896 | // https://bugzilla.mozilla.org/show_bug.cgi?id=307039 |
| 897 | // https://bug307039.bmoattachments.org/attachment.cgi?id=194893 |
| 898 | assertGreekUpper(u"Πατάτα", u"ΠΑΤΑΤΑ"); |
| 899 | assertGreekUpper(u"Αέρας, Μυστήριο, Ωραίο", u"ΑΕΡΑΣ, ΜΥΣΤΗΡΙΟ, ΩΡΑΙΟ"); |
| 900 | assertGreekUpper(u"Μαΐου, Πόρος, Ρύθμιση", u"ΜΑΪΟΥ, ΠΟΡΟΣ, ΡΥΘΜΙΣΗ"); |
| 901 | assertGreekUpper(u"ΰ, Τηρώ, Μάιος", u"Ϋ, ΤΗΡΩ, ΜΑΪΟΣ"); |
| 902 | assertGreekUpper(u"άυλος", u"ΑΫΛΟΣ"); |
| 903 | assertGreekUpper(u"ΑΫΛΟΣ", u"ΑΫΛΟΣ"); |
| 904 | assertGreekUpper(u"Άκλιτα ρήματα ή άκλιτες μετοχές", u"ΑΚΛΙΤΑ ΡΗΜΑΤΑ Ή ΑΚΛΙΤΕΣ ΜΕΤΟΧΕΣ"); |
| 905 | // http://www.unicode.org/udhr/d/udhr_ell_monotonic.html |
| 906 | assertGreekUpper(u"Επειδή η αναγνώριση της αξιοπρέπειας", u"ΕΠΕΙΔΗ Η ΑΝΑΓΝΩΡΙΣΗ ΤΗΣ ΑΞΙΟΠΡΕΠΕΙΑΣ"); |
| 907 | assertGreekUpper(u"νομικού ή διεθνούς", u"ΝΟΜΙΚΟΥ Ή ΔΙΕΘΝΟΥΣ"); |
| 908 | // http://unicode.org/udhr/d/udhr_ell_polytonic.html |
| 909 | assertGreekUpper(u"Ἐπειδὴ ἡ ἀναγνώριση", u"ΕΠΕΙΔΗ Η ΑΝΑΓΝΩΡΙΣΗ"); |
| 910 | assertGreekUpper(u"νομικοῦ ἢ διεθνοῦς", u"ΝΟΜΙΚΟΥ Ή ΔΙΕΘΝΟΥΣ"); |
| 911 | // From Google bug report |
| 912 | assertGreekUpper(u"Νέο, Δημιουργία", u"ΝΕΟ, ΔΗΜΙΟΥΡΓΙΑ"); |
| 913 | // http://crbug.com/234797 |
| 914 | assertGreekUpper(u"Ελάτε να φάτε τα καλύτερα παϊδάκια!", u"ΕΛΑΤΕ ΝΑ ΦΑΤΕ ΤΑ ΚΑΛΥΤΕΡΑ ΠΑΪΔΑΚΙΑ!"); |
| 915 | assertGreekUpper(u"Μαΐου, τρόλεϊ", u"ΜΑΪΟΥ, ΤΡΟΛΕΪ"); |
| 916 | assertGreekUpper(u"Το ένα ή το άλλο.", u"ΤΟ ΕΝΑ Ή ΤΟ ΑΛΛΟ."); |
| 917 | // http://multilingualtypesetting.co.uk/blog/greek-typesetting-tips/ |
| 918 | assertGreekUpper(u"ρωμέικα", u"ΡΩΜΕΪΚΑ"); |
| 919 | assertGreekUpper(u"ή.", u"Ή."); |
| 920 | } |
| 921 | |
| 922 | void StringCaseTest::TestArmenian() { |
| 923 | Locale hy("hy"); // Eastern Armenian |
| 924 | Locale hyw("hyw"); // Western Armenian |
| 925 | Locale root = Locale::getRoot(); |
| 926 | // See ICU-13416: |
| 927 | // և ligature ech-yiwn |
| 928 | // uppercases to ԵՒ=ech+yiwn by default and in Western Armenian, |
| 929 | // but to ԵՎ=ech+vew in Eastern Armenian. |
| 930 | UnicodeString s(u"և Երևանի"); |
| 931 | |
| 932 | assertEquals("upper root", u"ԵՒ ԵՐԵՒԱՆԻ", UnicodeString(s).toUpper(root)); |
| 933 | assertEquals("upper hy", u"ԵՎ ԵՐԵՎԱՆԻ", UnicodeString(s).toUpper(hy)); |
| 934 | assertEquals("upper hyw", u"ԵՒ ԵՐԵՒԱՆԻ", UnicodeString(s).toUpper(hyw)); |
| 935 | #if !UCONFIG_NO_BREAK_ITERATION |
| 936 | assertEquals("title root", u"Եւ Երևանի", UnicodeString(s).toTitle(nullptr, root)); |
| 937 | assertEquals("title hy", u"Եվ Երևանի", UnicodeString(s).toTitle(nullptr, hy)); |
| 938 | assertEquals("title hyw", u"Եւ Երևանի", UnicodeString(s).toTitle(nullptr, hyw)); |
| 939 | #endif |
| 940 | } |
| 941 | |
| 942 | void |
| 943 | StringCaseTest::TestLongUpper() { |
| 944 | if (quick) { |
| 945 | logln("not exhaustive mode: skipping this test"); |
| 946 | return; |
| 947 | } |
| 948 | // Ticket #12663, crash with an extremely long string where |
| 949 | // U+0390 maps to 0399 0308 0301 so that the result is three times as long |
| 950 | // and overflows an int32_t. |
| 951 | int32_t length = 0x40000004; // more than 1G UChars |
| 952 | UnicodeString s(length, (UChar32)0x390, length); |
| 953 | UnicodeString result; |
| 954 | UChar *dest = result.getBuffer(length + 1); |
| 955 | if (s.isBogus() || dest == NULL) { |
| 956 | logln("Out of memory, unable to run this test on this machine."); |
| 957 | return; |
| 958 | } |
| 959 | IcuTestErrorCode errorCode(*this, "TestLongUpper"); |
| 960 | int32_t destLength = u_strToUpper(dest, result.getCapacity(), |
| 961 | s.getBuffer(), s.length(), "", errorCode); |
| 962 | result.releaseBuffer(destLength); |
| 963 | if (errorCode.reset() != U_INDEX_OUTOFBOUNDS_ERROR) { |
| 964 | errln("expected U_INDEX_OUTOFBOUNDS_ERROR, got %s (destLength is undefined, got %ld)", |
| 965 | errorCode.errorName(), (long)destLength); |
| 966 | } |
| 967 | } |
| 968 | |
| 969 | void StringCaseTest::TestMalformedUTF8() { |
| 970 | // ticket #12639 |
| 971 | IcuTestErrorCode errorCode(*this, "TestMalformedUTF8"); |
| 972 | LocalUCaseMapPointer csm(ucasemap_open("en", U_TITLECASE_NO_BREAK_ADJUSTMENT, errorCode)); |
| 973 | if (errorCode.isFailure()) { |
| 974 | errln("ucasemap_open(English) failed - %s", errorCode.errorName()); |
| 975 | return; |
| 976 | } |
| 977 | char src[1] = { (char)0x85 }; // malformed UTF-8 |
| 978 | char dest[3] = { 0, 0, 0 }; |
| 979 | int32_t destLength; |
| 980 | #if !UCONFIG_NO_BREAK_ITERATION |
| 981 | destLength = ucasemap_utf8ToTitle(csm.getAlias(), dest, 3, src, 1, errorCode); |
| 982 | if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) { |
| 983 | errln("ucasemap_utf8ToTitle(\\x85) failed: %s destLength=%d dest[0]=0x%02x", |
| 984 | errorCode.errorName(), (int)destLength, dest[0]); |
| 985 | } |
| 986 | #endif |
| 987 | |
| 988 | errorCode.reset(); |
| 989 | dest[0] = 0; |
| 990 | destLength = ucasemap_utf8ToLower(csm.getAlias(), dest, 3, src, 1, errorCode); |
| 991 | if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) { |
| 992 | errln("ucasemap_utf8ToLower(\\x85) failed: %s destLength=%d dest[0]=0x%02x", |
| 993 | errorCode.errorName(), (int)destLength, dest[0]); |
| 994 | } |
| 995 | |
| 996 | errorCode.reset(); |
| 997 | dest[0] = 0; |
| 998 | destLength = ucasemap_utf8ToUpper(csm.getAlias(), dest, 3, src, 1, errorCode); |
| 999 | if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) { |
| 1000 | errln("ucasemap_utf8ToUpper(\\x85) failed: %s destLength=%d dest[0]=0x%02x", |
| 1001 | errorCode.errorName(), (int)destLength, dest[0]); |
| 1002 | } |
| 1003 | |
| 1004 | errorCode.reset(); |
| 1005 | dest[0] = 0; |
| 1006 | destLength = ucasemap_utf8FoldCase(csm.getAlias(), dest, 3, src, 1, errorCode); |
| 1007 | if (errorCode.isFailure() || destLength != 1 || dest[0] != src[0]) { |
| 1008 | errln("ucasemap_utf8FoldCase(\\x85) failed: %s destLength=%d dest[0]=0x%02x", |
| 1009 | errorCode.errorName(), (int)destLength, dest[0]); |
| 1010 | } |
| 1011 | } |
| 1012 | |
| 1013 | void StringCaseTest::TestBufferOverflow() { |
| 1014 | // Ticket #12849, incorrect result from Title Case preflight operation, |
| 1015 | // when buffer overflow error is expected. |
| 1016 | IcuTestErrorCode errorCode(*this, "TestBufferOverflow"); |
| 1017 | LocalUCaseMapPointer csm(ucasemap_open("en", 0, errorCode)); |
| 1018 | if (errorCode.isFailure()) { |
| 1019 | errln("ucasemap_open(English) failed - %s", errorCode.errorName()); |
| 1020 | return; |
| 1021 | } |
| 1022 | |
| 1023 | UnicodeString data("hello world"); |
| 1024 | int32_t result; |
| 1025 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1026 | result = ucasemap_toTitle(csm.getAlias(), NULL, 0, data.getBuffer(), data.length(), errorCode); |
| 1027 | if (errorCode.get() != U_BUFFER_OVERFLOW_ERROR || result != data.length()) { |
| 1028 | errln("%s:%d ucasemap_toTitle(\"hello world\") failed: " |
| 1029 | "expected (U_BUFFER_OVERFLOW_ERROR, %d), got (%s, %d)", |
| 1030 | __FILE__, __LINE__, data.length(), errorCode.errorName(), result); |
| 1031 | } |
| 1032 | #endif |
| 1033 | errorCode.reset(); |
| 1034 | |
| 1035 | std::string data_utf8; |
| 1036 | data.toUTF8String(data_utf8); |
| 1037 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1038 | result = ucasemap_utf8ToTitle(csm.getAlias(), NULL, 0, data_utf8.c_str(), static_cast<int32_t>(data_utf8.length()), errorCode); |
| 1039 | if (errorCode.get() != U_BUFFER_OVERFLOW_ERROR || result != (int32_t)data_utf8.length()) { |
| 1040 | errln("%s:%d ucasemap_toTitle(\"hello world\") failed: " |
| 1041 | "expected (U_BUFFER_OVERFLOW_ERROR, %d), got (%s, %d)", |
| 1042 | __FILE__, __LINE__, data_utf8.length(), errorCode.errorName(), result); |
| 1043 | } |
| 1044 | #endif |
| 1045 | errorCode.reset(); |
| 1046 | } |
| 1047 | |
| 1048 | void StringCaseTest::TestEdits() { |
| 1049 | IcuTestErrorCode errorCode(*this, "TestEdits"); |
| 1050 | Edits edits; |
| 1051 | assertFalse("new Edits hasChanges", edits.hasChanges()); |
| 1052 | assertEquals("new Edits numberOfChanges", 0, edits.numberOfChanges()); |
| 1053 | assertEquals("new Edits", 0, edits.lengthDelta()); |
| 1054 | edits.addUnchanged(1); // multiple unchanged ranges are combined |
| 1055 | edits.addUnchanged(10000); // too long, and they are split |
| 1056 | edits.addReplace(0, 0); |
| 1057 | edits.addUnchanged(2); |
| 1058 | assertFalse("unchanged 10003 hasChanges", edits.hasChanges()); |
| 1059 | assertEquals("unchanged 10003 numberOfChanges", 0, edits.numberOfChanges()); |
| 1060 | assertEquals("unchanged 10003", 0, edits.lengthDelta()); |
| 1061 | edits.addReplace(2, 1); // multiple short equal-lengths edits are compressed |
| 1062 | edits.addUnchanged(0); |
| 1063 | edits.addReplace(2, 1); |
| 1064 | edits.addReplace(2, 1); |
| 1065 | edits.addReplace(0, 10); |
| 1066 | edits.addReplace(100, 0); |
| 1067 | edits.addReplace(3000, 4000); // variable-length encoding |
| 1068 | edits.addReplace(100000, 100000); |
| 1069 | assertTrue("some edits hasChanges", edits.hasChanges()); |
| 1070 | assertEquals("some edits numberOfChanges", 7, edits.numberOfChanges()); |
| 1071 | assertEquals("some edits", -3 + 10 - 100 + 1000, edits.lengthDelta()); |
| 1072 | UErrorCode outErrorCode = U_ZERO_ERROR; |
| 1073 | assertFalse("edits done: copyErrorTo", edits.copyErrorTo(outErrorCode)); |
| 1074 | |
| 1075 | static const EditChange coarseExpectedChanges[] = { |
| 1076 | { FALSE, 10003, 10003 }, |
| 1077 | { TRUE, 103106, 104013 } |
| 1078 | }; |
| 1079 | TestUtility::checkEditsIter(*this, u"coarse", |
| 1080 | edits.getCoarseIterator(), edits.getCoarseIterator(), |
| 1081 | coarseExpectedChanges, UPRV_LENGTHOF(coarseExpectedChanges), TRUE, errorCode); |
| 1082 | TestUtility::checkEditsIter(*this, u"coarse changes", |
| 1083 | edits.getCoarseChangesIterator(), edits.getCoarseChangesIterator(), |
| 1084 | coarseExpectedChanges, UPRV_LENGTHOF(coarseExpectedChanges), FALSE, errorCode); |
| 1085 | |
| 1086 | static const EditChange fineExpectedChanges[] = { |
| 1087 | { FALSE, 10003, 10003 }, |
| 1088 | { TRUE, 2, 1 }, |
| 1089 | { TRUE, 2, 1 }, |
| 1090 | { TRUE, 2, 1 }, |
| 1091 | { TRUE, 0, 10 }, |
| 1092 | { TRUE, 100, 0 }, |
| 1093 | { TRUE, 3000, 4000 }, |
| 1094 | { TRUE, 100000, 100000 } |
| 1095 | }; |
| 1096 | TestUtility::checkEditsIter(*this, u"fine", |
| 1097 | edits.getFineIterator(), edits.getFineIterator(), |
| 1098 | fineExpectedChanges, UPRV_LENGTHOF(fineExpectedChanges), TRUE, errorCode); |
| 1099 | TestUtility::checkEditsIter(*this, u"fine changes", |
| 1100 | edits.getFineChangesIterator(), edits.getFineChangesIterator(), |
| 1101 | fineExpectedChanges, UPRV_LENGTHOF(fineExpectedChanges), FALSE, errorCode); |
| 1102 | |
| 1103 | edits.reset(); |
| 1104 | assertFalse("reset hasChanges", edits.hasChanges()); |
| 1105 | assertEquals("reset numberOfChanges", 0, edits.numberOfChanges()); |
| 1106 | assertEquals("reset", 0, edits.lengthDelta()); |
| 1107 | Edits::Iterator ei = edits.getCoarseChangesIterator(); |
| 1108 | assertFalse("reset then iterator", ei.next(errorCode)); |
| 1109 | } |
| 1110 | |
| 1111 | void StringCaseTest::TestCopyMoveEdits() { |
| 1112 | IcuTestErrorCode errorCode(*this, "TestCopyMoveEdits"); |
| 1113 | // Exceed the stack array capacity. |
| 1114 | Edits a; |
| 1115 | for (int32_t i = 0; i < 250; ++i) { |
| 1116 | a.addReplace(i % 10, (i % 10) + 1); |
| 1117 | } |
| 1118 | assertEquals("a: many edits, length delta", 250, a.lengthDelta()); |
| 1119 | |
| 1120 | // copy |
| 1121 | Edits b(a); |
| 1122 | assertEquals("b: copy of many edits, length delta", 250, b.lengthDelta()); |
| 1123 | assertEquals("a remains: many edits, length delta", 250, a.lengthDelta()); |
| 1124 | TestUtility::checkEqualEdits(*this, u"b copy of a", a, b, errorCode); |
| 1125 | |
| 1126 | // assign |
| 1127 | Edits c; |
| 1128 | c.addUnchanged(99); |
| 1129 | c.addReplace(88, 77); |
| 1130 | c = b; |
| 1131 | assertEquals("c: assigned many edits, length delta", 250, c.lengthDelta()); |
| 1132 | assertEquals("b remains: many edits, length delta", 250, b.lengthDelta()); |
| 1133 | TestUtility::checkEqualEdits(*this, u"c = b", b, c, errorCode); |
| 1134 | |
| 1135 | // std::move trouble on these platforms. |
| 1136 | // See https://unicode-org.atlassian.net/browse/ICU-13393 |
| 1137 | #if !(U_PLATFORM == U_PF_AIX || U_PLATFORM == U_PF_OS390) |
| 1138 | // move constructor empties object with heap array |
| 1139 | Edits d(std::move(a)); |
| 1140 | assertEquals("d: move-constructed many edits, length delta", 250, d.lengthDelta()); |
| 1141 | assertFalse("a moved away: no more hasChanges", a.hasChanges()); |
| 1142 | TestUtility::checkEqualEdits(*this, u"d() <- a", d, b, errorCode); |
| 1143 | Edits empty; |
| 1144 | TestUtility::checkEqualEdits(*this, u"a moved away", empty, a, errorCode); |
| 1145 | |
| 1146 | // move assignment empties object with heap array |
| 1147 | Edits e; |
| 1148 | e.addReplace(0, 1000); |
| 1149 | e = std::move(b); |
| 1150 | assertEquals("e: move-assigned many edits, length delta", 250, e.lengthDelta()); |
| 1151 | assertFalse("b moved away: no more hasChanges", b.hasChanges()); |
| 1152 | TestUtility::checkEqualEdits(*this, u"e <- b", e, c, errorCode); |
| 1153 | TestUtility::checkEqualEdits(*this, u"b moved away", empty, b, errorCode); |
| 1154 | |
| 1155 | // Edits::Iterator default constructor. |
| 1156 | Edits::Iterator iter; |
| 1157 | assertFalse("Edits::Iterator().next()", iter.next(errorCode)); |
| 1158 | assertSuccess("Edits::Iterator().next()", errorCode); |
| 1159 | iter = e.getFineChangesIterator(); |
| 1160 | assertTrue("iter.next()", iter.next(errorCode)); |
| 1161 | assertSuccess("iter.next()", errorCode); |
| 1162 | assertTrue("iter.hasChange()", iter.hasChange()); |
| 1163 | assertEquals("iter.newLength()", 1, iter.newLength()); |
| 1164 | #endif |
| 1165 | } |
| 1166 | |
| 1167 | void StringCaseTest::TestEditsFindFwdBwd() { |
| 1168 | IcuTestErrorCode errorCode(*this, "TestEditsFindFwdBwd"); |
| 1169 | // Some users need index mappings to be efficient when they are out of order. |
| 1170 | // The most interesting failure case for this test is it taking a very long time. |
| 1171 | Edits e; |
| 1172 | constexpr int32_t N = 200000; |
| 1173 | for (int32_t i = 0; i < N; ++i) { |
| 1174 | e.addUnchanged(1); |
| 1175 | e.addReplace(3, 1); |
| 1176 | } |
| 1177 | Edits::Iterator iter = e.getFineIterator(); |
| 1178 | for (int32_t i = 0; i <= N; i += 2) { |
| 1179 | assertEquals("ascending", i * 2, iter.sourceIndexFromDestinationIndex(i, errorCode)); |
| 1180 | assertEquals("ascending", i * 2 + 1, iter.sourceIndexFromDestinationIndex(i + 1, errorCode)); |
| 1181 | } |
| 1182 | for (int32_t i = N; i >= 0; i -= 2) { |
| 1183 | assertEquals("descending", i * 2 + 1, iter.sourceIndexFromDestinationIndex(i + 1, errorCode)); |
| 1184 | assertEquals("descending", i * 2, iter.sourceIndexFromDestinationIndex(i, errorCode)); |
| 1185 | } |
| 1186 | } |
| 1187 | |
| 1188 | void StringCaseTest::TestMergeEdits() { |
| 1189 | // For debugging, set -v to see matching edits up to a failure. |
| 1190 | IcuTestErrorCode errorCode(*this, "TestMergeEdits"); |
| 1191 | Edits ab, bc, ac, expected_ac; |
| 1192 | |
| 1193 | // Simple: Two parallel non-changes. |
| 1194 | ab.addUnchanged(2); |
| 1195 | bc.addUnchanged(2); |
| 1196 | expected_ac.addUnchanged(2); |
| 1197 | |
| 1198 | // Simple: Two aligned changes. |
| 1199 | ab.addReplace(3, 2); |
| 1200 | bc.addReplace(2, 1); |
| 1201 | expected_ac.addReplace(3, 1); |
| 1202 | |
| 1203 | // Unequal non-changes. |
| 1204 | ab.addUnchanged(5); |
| 1205 | bc.addUnchanged(3); |
| 1206 | expected_ac.addUnchanged(3); |
| 1207 | // ab ahead by 2 |
| 1208 | |
| 1209 | // Overlapping changes accumulate until they share a boundary. |
| 1210 | ab.addReplace(4, 3); |
| 1211 | bc.addReplace(3, 2); |
| 1212 | ab.addReplace(4, 3); |
| 1213 | bc.addReplace(3, 2); |
| 1214 | ab.addReplace(4, 3); |
| 1215 | bc.addReplace(3, 2); |
| 1216 | bc.addUnchanged(4); |
| 1217 | expected_ac.addReplace(14, 8); |
| 1218 | // bc ahead by 2 |
| 1219 | |
| 1220 | // Balance out intermediate-string lengths. |
| 1221 | ab.addUnchanged(2); |
| 1222 | expected_ac.addUnchanged(2); |
| 1223 | |
| 1224 | // Insert something and delete it: Should disappear. |
| 1225 | ab.addReplace(0, 5); |
| 1226 | ab.addReplace(0, 2); |
| 1227 | bc.addReplace(7, 0); |
| 1228 | |
| 1229 | // Parallel change to make a new boundary. |
| 1230 | ab.addReplace(1, 2); |
| 1231 | bc.addReplace(2, 3); |
| 1232 | expected_ac.addReplace(1, 3); |
| 1233 | |
| 1234 | // Multiple ab deletions should remain separate at the boundary. |
| 1235 | ab.addReplace(1, 0); |
| 1236 | ab.addReplace(2, 0); |
| 1237 | ab.addReplace(3, 0); |
| 1238 | expected_ac.addReplace(1, 0); |
| 1239 | expected_ac.addReplace(2, 0); |
| 1240 | expected_ac.addReplace(3, 0); |
| 1241 | |
| 1242 | // Unequal non-changes can be split for another boundary. |
| 1243 | ab.addUnchanged(2); |
| 1244 | bc.addUnchanged(1); |
| 1245 | expected_ac.addUnchanged(1); |
| 1246 | // ab ahead by 1 |
| 1247 | |
| 1248 | // Multiple bc insertions should create a boundary and remain separate. |
| 1249 | bc.addReplace(0, 4); |
| 1250 | bc.addReplace(0, 5); |
| 1251 | bc.addReplace(0, 6); |
| 1252 | expected_ac.addReplace(0, 4); |
| 1253 | expected_ac.addReplace(0, 5); |
| 1254 | expected_ac.addReplace(0, 6); |
| 1255 | // ab ahead by 1 |
| 1256 | |
| 1257 | // Multiple ab deletions in the middle of a bc change are merged. |
| 1258 | bc.addReplace(2, 2); |
| 1259 | // bc ahead by 1 |
| 1260 | ab.addReplace(1, 0); |
| 1261 | ab.addReplace(2, 0); |
| 1262 | ab.addReplace(3, 0); |
| 1263 | ab.addReplace(4, 1); |
| 1264 | expected_ac.addReplace(11, 2); |
| 1265 | |
| 1266 | // Multiple bc insertions in the middle of an ab change are merged. |
| 1267 | ab.addReplace(5, 6); |
| 1268 | bc.addReplace(3, 3); |
| 1269 | // ab ahead by 3 |
| 1270 | bc.addReplace(0, 4); |
| 1271 | bc.addReplace(0, 5); |
| 1272 | bc.addReplace(0, 6); |
| 1273 | bc.addReplace(3, 7); |
| 1274 | expected_ac.addReplace(5, 25); |
| 1275 | |
| 1276 | // Delete around a deletion. |
| 1277 | ab.addReplace(4, 4); |
| 1278 | ab.addReplace(3, 0); |
| 1279 | ab.addUnchanged(2); |
| 1280 | bc.addReplace(2, 2); |
| 1281 | bc.addReplace(4, 0); |
| 1282 | expected_ac.addReplace(9, 2); |
| 1283 | |
| 1284 | // Insert into an insertion. |
| 1285 | ab.addReplace(0, 2); |
| 1286 | bc.addReplace(1, 1); |
| 1287 | bc.addReplace(0, 8); |
| 1288 | bc.addUnchanged(4); |
| 1289 | expected_ac.addReplace(0, 10); |
| 1290 | // bc ahead by 3 |
| 1291 | |
| 1292 | // Balance out intermediate-string lengths. |
| 1293 | ab.addUnchanged(3); |
| 1294 | expected_ac.addUnchanged(3); |
| 1295 | |
| 1296 | // Deletions meet insertions. |
| 1297 | // Output order is arbitrary in principle, but we expect insertions first |
| 1298 | // and want to keep it that way. |
| 1299 | ab.addReplace(2, 0); |
| 1300 | ab.addReplace(4, 0); |
| 1301 | ab.addReplace(6, 0); |
| 1302 | bc.addReplace(0, 1); |
| 1303 | bc.addReplace(0, 3); |
| 1304 | bc.addReplace(0, 5); |
| 1305 | expected_ac.addReplace(0, 1); |
| 1306 | expected_ac.addReplace(0, 3); |
| 1307 | expected_ac.addReplace(0, 5); |
| 1308 | expected_ac.addReplace(2, 0); |
| 1309 | expected_ac.addReplace(4, 0); |
| 1310 | expected_ac.addReplace(6, 0); |
| 1311 | |
| 1312 | // End with a non-change, so that further edits are never reordered. |
| 1313 | ab.addUnchanged(1); |
| 1314 | bc.addUnchanged(1); |
| 1315 | expected_ac.addUnchanged(1); |
| 1316 | |
| 1317 | ac.mergeAndAppend(ab, bc, errorCode); |
| 1318 | assertSuccess("ab+bc", errorCode); |
| 1319 | if (!TestUtility::checkEqualEdits(*this, u"ab+bc", expected_ac, ac, errorCode)) { |
| 1320 | return; |
| 1321 | } |
| 1322 | |
| 1323 | // Append more Edits. |
| 1324 | Edits ab2, bc2; |
| 1325 | ab2.addUnchanged(5); |
| 1326 | bc2.addReplace(1, 2); |
| 1327 | bc2.addUnchanged(4); |
| 1328 | expected_ac.addReplace(1, 2); |
| 1329 | expected_ac.addUnchanged(4); |
| 1330 | ac.mergeAndAppend(ab2, bc2, errorCode); |
| 1331 | assertSuccess("ab2+bc2", errorCode); |
| 1332 | if (!TestUtility::checkEqualEdits(*this, u"ab2+bc2", expected_ac, ac, errorCode)) { |
| 1333 | return; |
| 1334 | } |
| 1335 | |
| 1336 | // Append empty edits. |
| 1337 | Edits empty; |
| 1338 | ac.mergeAndAppend(empty, empty, errorCode); |
| 1339 | assertSuccess("empty+empty", errorCode); |
| 1340 | if (!TestUtility::checkEqualEdits(*this, u"empty+empty", expected_ac, ac, errorCode)) { |
| 1341 | return; |
| 1342 | } |
| 1343 | |
| 1344 | // Error: Append more edits with mismatched intermediate-string lengths. |
| 1345 | Edits mismatch; |
| 1346 | mismatch.addReplace(1, 1); |
| 1347 | ac.mergeAndAppend(ab2, mismatch, errorCode); |
| 1348 | assertEquals("ab2+mismatch", U_ILLEGAL_ARGUMENT_ERROR, errorCode.get()); |
| 1349 | errorCode.reset(); |
| 1350 | ac.mergeAndAppend(mismatch, bc2, errorCode); |
| 1351 | assertEquals("mismatch+bc2", U_ILLEGAL_ARGUMENT_ERROR, errorCode.get()); |
| 1352 | errorCode.reset(); |
| 1353 | } |
| 1354 | |
| 1355 | void StringCaseTest::TestCaseMapWithEdits() { |
| 1356 | IcuTestErrorCode errorCode(*this, "TestCaseMapWithEdits"); |
| 1357 | UChar dest[20]; |
| 1358 | Edits edits; |
| 1359 | |
| 1360 | int32_t length = CaseMap::toLower("tr", U_OMIT_UNCHANGED_TEXT, |
| 1361 | u"IstanBul", 8, dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1362 | assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıb"), UnicodeString(TRUE, dest, length)); |
| 1363 | static const EditChange lowerExpectedChanges[] = { |
| 1364 | { TRUE, 1, 1 }, |
| 1365 | { FALSE, 4, 4 }, |
| 1366 | { TRUE, 1, 1 }, |
| 1367 | { FALSE, 2, 2 } |
| 1368 | }; |
| 1369 | TestUtility::checkEditsIter(*this, u"toLower(IstanBul)", |
| 1370 | edits.getFineIterator(), edits.getFineIterator(), |
| 1371 | lowerExpectedChanges, UPRV_LENGTHOF(lowerExpectedChanges), |
| 1372 | TRUE, errorCode); |
| 1373 | |
| 1374 | edits.reset(); |
| 1375 | length = CaseMap::toUpper("el", U_OMIT_UNCHANGED_TEXT, |
| 1376 | u"Πατάτα", 6, dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1377 | assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΑΤΑΤΑ"), UnicodeString(TRUE, dest, length)); |
| 1378 | static const EditChange upperExpectedChanges[] = { |
| 1379 | { FALSE, 1, 1 }, |
| 1380 | { TRUE, 1, 1 }, |
| 1381 | { TRUE, 1, 1 }, |
| 1382 | { TRUE, 1, 1 }, |
| 1383 | { TRUE, 1, 1 }, |
| 1384 | { TRUE, 1, 1 } |
| 1385 | }; |
| 1386 | TestUtility::checkEditsIter(*this, u"toUpper(Πατάτα)", |
| 1387 | edits.getFineIterator(), edits.getFineIterator(), |
| 1388 | upperExpectedChanges, UPRV_LENGTHOF(upperExpectedChanges), |
| 1389 | TRUE, errorCode); |
| 1390 | |
| 1391 | edits.reset(); |
| 1392 | |
| 1393 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1394 | length = CaseMap::toTitle("nl", |
| 1395 | U_OMIT_UNCHANGED_TEXT | |
| 1396 | U_TITLECASE_NO_BREAK_ADJUSTMENT | |
| 1397 | U_TITLECASE_NO_LOWERCASE, |
| 1398 | nullptr, u"IjssEL IglOo", 12, |
| 1399 | dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1400 | assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"J"), UnicodeString(TRUE, dest, length)); |
| 1401 | static const EditChange titleExpectedChanges[] = { |
| 1402 | { FALSE, 1, 1 }, |
| 1403 | { TRUE, 1, 1 }, |
| 1404 | { FALSE, 10, 10 } |
| 1405 | }; |
| 1406 | TestUtility::checkEditsIter(*this, u"toTitle(IjssEL IglOo)", |
| 1407 | edits.getFineIterator(), edits.getFineIterator(), |
| 1408 | titleExpectedChanges, UPRV_LENGTHOF(titleExpectedChanges), |
| 1409 | TRUE, errorCode); |
| 1410 | #endif |
| 1411 | |
| 1412 | // No explicit nor automatic edits.reset(). Edits should be appended. |
| 1413 | length = CaseMap::fold(U_OMIT_UNCHANGED_TEXT | U_EDITS_NO_RESET | U_FOLD_CASE_EXCLUDE_SPECIAL_I, |
| 1414 | u"IßtanBul", 8, dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1415 | assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ıssb"), UnicodeString(TRUE, dest, length)); |
| 1416 | static const EditChange foldExpectedChanges[] = { |
| 1417 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1418 | // From titlecasing. |
| 1419 | { FALSE, 1, 1 }, |
| 1420 | { TRUE, 1, 1 }, |
| 1421 | { FALSE, 10, 10 }, |
| 1422 | #endif |
| 1423 | // From case folding. |
| 1424 | { TRUE, 1, 1 }, |
| 1425 | { TRUE, 1, 2 }, |
| 1426 | { FALSE, 3, 3 }, |
| 1427 | { TRUE, 1, 1 }, |
| 1428 | { FALSE, 2, 2 } |
| 1429 | }; |
| 1430 | TestUtility::checkEditsIter(*this, u"foldCase(no Edits reset, IßtanBul)", |
| 1431 | edits.getFineIterator(), edits.getFineIterator(), |
| 1432 | foldExpectedChanges, UPRV_LENGTHOF(foldExpectedChanges), |
| 1433 | TRUE, errorCode); |
| 1434 | } |
| 1435 | |
| 1436 | void StringCaseTest::TestCaseMapUTF8WithEdits() { |
| 1437 | IcuTestErrorCode errorCode(*this, "TestCaseMapUTF8WithEdits"); |
| 1438 | char dest[50]; |
| 1439 | Edits edits; |
| 1440 | |
| 1441 | int32_t length = CaseMap::utf8ToLower("tr", U_OMIT_UNCHANGED_TEXT, |
| 1442 | reinterpret_cast<const char*>(u8"IstanBul"), 8, |
| 1443 | dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1444 | assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıb"), |
| 1445 | UnicodeString::fromUTF8(StringPiece(dest, length))); |
| 1446 | static const EditChange lowerExpectedChanges[] = { |
| 1447 | { TRUE, 1, 2 }, |
| 1448 | { FALSE, 4, 4 }, |
| 1449 | { TRUE, 1, 1 }, |
| 1450 | { FALSE, 2, 2 } |
| 1451 | }; |
| 1452 | TestUtility::checkEditsIter(*this, u"toLower(IstanBul)", |
| 1453 | edits.getFineIterator(), edits.getFineIterator(), |
| 1454 | lowerExpectedChanges, UPRV_LENGTHOF(lowerExpectedChanges), |
| 1455 | TRUE, errorCode); |
| 1456 | |
| 1457 | edits.reset(); |
| 1458 | length = CaseMap::utf8ToUpper("el", U_OMIT_UNCHANGED_TEXT, |
| 1459 | reinterpret_cast<const char*>(u8"Πατάτα"), 6 * 2, |
| 1460 | dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1461 | assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΑΤΑΤΑ"), |
| 1462 | UnicodeString::fromUTF8(StringPiece(dest, length))); |
| 1463 | static const EditChange upperExpectedChanges[] = { |
| 1464 | { FALSE, 2, 2 }, |
| 1465 | { TRUE, 2, 2 }, |
| 1466 | { TRUE, 2, 2 }, |
| 1467 | { TRUE, 2, 2 }, |
| 1468 | { TRUE, 2, 2 }, |
| 1469 | { TRUE, 2, 2 } |
| 1470 | }; |
| 1471 | TestUtility::checkEditsIter(*this, u"toUpper(Πατάτα)", |
| 1472 | edits.getFineIterator(), edits.getFineIterator(), |
| 1473 | upperExpectedChanges, UPRV_LENGTHOF(upperExpectedChanges), |
| 1474 | TRUE, errorCode); |
| 1475 | |
| 1476 | edits.reset(); |
| 1477 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1478 | length = CaseMap::utf8ToTitle("nl", |
| 1479 | U_OMIT_UNCHANGED_TEXT | |
| 1480 | U_TITLECASE_NO_BREAK_ADJUSTMENT | |
| 1481 | U_TITLECASE_NO_LOWERCASE, |
| 1482 | nullptr, reinterpret_cast<const char*>(u8"IjssEL IglOo"), 12, |
| 1483 | dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1484 | assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"J"), |
| 1485 | UnicodeString::fromUTF8(StringPiece(dest, length))); |
| 1486 | static const EditChange titleExpectedChanges[] = { |
| 1487 | { FALSE, 1, 1 }, |
| 1488 | { TRUE, 1, 1 }, |
| 1489 | { FALSE, 10, 10 } |
| 1490 | }; |
| 1491 | TestUtility::checkEditsIter(*this, u"toTitle(IjssEL IglOo)", |
| 1492 | edits.getFineIterator(), edits.getFineIterator(), |
| 1493 | titleExpectedChanges, UPRV_LENGTHOF(titleExpectedChanges), |
| 1494 | TRUE, errorCode); |
| 1495 | #endif |
| 1496 | |
| 1497 | // No explicit nor automatic edits.reset(). Edits should be appended. |
| 1498 | length = CaseMap::utf8Fold(U_OMIT_UNCHANGED_TEXT | U_EDITS_NO_RESET | |
| 1499 | U_FOLD_CASE_EXCLUDE_SPECIAL_I, |
| 1500 | reinterpret_cast<const char*>(u8"IßtanBul"), 1 + 2 + 6, |
| 1501 | dest, UPRV_LENGTHOF(dest), &edits, errorCode); |
| 1502 | assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ıssb"), |
| 1503 | UnicodeString::fromUTF8(StringPiece(dest, length))); |
| 1504 | static const EditChange foldExpectedChanges[] = { |
| 1505 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1506 | // From titlecasing. |
| 1507 | { FALSE, 1, 1 }, |
| 1508 | { TRUE, 1, 1 }, |
| 1509 | { FALSE, 10, 10 }, |
| 1510 | #endif |
| 1511 | // From case folding. |
| 1512 | { TRUE, 1, 2 }, |
| 1513 | { TRUE, 2, 2 }, |
| 1514 | { FALSE, 3, 3 }, |
| 1515 | { TRUE, 1, 1 }, |
| 1516 | { FALSE, 2, 2 } |
| 1517 | }; |
| 1518 | TestUtility::checkEditsIter(*this, u"foldCase(IßtanBul)", |
| 1519 | edits.getFineIterator(), edits.getFineIterator(), |
| 1520 | foldExpectedChanges, UPRV_LENGTHOF(foldExpectedChanges), |
| 1521 | TRUE, errorCode); |
| 1522 | } |
| 1523 | |
| 1524 | void StringCaseTest::TestCaseMapToString() { |
| 1525 | // This test function name is parallel with one in UCharacterCaseTest.java. |
| 1526 | // It is a bit of a misnomer until we have CaseMap API that writes to |
| 1527 | // a UnicodeString, at which point we should change this code here. |
| 1528 | IcuTestErrorCode errorCode(*this, "TestCaseMapToString"); |
| 1529 | UChar dest[20]; |
| 1530 | |
| 1531 | // Omit unchanged text. |
| 1532 | int32_t length = CaseMap::toLower("tr", U_OMIT_UNCHANGED_TEXT, |
| 1533 | u"IstanBul", 8, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1534 | assertEquals(u"toLower(IstanBul)", |
| 1535 | UnicodeString(u"ıb"), UnicodeString(TRUE, dest, length)); |
| 1536 | length = CaseMap::toUpper("el", U_OMIT_UNCHANGED_TEXT, |
| 1537 | u"Πατάτα", 6, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1538 | assertEquals(u"toUpper(Πατάτα)", |
| 1539 | UnicodeString(u"ΑΤΑΤΑ"), UnicodeString(TRUE, dest, length)); |
| 1540 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1541 | length = CaseMap::toTitle("nl", |
| 1542 | U_OMIT_UNCHANGED_TEXT | |
| 1543 | U_TITLECASE_NO_BREAK_ADJUSTMENT | |
| 1544 | U_TITLECASE_NO_LOWERCASE, |
| 1545 | nullptr, u"IjssEL IglOo", 12, |
| 1546 | dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1547 | assertEquals(u"toTitle(IjssEL IglOo)", |
| 1548 | UnicodeString(u"J"), UnicodeString(TRUE, dest, length)); |
| 1549 | #endif |
| 1550 | length = CaseMap::fold(U_OMIT_UNCHANGED_TEXT | U_FOLD_CASE_EXCLUDE_SPECIAL_I, |
| 1551 | u"IßtanBul", 8, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1552 | assertEquals(u"foldCase(IßtanBul)", |
| 1553 | UnicodeString(u"ıssb"), UnicodeString(TRUE, dest, length)); |
| 1554 | |
| 1555 | // Return the whole result string. |
| 1556 | length = CaseMap::toLower("tr", 0, |
| 1557 | u"IstanBul", 8, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1558 | assertEquals(u"toLower(IstanBul)", |
| 1559 | UnicodeString(u"ıstanbul"), UnicodeString(TRUE, dest, length)); |
| 1560 | length = CaseMap::toUpper("el", 0, |
| 1561 | u"Πατάτα", 6, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1562 | assertEquals(u"toUpper(Πατάτα)", |
| 1563 | UnicodeString(u"ΠΑΤΑΤΑ"), UnicodeString(TRUE, dest, length)); |
| 1564 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1565 | length = CaseMap::toTitle("nl", |
| 1566 | U_TITLECASE_NO_BREAK_ADJUSTMENT | |
| 1567 | U_TITLECASE_NO_LOWERCASE, |
| 1568 | nullptr, u"IjssEL IglOo", 12, |
| 1569 | dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1570 | assertEquals(u"toTitle(IjssEL IglOo)", |
| 1571 | UnicodeString(u"IJssEL IglOo"), UnicodeString(TRUE, dest, length)); |
| 1572 | #endif |
| 1573 | length = CaseMap::fold(U_FOLD_CASE_EXCLUDE_SPECIAL_I, |
| 1574 | u"IßtanBul", 8, dest, UPRV_LENGTHOF(dest), nullptr, errorCode); |
| 1575 | assertEquals(u"foldCase(IßtanBul)", |
| 1576 | UnicodeString(u"ısstanbul"), UnicodeString(TRUE, dest, length)); |
| 1577 | } |
| 1578 | |
| 1579 | void StringCaseTest::TestCaseMapUTF8ToString() { |
| 1580 | IcuTestErrorCode errorCode(*this, "TestCaseMapUTF8ToString"); |
| 1581 | std::string dest; |
| 1582 | StringByteSink<std::string> sink(&dest); |
| 1583 | |
| 1584 | // Omit unchanged text. |
| 1585 | CaseMap::utf8ToLower("tr", U_OMIT_UNCHANGED_TEXT, u8"IstanBul", sink, nullptr, errorCode); |
| 1586 | assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıb"), UnicodeString::fromUTF8(dest)); |
| 1587 | dest.clear(); |
| 1588 | CaseMap::utf8ToUpper("el", U_OMIT_UNCHANGED_TEXT, u8"Πατάτα", sink, nullptr, errorCode); |
| 1589 | assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΑΤΑΤΑ"), |
| 1590 | UnicodeString::fromUTF8(dest)); |
| 1591 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1592 | dest.clear(); |
| 1593 | CaseMap::utf8ToTitle( |
| 1594 | "nl", U_OMIT_UNCHANGED_TEXT | U_TITLECASE_NO_BREAK_ADJUSTMENT | U_TITLECASE_NO_LOWERCASE, |
| 1595 | nullptr, u8"IjssEL IglOo", sink, nullptr, errorCode); |
| 1596 | assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"J"), |
| 1597 | UnicodeString::fromUTF8(dest)); |
| 1598 | #endif |
| 1599 | dest.clear(); |
| 1600 | CaseMap::utf8Fold(U_OMIT_UNCHANGED_TEXT | U_FOLD_CASE_EXCLUDE_SPECIAL_I, |
| 1601 | u8"IßtanBul", sink, nullptr, errorCode); |
| 1602 | assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ıssb"), |
| 1603 | UnicodeString::fromUTF8(dest)); |
| 1604 | |
| 1605 | // Return the whole result string. |
| 1606 | dest.clear(); |
| 1607 | CaseMap::utf8ToLower("tr", 0, u8"IstanBul", sink, nullptr, errorCode); |
| 1608 | assertEquals(u"toLower(IstanBul)", UnicodeString(u"ıstanbul"), |
| 1609 | UnicodeString::fromUTF8(dest)); |
| 1610 | dest.clear(); |
| 1611 | CaseMap::utf8ToUpper("el", 0, u8"Πατάτα", sink, nullptr, errorCode); |
| 1612 | assertEquals(u"toUpper(Πατάτα)", UnicodeString(u"ΠΑΤΑΤΑ"), |
| 1613 | UnicodeString::fromUTF8(dest)); |
| 1614 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1615 | dest.clear(); |
| 1616 | CaseMap::utf8ToTitle("nl", U_TITLECASE_NO_BREAK_ADJUSTMENT | U_TITLECASE_NO_LOWERCASE, |
| 1617 | nullptr, u8"IjssEL IglOo", sink, nullptr, errorCode); |
| 1618 | assertEquals(u"toTitle(IjssEL IglOo)", UnicodeString(u"IJssEL IglOo"), |
| 1619 | UnicodeString::fromUTF8(dest)); |
| 1620 | #endif |
| 1621 | dest.clear(); |
| 1622 | CaseMap::utf8Fold(U_FOLD_CASE_EXCLUDE_SPECIAL_I, u8"IßtanBul", sink, nullptr, errorCode); |
| 1623 | assertEquals(u"foldCase(IßtanBul)", UnicodeString(u"ısstanbul"), |
| 1624 | UnicodeString::fromUTF8(dest)); |
| 1625 | } |
| 1626 | |
| 1627 | void StringCaseTest::TestLongUnicodeString() { |
| 1628 | // Code coverage for UnicodeString case mapping code handling |
| 1629 | // long strings or many changes in a string. |
| 1630 | UnicodeString s(TRUE, |
| 1631 | (const UChar *) |
| 1632 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF" |
| 1633 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF" |
| 1634 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF" |
| 1635 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF" |
| 1636 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF" |
| 1637 | u"aaaaaaaaaabbbbbbbbbbccccccccccddddddddddeeeeeeeeeeF", 6 * 51); |
| 1638 | UnicodeString expected(TRUE, |
| 1639 | (const UChar *) |
| 1640 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF" |
| 1641 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF" |
| 1642 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF" |
| 1643 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF" |
| 1644 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF" |
| 1645 | u"AAAAAAAAAABBBBBBBBBBCCCCCCCCCCDDDDDDDDDDEEEEEEEEEEF", 6 * 51); |
| 1646 | s.toUpper(Locale::getRoot()); |
| 1647 | assertEquals("string length 306", expected, s); |
| 1648 | } |
| 1649 | |
| 1650 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1651 | void StringCaseTest::TestBug13127() { |
| 1652 | // Test case crashed when the bug was present. |
| 1653 | const char16_t *s16 = u"日本語"; |
| 1654 | UnicodeString s(TRUE, s16, -1); |
| 1655 | s.toTitle(0, Locale::getEnglish()); |
| 1656 | } |
| 1657 | |
| 1658 | void StringCaseTest::TestInPlaceTitle() { |
| 1659 | // Similar to TestBug13127. u_strToTitle() can modify the buffer in-place. |
| 1660 | IcuTestErrorCode errorCode(*this, "TestInPlaceTitle"); |
| 1661 | char16_t s[32] = u"ß ß ß日本語 abcdef"; |
| 1662 | const char16_t *expected = u"Ss Ss Ss日本語 Abcdef"; |
| 1663 | int32_t length = u_strToTitle(s, UPRV_LENGTHOF(s), s, -1, nullptr, "", errorCode); |
| 1664 | assertEquals("u_strToTitle(in-place) length", u_strlen(expected), length); |
| 1665 | assertEquals("u_strToTitle(in-place)", expected, s); |
| 1666 | } |
| 1667 | #endif |
| 1668 | |
| 1669 | void StringCaseTest::TestCaseMapEditsIteratorDocs() { |
| 1670 | IcuTestErrorCode status(*this, "TestCaseMapEditsIteratorDocs"); |
| 1671 | const char16_t* input = u"abcßDeF"; |
| 1672 | int32_t inputLength = u_strlen(input); |
| 1673 | // output: "abcssdef" |
| 1674 | |
| 1675 | char16_t output[10]; |
| 1676 | Edits edits; |
| 1677 | CaseMap::fold(0, input, -1, output, 10, &edits, status); |
| 1678 | |
| 1679 | static const char16_t* fineIteratorExpected[] = { |
| 1680 | u"{ src[0..3] ≡ dest[0..3] (no-change) }", |
| 1681 | u"{ src[3..4] ⇝ dest[3..5], repl[0..2] }", |
| 1682 | u"{ src[4..5] ⇝ dest[5..6], repl[2..3] }", |
| 1683 | u"{ src[5..6] ≡ dest[6..7] (no-change) }", |
| 1684 | u"{ src[6..7] ⇝ dest[7..8], repl[3..4] }", |
| 1685 | }; |
| 1686 | static const char16_t* fineChangesIteratorExpected[] = { |
| 1687 | u"{ src[3..4] ⇝ dest[3..5], repl[0..2] }", |
| 1688 | u"{ src[4..5] ⇝ dest[5..6], repl[2..3] }", |
| 1689 | u"{ src[6..7] ⇝ dest[7..8], repl[3..4] }", |
| 1690 | }; |
| 1691 | static const char16_t* coarseIteratorExpected[] = { |
| 1692 | u"{ src[0..3] ≡ dest[0..3] (no-change) }", |
| 1693 | u"{ src[3..5] ⇝ dest[3..6], repl[0..3] }", |
| 1694 | u"{ src[5..6] ≡ dest[6..7] (no-change) }", |
| 1695 | u"{ src[6..7] ⇝ dest[7..8], repl[3..4] }", |
| 1696 | }; |
| 1697 | static const char16_t* coarseChangesIteratorExpected[] = { |
| 1698 | u"{ src[3..5] ⇝ dest[3..6], repl[0..3] }", |
| 1699 | u"{ src[6..7] ⇝ dest[7..8], repl[3..4] }", |
| 1700 | }; |
| 1701 | |
| 1702 | // Expected destination indices when source index is queried |
| 1703 | static int32_t expectedDestFineEditIndices[] = {0, 0, 0, 3, 5, 6, 7}; |
| 1704 | static int32_t expectedDestCoarseEditIndices[] = {0, 0, 0, 3, 3, 6, 7}; |
| 1705 | static int32_t expectedDestFineStringIndices[] = {0, 1, 2, 3, 5, 6, 7}; |
| 1706 | static int32_t expectedDestCoarseStringIndices[] = {0, 1, 2, 3, 6, 6, 7}; |
| 1707 | |
| 1708 | // Expected source indices when destination index is queried |
| 1709 | static int32_t expectedSrcFineEditIndices[] = { 0, 0, 0, 3, 3, 4, 5, 6 }; |
| 1710 | static int32_t expectedSrcCoarseEditIndices[] = { 0, 0, 0, 3, 3, 3, 5, 6 }; |
| 1711 | static int32_t expectedSrcFineStringIndices[] = { 0, 1, 2, 3, 4, 4, 5, 6 }; |
| 1712 | static int32_t expectedSrcCoarseStringIndices[] = { 0, 1, 2, 3, 5, 5, 5, 6 }; |
| 1713 | |
| 1714 | // Demonstrate the iterator next() method: |
| 1715 | Edits::Iterator fineIterator = edits.getFineIterator(); |
| 1716 | int i = 0; |
| 1717 | UnicodeString toString; |
| 1718 | while (fineIterator.next(status)) { |
| 1719 | UnicodeString expected = fineIteratorExpected[i++]; |
| 1720 | assertEquals(UnicodeString(u"Iteration #") + i, |
| 1721 | expected, |
| 1722 | fineIterator.toString(toString.remove())); |
| 1723 | } |
| 1724 | Edits::Iterator fineChangesIterator = edits.getFineChangesIterator(); |
| 1725 | i = 0; |
| 1726 | while (fineChangesIterator.next(status)) { |
| 1727 | UnicodeString expected = fineChangesIteratorExpected[i++]; |
| 1728 | assertEquals(UnicodeString(u"Iteration #") + i, |
| 1729 | expected, |
| 1730 | fineChangesIterator.toString(toString.remove())); |
| 1731 | } |
| 1732 | Edits::Iterator coarseIterator = edits.getCoarseIterator(); |
| 1733 | i = 0; |
| 1734 | while (coarseIterator.next(status)) { |
| 1735 | UnicodeString expected = coarseIteratorExpected[i++]; |
| 1736 | assertEquals(UnicodeString(u"Iteration #") + i, |
| 1737 | expected, |
| 1738 | coarseIterator.toString(toString.remove())); |
| 1739 | } |
| 1740 | Edits::Iterator coarseChangesIterator = edits.getCoarseChangesIterator(); |
| 1741 | i = 0; |
| 1742 | while (coarseChangesIterator.next(status)) { |
| 1743 | UnicodeString expected = coarseChangesIteratorExpected[i++]; |
| 1744 | assertEquals(UnicodeString(u"Iteration #") + i, |
| 1745 | expected, |
| 1746 | coarseChangesIterator.toString(toString.remove())); |
| 1747 | } |
| 1748 | |
| 1749 | // Demonstrate the iterator indexing methods: |
| 1750 | // fineIterator should have the same behavior as fineChangesIterator, and |
| 1751 | // coarseIterator should have the same behavior as coarseChangesIterator. |
| 1752 | for (int32_t srcIndex=0; srcIndex<inputLength; srcIndex++) { |
| 1753 | fineIterator.findSourceIndex(srcIndex, status); |
| 1754 | fineChangesIterator.findSourceIndex(srcIndex, status); |
| 1755 | coarseIterator.findSourceIndex(srcIndex, status); |
| 1756 | coarseChangesIterator.findSourceIndex(srcIndex, status); |
| 1757 | |
| 1758 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1759 | expectedDestFineEditIndices[srcIndex], |
| 1760 | fineIterator.destinationIndex()); |
| 1761 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1762 | expectedDestFineEditIndices[srcIndex], |
| 1763 | fineChangesIterator.destinationIndex()); |
| 1764 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1765 | expectedDestCoarseEditIndices[srcIndex], |
| 1766 | coarseIterator.destinationIndex()); |
| 1767 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1768 | expectedDestCoarseEditIndices[srcIndex], |
| 1769 | coarseChangesIterator.destinationIndex()); |
| 1770 | |
| 1771 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1772 | expectedDestFineStringIndices[srcIndex], |
| 1773 | fineIterator.destinationIndexFromSourceIndex(srcIndex, status)); |
| 1774 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1775 | expectedDestFineStringIndices[srcIndex], |
| 1776 | fineChangesIterator.destinationIndexFromSourceIndex(srcIndex, status)); |
| 1777 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1778 | expectedDestCoarseStringIndices[srcIndex], |
| 1779 | coarseIterator.destinationIndexFromSourceIndex(srcIndex, status)); |
| 1780 | assertEquals(UnicodeString("Source index: ") + srcIndex, |
| 1781 | expectedDestCoarseStringIndices[srcIndex], |
| 1782 | coarseChangesIterator.destinationIndexFromSourceIndex(srcIndex, status)); |
| 1783 | } |
| 1784 | for (int32_t destIndex=0; destIndex<inputLength; destIndex++) { |
| 1785 | fineIterator.findDestinationIndex(destIndex, status); |
| 1786 | fineChangesIterator.findDestinationIndex(destIndex, status); |
| 1787 | coarseIterator.findDestinationIndex(destIndex, status); |
| 1788 | coarseChangesIterator.findDestinationIndex(destIndex, status); |
| 1789 | |
| 1790 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1791 | expectedSrcFineEditIndices[destIndex], |
| 1792 | fineIterator.sourceIndex()); |
| 1793 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1794 | expectedSrcFineEditIndices[destIndex], |
| 1795 | fineChangesIterator.sourceIndex()); |
| 1796 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1797 | expectedSrcCoarseEditIndices[destIndex], |
| 1798 | coarseIterator.sourceIndex()); |
| 1799 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1800 | expectedSrcCoarseEditIndices[destIndex], |
| 1801 | coarseChangesIterator.sourceIndex()); |
| 1802 | |
| 1803 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1804 | expectedSrcFineStringIndices[destIndex], |
| 1805 | fineIterator.sourceIndexFromDestinationIndex(destIndex, status)); |
| 1806 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1807 | expectedSrcFineStringIndices[destIndex], |
| 1808 | fineChangesIterator.sourceIndexFromDestinationIndex(destIndex, status)); |
| 1809 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1810 | expectedSrcCoarseStringIndices[destIndex], |
| 1811 | coarseIterator.sourceIndexFromDestinationIndex(destIndex, status)); |
| 1812 | assertEquals(UnicodeString("Destination index: ") + destIndex, |
| 1813 | expectedSrcCoarseStringIndices[destIndex], |
| 1814 | coarseChangesIterator.sourceIndexFromDestinationIndex(destIndex, status)); |
| 1815 | } |
| 1816 | } |
| 1817 | |
| 1818 | void StringCaseTest::TestCaseMapGreekExtended() { |
| 1819 | // Ticket 13851 |
| 1820 | UnicodeString s(u"\u1F80\u1F88\u1FFC"); |
| 1821 | UnicodeString result(s); |
| 1822 | result.toLower(Locale::getRoot()); |
| 1823 | assertEquals(u"lower", u"\u1F80\u1F80\u1FF3", result); |
| 1824 | #if !UCONFIG_NO_BREAK_ITERATION |
| 1825 | result = s; |
| 1826 | result.toTitle(nullptr, Locale::getRoot()); |
| 1827 | assertEquals(u"title", u"\u1F88\u1F80\u1FF3", result); |
| 1828 | #endif |
| 1829 | } |
| 1830 | |
| 1831 | //#endif |