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 | * Copyright (c) 2001-2009, International Business Machines |
| 5 | * Corporation and others. All Rights Reserved. |
| 6 | ********************************************************************* |
| 7 | * This test program is intended for testing error conditions of the |
| 8 | * transliterator APIs to make sure the exceptions are raised where |
| 9 | * necessary. |
| 10 | * |
| 11 | * Date Name Description |
| 12 | * 11/14/2001 hshih Creation. |
| 13 | * |
| 14 | ********************************************************************/ |
| 15 | |
| 16 | #include "unicode/utypes.h" |
| 17 | |
| 18 | #if !UCONFIG_NO_TRANSLITERATION |
| 19 | |
| 20 | #include "ittrans.h" |
| 21 | #include "trnserr.h" |
| 22 | #include "unicode/utypes.h" |
| 23 | #include "unicode/translit.h" |
| 24 | #include "unicode/uniset.h" |
| 25 | #include "unicode/unifilt.h" |
| 26 | #include "cpdtrans.h" |
| 27 | #include <string.h> |
| 28 | #include <stdio.h> |
| 29 | #include <stdlib.h> |
| 30 | #include "unicode/rep.h" |
| 31 | #include "unicode/locid.h" |
| 32 | |
| 33 | //--------------------------------------------- |
| 34 | // runIndexedTest |
| 35 | //--------------------------------------------- |
| 36 | |
| 37 | void |
| 38 | TransliteratorErrorTest::runIndexedTest(int32_t index, UBool exec, |
| 39 | const char* &name, char* /*par*/) { |
| 40 | switch (index) { |
| 41 | TESTCASE(0,TestTransliteratorErrors); |
| 42 | TESTCASE(1, TestUnicodeSetErrors); |
| 43 | TESTCASE(2, TestRBTErrors); |
| 44 | TESTCASE(3, TestCoverage); |
| 45 | //TESTCASE(3, TestUniToHexErrors); |
| 46 | //TESTCASE(4, TestHexToUniErrors); |
| 47 | // TODO: Add a subclass to test clone(). |
| 48 | default: name = ""; break; |
| 49 | } |
| 50 | } |
| 51 | |
| 52 | |
| 53 | void TransliteratorErrorTest::TestTransliteratorErrors() { |
| 54 | UnicodeString trans="Latin-Greek"; |
| 55 | UnicodeString bogusID="LATINGREEK-GREEKLATIN"; |
| 56 | UnicodeString newID="Bogus-Latin"; |
| 57 | UnicodeString newIDRules="zzz > Z; f <> ph"; |
| 58 | UnicodeString bogusRules="a } [b-g m-p "; |
| 59 | UParseError parseError; |
| 60 | UErrorCode status = U_ZERO_ERROR; |
| 61 | UnicodeString testString="A quick fox jumped over the lazy dog."; |
| 62 | UnicodeString insertString="cats and dogs"; |
| 63 | int32_t stoppedAt = 0, len; |
| 64 | UTransPosition pos; |
| 65 | |
| 66 | Transliterator* t= Transliterator::createInstance(trans, UTRANS_FORWARD, parseError, status); |
| 67 | if(t==0 || U_FAILURE(status)){ |
| 68 | dataerrln("FAIL: construction of Latin-Greek - %s", u_errorName(status)); |
| 69 | return; |
| 70 | } |
| 71 | pos.contextLimit = 0; |
| 72 | pos.contextStart = 0; |
| 73 | pos.limit = 0; |
| 74 | pos.start = 0; |
| 75 | len = testString.length(); |
| 76 | stoppedAt = t->transliterate(testString, 0, 100); |
| 77 | if (stoppedAt != -1) { |
| 78 | errln("FAIL: Out of bounds check failed (1)."); |
| 79 | } else if (testString.length() != len) { |
| 80 | testString="A quick fox jumped over the lazy dog."; |
| 81 | errln("FAIL: Transliterate fails and the target string was modified."); |
| 82 | } |
| 83 | stoppedAt = t->transliterate(testString, 100, testString.length()-1); |
| 84 | if (stoppedAt != -1) |
| 85 | errln("FAIL: Out of bounds check failed (2)."); |
| 86 | else if (testString.length() != len) { |
| 87 | testString="A quick fox jumped over the lazy dog."; |
| 88 | errln("FAIL: Transliterate fails and the target string was modified."); |
| 89 | } |
| 90 | pos.start = 100; |
| 91 | pos.limit = testString.length(); |
| 92 | t->transliterate(testString, pos, status); |
| 93 | if (U_SUCCESS(status)) { |
| 94 | errln("FAIL: Start offset is out of bounds, error not reported.\n"); |
| 95 | } |
| 96 | status = U_ZERO_ERROR; |
| 97 | pos.limit = 100; |
| 98 | pos.start = 0; |
| 99 | t->transliterate(testString, pos, status); |
| 100 | if (U_SUCCESS(status)) { |
| 101 | errln("FAIL: Limit offset is out of bounds, error not reported.\n"); |
| 102 | } |
| 103 | status = U_ZERO_ERROR; |
| 104 | len = pos.contextLimit = testString.length(); |
| 105 | pos.contextStart = 0; |
| 106 | pos.limit = len - 1; |
| 107 | pos.start = 5; |
| 108 | t->transliterate(testString, pos, insertString, status); |
| 109 | if (len == pos.limit) { |
| 110 | errln("FAIL: Test insertion with string: the transliteration position limit didn't change as expected."); |
| 111 | if (U_SUCCESS(status)) { |
| 112 | errln("FAIL: Error code wasn't set either."); |
| 113 | } |
| 114 | } |
| 115 | status = U_ZERO_ERROR; |
| 116 | pos.contextStart = 0; |
| 117 | pos.contextLimit = testString.length(); |
| 118 | pos.limit = testString.length() -1; |
| 119 | pos.start = 5; |
| 120 | t->transliterate(testString, pos, (UChar32)0x0061, status); |
| 121 | if (len == pos.limit) { |
| 122 | errln("FAIL: Test insertion with character: the transliteration position limit didn't change as expected."); |
| 123 | if (U_SUCCESS(status)) { |
| 124 | errln("FAIL: Error code wasn't set either."); |
| 125 | } |
| 126 | } |
| 127 | status = U_ZERO_ERROR; |
| 128 | len = pos.limit = testString.length(); |
| 129 | pos.contextStart = 0; |
| 130 | pos.contextLimit = testString.length() - 1; |
| 131 | pos.start = 5; |
| 132 | t->transliterate(testString, pos, insertString, status); |
| 133 | if (U_SUCCESS(status)) { |
| 134 | errln("FAIL: Out of bounds check failed (3)."); |
| 135 | if (testString.length() != len) |
| 136 | errln("FAIL: The input string was modified though the offsets were out of bounds."); |
| 137 | } |
| 138 | Transliterator* t1= Transliterator::createInstance(bogusID, UTRANS_FORWARD, parseError, status); |
| 139 | if(t1!=0 || U_SUCCESS(status)){ |
| 140 | delete t1; |
| 141 | errln("FAIL: construction of bogus ID \"LATINGREEK-GREEKLATIN\""); |
| 142 | } |
| 143 | status = U_ZERO_ERROR; |
| 144 | Transliterator* t2 = Transliterator::createFromRules(newID, newIDRules, UTRANS_FORWARD, parseError, status); |
| 145 | if (U_SUCCESS(status)) { |
| 146 | Transliterator* t3 = t2->createInverse(status); |
| 147 | if (U_SUCCESS(status)) { |
| 148 | delete t3; |
| 149 | errln("FAIL: The newID transliterator was not registered so createInverse should fail."); |
| 150 | } else { |
| 151 | delete t3; |
| 152 | } |
| 153 | } |
| 154 | status = U_ZERO_ERROR; |
| 155 | Transliterator* t4 = Transliterator::createFromRules(newID, bogusRules, UTRANS_FORWARD, parseError, status); |
| 156 | if (t4 != NULL || U_SUCCESS(status)) { |
| 157 | errln("FAIL: The rules is malformed but error was not reported."); |
| 158 | if (parseError.offset != -1) { |
| 159 | errln("FAIL: The parse error offset isn't set correctly when fails."); |
| 160 | } else if (parseError.postContext[0] == 0 || parseError.preContext[0] == 0) { |
| 161 | errln("FAIL: The parse error pre/post context isn't reset properly."); |
| 162 | } |
| 163 | delete t4; |
| 164 | } |
| 165 | delete t; |
| 166 | delete t2; |
| 167 | } |
| 168 | |
| 169 | void TransliteratorErrorTest::TestUnicodeSetErrors() { |
| 170 | UnicodeString badPattern="[[:L:]-[0x0300-0x0400]"; |
| 171 | UnicodeSet set; |
| 172 | UErrorCode status = U_ZERO_ERROR; |
| 173 | UnicodeString result; |
| 174 | |
| 175 | if (!set.isEmpty()) { |
| 176 | errln("FAIL: The default ctor of UnicodeSet created a non-empty object."); |
| 177 | } |
| 178 | set.applyPattern(badPattern, status); |
| 179 | if (U_SUCCESS(status)) { |
| 180 | errln("FAIL: Applied a bad pattern to the UnicodeSet object okay."); |
| 181 | } |
| 182 | status = U_ZERO_ERROR; |
| 183 | UnicodeSet *set1 = new UnicodeSet(badPattern, status); |
| 184 | if (U_SUCCESS(status)) { |
| 185 | errln("FAIL: Created a UnicodeSet based on bad patterns."); |
| 186 | } |
| 187 | delete set1; |
| 188 | } |
| 189 | |
| 190 | //void TransliteratorErrorTest::TestUniToHexErrors() { |
| 191 | // UErrorCode status = U_ZERO_ERROR; |
| 192 | // Transliterator *t = new UnicodeToHexTransliterator("", TRUE, NULL, status); |
| 193 | // if (U_SUCCESS(status)) { |
| 194 | // errln("FAIL: Created a UnicodeToHexTransliterator with an empty pattern."); |
| 195 | // } |
| 196 | // delete t; |
| 197 | // |
| 198 | // status = U_ZERO_ERROR; |
| 199 | // t = new UnicodeToHexTransliterator("\\x", TRUE, NULL, status); |
| 200 | // if (U_SUCCESS(status)) { |
| 201 | // errln("FAIL: Created a UnicodeToHexTransliterator with a bad pattern."); |
| 202 | // } |
| 203 | // delete t; |
| 204 | // |
| 205 | // status = U_ZERO_ERROR; |
| 206 | // t = new UnicodeToHexTransliterator(); |
| 207 | // ((UnicodeToHexTransliterator*)t)->applyPattern("\\x", status); |
| 208 | // if (U_SUCCESS(status)) { |
| 209 | // errln("FAIL: UnicodeToHexTransliterator::applyPattern succeeded with a bad pattern."); |
| 210 | // } |
| 211 | // delete t; |
| 212 | //} |
| 213 | |
| 214 | void TransliteratorErrorTest::TestRBTErrors() { |
| 215 | |
| 216 | UnicodeString rules="ab>y"; |
| 217 | UnicodeString id="MyRandom-YReverse"; |
| 218 | //UnicodeString goodPattern="[[:L:]&[\\u0000-\\uFFFF]]"; /* all BMP letters */ |
| 219 | UErrorCode status = U_ZERO_ERROR; |
| 220 | UParseError parseErr; |
| 221 | /*UnicodeSet *set = new UnicodeSet(goodPattern, status); |
| 222 | if (U_FAILURE(status)) { |
| 223 | errln("FAIL: Was not able to create a good UnicodeSet based on valid patterns."); |
| 224 | return; |
| 225 | }*/ |
| 226 | Transliterator *t = Transliterator::createFromRules(id, rules, UTRANS_REVERSE, parseErr, status); |
| 227 | if (U_FAILURE(status)) { |
| 228 | errln("FAIL: Was not able to create a good RBT to test registration."); |
| 229 | //delete set; |
| 230 | return; |
| 231 | } |
| 232 | Transliterator::registerInstance(t); |
| 233 | Transliterator::unregister(id); |
| 234 | status = U_ZERO_ERROR; |
| 235 | Transliterator* t1= Transliterator::createInstance(id, UTRANS_REVERSE, parseErr, status); |
| 236 | if(U_SUCCESS(status)){ |
| 237 | delete t1; |
| 238 | errln("FAIL: construction of unregistered ID failed."); |
| 239 | } |
| 240 | } |
| 241 | |
| 242 | //void TransliteratorErrorTest::TestHexToUniErrors() { |
| 243 | // UErrorCode status = U_ZERO_ERROR; |
| 244 | // Transliterator *t = new HexToUnicodeTransliterator("", NULL, status); |
| 245 | // if (U_FAILURE(status)) { |
| 246 | // errln("FAIL: Could not create a HexToUnicodeTransliterator with an empty pattern."); |
| 247 | // } |
| 248 | // delete t; |
| 249 | // status = U_ZERO_ERROR; |
| 250 | // t = new HexToUnicodeTransliterator("\\x", NULL, status); |
| 251 | // if (U_SUCCESS(status)) { |
| 252 | // errln("FAIL: Created a HexToUnicodeTransliterator with a bad pattern."); |
| 253 | // } |
| 254 | // delete t; |
| 255 | // status = U_ZERO_ERROR; |
| 256 | // t = new HexToUnicodeTransliterator(); |
| 257 | // ((HexToUnicodeTransliterator*)t)->applyPattern("\\x", status); |
| 258 | // if (U_SUCCESS(status)) { |
| 259 | // errln("FAIL: HexToUnicodeTransliterator::applyPattern succeeded with a bad pattern."); |
| 260 | // } |
| 261 | // delete t; |
| 262 | //} |
| 263 | |
| 264 | class StubTransliterator: public Transliterator{ |
| 265 | public: |
| 266 | StubTransliterator(): Transliterator(UNICODE_STRING_SIMPLE("Any-Null"), 0) {} |
| 267 | virtual void handleTransliterate(Replaceable& ,UTransPosition& offsets,UBool) const override { |
| 268 | offsets.start = offsets.limit; |
| 269 | } |
| 270 | |
| 271 | virtual UClassID getDynamicClassID() const override { |
| 272 | static char classID = 0; |
| 273 | return (UClassID)&classID; |
| 274 | } |
| 275 | }; |
| 276 | |
| 277 | void TransliteratorErrorTest::TestCoverage() { |
| 278 | StubTransliterator stub; |
| 279 | |
| 280 | if (stub.clone() != NULL){ |
| 281 | errln("FAIL: default Transliterator::clone() should return NULL"); |
| 282 | } |
| 283 | } |
| 284 | |
| 285 | #endif /* #if !UCONFIG_NO_TRANSLITERATION */ |