1 // Copyright (C) 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
4 **********************************************************************
5 * Copyright (C) 2011-2015, International Business Machines Corporation
6 * and others. All Rights Reserved.
7 **********************************************************************
10 * IntlTestSpoof tests for USpoofDetector
13 #include "unicode/utypes.h"
15 #if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO
19 #include "unicode/normlzr.h"
20 #include "unicode/regex.h"
21 #include "unicode/unistr.h"
22 #include "unicode/uscript.h"
23 #include "unicode/uspoof.h"
26 #include "scriptset.h"
32 #define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \
33 errcheckln(status, "Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}}
35 #define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
36 errln("Test Failure at file %s, line %d: \"%s\" is false.", __FILE__, __LINE__, #expr);};}
38 #define TEST_ASSERT_MSG(expr, msg) {if ((expr)==FALSE) { \
39 dataerrln("Test Failure at file %s, line %d, %s: \"%s\" is false.", __FILE__, __LINE__, msg, #expr);};}
41 #define TEST_ASSERT_EQ(a, b) { if ((a) != (b)) { \
42 errln("Test Failure at file %s, line %d: \"%s\" (%d) != \"%s\" (%d)", \
43 __FILE__, __LINE__, #a, (a), #b, (b)); }}
45 #define TEST_ASSERT_NE(a, b) { if ((a) == (b)) { \
46 errln("Test Failure at file %s, line %d: \"%s\" (%d) == \"%s\" (%d)", \
47 __FILE__, __LINE__, #a, (a), #b, (b)); }}
50 * TEST_SETUP and TEST_TEARDOWN
51 * macros to handle the boilerplate around setting up test case.
52 * Put arbitrary test code between SETUP and TEARDOWN.
53 * "sc" is the ready-to-go SpoofChecker for use in the tests.
55 #define TEST_SETUP { \
56 UErrorCode status = U_ZERO_ERROR; \
58 sc = uspoof_open(&status); \
59 TEST_ASSERT_SUCCESS(status); \
60 USpoofCheckResult *checkResult; \
61 checkResult = uspoof_openCheckResult(&status); \
62 TEST_ASSERT_SUCCESS(status); \
63 if (U_SUCCESS(status)){
65 #define TEST_TEARDOWN \
67 TEST_ASSERT_SUCCESS(status); \
68 uspoof_closeCheckResult(checkResult); \
75 void IntlTestSpoof::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
78 logln("TestSuite spoof: ");
81 TESTCASE_AUTO(testSpoofAPI);
82 TESTCASE_AUTO(testSkeleton);
83 TESTCASE_AUTO(testAreConfusable);
84 TESTCASE_AUTO(testInvisible);
85 TESTCASE_AUTO(testConfData);
86 TESTCASE_AUTO(testBug8654);
87 TESTCASE_AUTO(testScriptSet);
88 TESTCASE_AUTO(testRestrictionLevel);
89 TESTCASE_AUTO(testMixedNumbers);
90 TESTCASE_AUTO(testBug12153);
94 void IntlTestSpoof::testSpoofAPI() {
97 UnicodeString s("xyz"); // Many latin ranges are whole-script confusable with other scripts.
98 // If this test starts failing, consult confusablesWholeScript.txt
99 int32_t position = 666;
100 int32_t checkResults = uspoof_checkUnicodeString(sc, s, &position, &status);
101 TEST_ASSERT_SUCCESS(status);
102 TEST_ASSERT_EQ(0, checkResults);
103 TEST_ASSERT_EQ(0, position);
107 UnicodeString s1("cxs");
108 UnicodeString s2 = UnicodeString("\\u0441\\u0445\\u0455").unescape(); // Cyrillic "cxs"
109 int32_t checkResults = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
110 TEST_ASSERT_SUCCESS(status);
111 TEST_ASSERT_EQ(USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE, checkResults);
116 UnicodeString s("I1l0O");
118 UnicodeString &retStr = uspoof_getSkeletonUnicodeString(sc, USPOOF_ANY_CASE, s, dest, &status);
119 TEST_ASSERT_SUCCESS(status);
120 TEST_ASSERT(UnicodeString("lllOO") == dest);
121 TEST_ASSERT(&dest == &retStr);
126 #define CHECK_SKELETON(type, input, expected) { \
127 checkSkeleton(sc, type, input, expected, __LINE__); \
131 // testSkeleton. Spot check a number of confusable skeleton substitutions from the
132 // Unicode data file confusables.txt
133 // Test cases chosen for substitutions of various lengths, and
134 // membership in different mapping tables.
135 // Note: for ICU 55, all tables collapsed to the MA table data.
136 // TODO: for ICU 56 with Unicode 8, revisit this test.
138 void IntlTestSpoof::testSkeleton() {
139 const uint32_t ML = 0;
140 const uint32_t SL = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
141 const uint32_t MA = USPOOF_ANY_CASE;
142 const uint32_t SA = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
145 CHECK_SKELETON(SL, "nochange", "nochange");
146 CHECK_SKELETON(SA, "nochange", "nochange");
147 CHECK_SKELETON(ML, "nochange", "nochange");
148 CHECK_SKELETON(MA, "nochange", "nochange");
149 CHECK_SKELETON(MA, "love", "love");
150 CHECK_SKELETON(MA, "1ove", "love"); // Digit 1 to letter l
151 CHECK_SKELETON(ML, "OOPS", "OOPS");
152 CHECK_SKELETON(ML, "00PS", "OOPS");
153 CHECK_SKELETON(MA, "OOPS", "OOPS");
154 CHECK_SKELETON(MA, "00PS", "OOPS"); // Digit 0 to letter O in any case mode only
155 CHECK_SKELETON(SL, "\\u059c", "\\u0301");
156 CHECK_SKELETON(SL, "\\u2A74", "\\u003A\\u003A\\u003D");
157 CHECK_SKELETON(SL, "\\u247E", "\\u0028\\u006C\\u006C\\u0029"); // "(ll)"
158 CHECK_SKELETON(SL, "\\uFDFB", "\\u062C\\u0644\\u0020\\u062C\\u0644\\u006c\\u0644\\u006f");
160 // This mapping exists in the ML and MA tables, does not exist in SL, SA
165 CHECK_SKELETON(SL, "\\u0C83", "\\u0983");
166 CHECK_SKELETON(SA, "\\u0C83", "\\u0983");
167 CHECK_SKELETON(ML, "\\u0C83", "\\u0983");
168 CHECK_SKELETON(MA, "\\u0C83", "\\u0983");
170 // 0391 mappings exist only in MA and SA tables.
171 CHECK_SKELETON(MA, "\\u0391", "A");
172 CHECK_SKELETON(SA, "\\u0391", "A");
173 CHECK_SKELETON(ML, "\\u0391", "A");
174 CHECK_SKELETON(SL, "\\u0391", "A");
176 // 13CF Mappings in all four tables, different in MA.
177 CHECK_SKELETON(ML, "\\u13CF", "b");
178 CHECK_SKELETON(MA, "\\u13CF", "b");
179 CHECK_SKELETON(SL, "\\u13CF", "b");
180 CHECK_SKELETON(SA, "\\u13CF", "b");
182 // 0022 ; 0027 0027 ;
184 CHECK_SKELETON(SL, "\\u0022", "\\u0027\\u0027");
185 CHECK_SKELETON(SA, "\\u0022", "\\u0027\\u0027");
186 CHECK_SKELETON(ML, "\\u0022", "\\u0027\\u0027");
187 CHECK_SKELETON(MA, "\\u0022", "\\u0027\\u0027");
189 // 017F mappings exist only in MA and SA tables.
190 CHECK_SKELETON(MA, "\\u017F", "f");
191 CHECK_SKELETON(SA, "\\u017F", "f");
192 CHECK_SKELETON(ML, "\\u017F", "f");
193 CHECK_SKELETON(SL, "\\u017F", "f");
200 // Run a single confusable skeleton transformation test case.
202 void IntlTestSpoof::checkSkeleton(const USpoofChecker *sc, uint32_t type,
203 const char *input, const char *expected, int32_t lineNum) {
204 UnicodeString uInput = UnicodeString(input).unescape();
205 UnicodeString uExpected = UnicodeString(expected).unescape();
207 UErrorCode status = U_ZERO_ERROR;
208 UnicodeString actual;
209 uspoof_getSkeletonUnicodeString(sc, type, uInput, actual, &status);
210 if (U_FAILURE(status)) {
211 errln("File %s, Line %d, Test case from line %d, status is %s", __FILE__, __LINE__, lineNum,
212 u_errorName(status));
215 if (uExpected != actual) {
216 errln("File %s, Line %d, Test case from line %d, Actual and Expected skeletons differ.",
217 __FILE__, __LINE__, lineNum);
218 errln(UnicodeString(" Actual Skeleton: \"") + actual + UnicodeString("\"\n") +
219 UnicodeString(" Expected Skeleton: \"") + uExpected + UnicodeString("\""));
223 void IntlTestSpoof::testAreConfusable() {
225 UnicodeString s1("A long string that will overflow stack buffers. A long string that will overflow stack buffers. "
226 "A long string that will overflow stack buffers. A long string that will overflow stack buffers. ");
227 UnicodeString s2("A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. "
228 "A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. ");
229 int32_t result = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
230 TEST_ASSERT_SUCCESS(status);
231 TEST_ASSERT_EQ(USPOOF_SINGLE_SCRIPT_CONFUSABLE, result);
236 void IntlTestSpoof::testInvisible() {
238 UnicodeString s = UnicodeString("abcd\\u0301ef").unescape();
239 int32_t position = -42;
240 TEST_ASSERT_EQ(0, uspoof_checkUnicodeString(sc, s, &position, &status));
241 TEST_ASSERT_SUCCESS(status);
242 TEST_ASSERT(0 == position);
244 UnicodeString s2 = UnicodeString("abcd\\u0301\\u0302\\u0301ef").unescape();
245 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s2, &position, &status));
246 TEST_ASSERT_SUCCESS(status);
247 TEST_ASSERT_EQ(0, position);
249 // Two acute accents, one from the composed a with acute accent, \u00e1,
252 UnicodeString s3 = UnicodeString("abcd\\u00e1\\u0301xyz").unescape();
253 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s3, &position, &status));
254 TEST_ASSERT_SUCCESS(status);
255 TEST_ASSERT_EQ(0, position);
259 void IntlTestSpoof::testBug8654() {
261 UnicodeString s = UnicodeString("B\\u00c1\\u0301").unescape();
262 int32_t position = -42;
263 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s, &position, &status) & USPOOF_INVISIBLE );
264 TEST_ASSERT_SUCCESS(status);
265 TEST_ASSERT_EQ(0, position);
269 static UnicodeString parseHex(const UnicodeString &in) {
270 // Convert a series of hex numbers in a Unicode String to a string with the
271 // corresponding characters.
272 // The conversion is _really_ annoying. There must be some function to just do it.
273 UnicodeString result;
275 for (int32_t i=0; i<in.length(); i++) {
276 UChar c = in.charAt(i);
277 if (c == 0x20) { // Space
282 } else if (c>=0x30 && c<=0x39) {
283 cc = (cc<<4) + (c - 0x30);
284 } else if ((c>=0x41 && c<=0x46) || (c>=0x61 && c<=0x66)) {
285 cc = (cc<<4) + (c & 0x0f)+9;
287 // else do something with bad input.
297 // Append the hex form of a UChar32 to a UnicodeString.
298 // Used in formatting error messages.
299 // Match the formatting of numbers in confusables.txt
300 // Minimum of 4 digits, no leading zeroes for positions 5 and up.
302 static void appendHexUChar(UnicodeString &dest, UChar32 c) {
303 UBool doZeroes = FALSE;
304 for (int bitNum=28; bitNum>=0; bitNum-=4) {
308 int hexDigit = (c>>bitNum) & 0x0f;
309 if (hexDigit != 0 || doZeroes) {
311 dest.append((UChar)(hexDigit<=9? hexDigit + 0x30: hexDigit -10 + 0x41));
314 dest.append((UChar)0x20);
317 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
319 // testConfData - Check each data item from the Unicode confusables.txt file,
320 // verify that it transforms correctly in a skeleton.
322 void IntlTestSpoof::testConfData() {
324 if (getUnidataPath(buffer) == NULL) {
325 errln("Skipping test spoof/testConfData. Unable to find path to source/data/unidata/.");
328 uprv_strcat(buffer, "confusables.txt");
330 LocalStdioFilePointer f(fopen(buffer, "rb"));
332 errln("Skipping test spoof/testConfData. File confusables.txt not accessible.");
335 fseek(f.getAlias(), 0, SEEK_END);
336 int32_t fileSize = ftell(f.getAlias());
337 LocalArray<char> fileBuf(new char[fileSize]);
338 fseek(f.getAlias(), 0, SEEK_SET);
339 int32_t amt_read = fread(fileBuf.getAlias(), 1, fileSize, f.getAlias());
340 TEST_ASSERT_EQ(amt_read, fileSize);
341 TEST_ASSERT(fileSize>0);
342 if (amt_read != fileSize || fileSize <=0) {
345 UnicodeString confusablesTxt = UnicodeString::fromUTF8(StringPiece(fileBuf.getAlias(), fileSize));
347 UErrorCode status = U_ZERO_ERROR;
348 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
349 TEST_ASSERT_SUCCESS(status);
351 // Parse lines from the confusables.txt file. Example Line:
352 // FF44 ; 0064 ; SL # ( d -> d ) FULLWIDTH ....
353 // Three fields. The hex fields can contain more than one character,
354 // and each character may be more than 4 digits (for supplemntals)
355 // This regular expression matches lines and splits the fields into capture groups.
356 RegexMatcher parseLine("(?m)^([0-9A-F]{4}[^#;]*?);([^#;]*?);([^#]*)", confusablesTxt, 0, status);
357 TEST_ASSERT_SUCCESS(status);
358 while (parseLine.find()) {
359 UnicodeString from = parseHex(parseLine.group(1, status));
360 if (!Normalizer::isNormalized(from, UNORM_NFD, status)) {
361 // The source character was not NFD.
362 // Skip this case; the first step in obtaining a skeleton is to NFD the input,
363 // so the mapping in this line of confusables.txt will never be applied.
367 UnicodeString rawExpected = parseHex(parseLine.group(2, status));
368 UnicodeString expected;
369 Normalizer::decompose(rawExpected, FALSE /*NFD*/, 0, expected, status);
370 TEST_ASSERT_SUCCESS(status);
372 int32_t skeletonType = 0;
373 UnicodeString tableType = parseLine.group(3, status);
374 TEST_ASSERT_SUCCESS(status);
375 if (tableType.indexOf("SL") >= 0) {
376 skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
377 } else if (tableType.indexOf("SA") >= 0) {
378 skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
379 } else if (tableType.indexOf("ML") >= 0) {
381 } else if (tableType.indexOf("MA") >= 0) {
382 skeletonType = USPOOF_ANY_CASE;
385 UnicodeString actual;
386 uspoof_getSkeletonUnicodeString(sc.getAlias(), skeletonType, from, actual, &status);
387 TEST_ASSERT_SUCCESS(status);
388 TEST_ASSERT(actual == expected);
389 if (actual != expected) {
390 errln(parseLine.group(0, status));
391 UnicodeString line = "Actual: ";
393 while (i < actual.length()) {
394 appendHexUChar(line, actual.char32At(i));
395 i = actual.moveIndex32(i, 1);
399 if (U_FAILURE(status)) {
406 void IntlTestSpoof::testScriptSet() {
409 UErrorCode status = U_ZERO_ERROR;
411 TEST_ASSERT(s1 == s2);
412 s1.set(USCRIPT_ARABIC,status);
413 TEST_ASSERT_SUCCESS(status);
414 TEST_ASSERT(!(s1 == s2));
415 TEST_ASSERT(s1.test(USCRIPT_ARABIC, status));
416 TEST_ASSERT(s1.test(USCRIPT_GREEK, status) == FALSE);
418 status = U_ZERO_ERROR;
419 s1.reset(USCRIPT_ARABIC, status);
420 TEST_ASSERT(s1 == s2);
422 status = U_ZERO_ERROR;
424 TEST_ASSERT(s1.test(USCRIPT_COMMON, status));
425 TEST_ASSERT(s1.test(USCRIPT_ETHIOPIC, status));
426 TEST_ASSERT(s1.test(USCRIPT_CODE_LIMIT, status));
428 TEST_ASSERT(!s1.test(USCRIPT_COMMON, status));
429 TEST_ASSERT(!s1.test(USCRIPT_ETHIOPIC, status));
430 TEST_ASSERT(!s1.test(USCRIPT_CODE_LIMIT, status));
432 status = U_ZERO_ERROR;
433 s1.set(USCRIPT_TAKRI, status);
434 s1.set(USCRIPT_BLISSYMBOLS, status);
436 TEST_ASSERT(s2.contains(s1));
437 TEST_ASSERT(!s1.contains(s2));
438 TEST_ASSERT(s2.intersects(s1));
439 TEST_ASSERT(s1.intersects(s2));
440 s2.reset(USCRIPT_TAKRI, status);
441 TEST_ASSERT(!s2.contains(s1));
442 TEST_ASSERT(!s1.contains(s2));
443 TEST_ASSERT(s1.intersects(s2));
444 TEST_ASSERT(s2.intersects(s1));
445 TEST_ASSERT_SUCCESS(status);
447 status = U_ZERO_ERROR;
449 s1.set(USCRIPT_NKO, status);
450 s1.set(USCRIPT_COMMON, status);
452 TEST_ASSERT(s2 == s1);
453 TEST_ASSERT_EQ(2, s2.countMembers());
455 TEST_ASSERT(s2 == s1);
457 TEST_ASSERT(!(s2 == s1));
458 TEST_ASSERT(s2.countMembers() >= USCRIPT_CODE_LIMIT);
460 TEST_ASSERT(s2 == s1);
463 s2.reset(USCRIPT_COMMON, status);
465 TEST_ASSERT(s2.countMembers() == 1);
468 TEST_ASSERT(s1.isEmpty());
469 s1.set(USCRIPT_LATIN, status);
470 TEST_ASSERT(!s1.isEmpty());
472 TEST_ASSERT(!s1.isEmpty());
473 TEST_ASSERT_SUCCESS(status);
476 s1.set(USCRIPT_AFAKA, status);
477 s1.set(USCRIPT_VAI, status);
478 s1.set(USCRIPT_INHERITED, status);
480 for (int32_t i=0; i<4; i++) {
481 n = s1.nextSetBit(n+1);
483 case 0: TEST_ASSERT_EQ(USCRIPT_INHERITED, n); break;
484 case 1: TEST_ASSERT_EQ(USCRIPT_VAI, n); break;
485 case 2: TEST_ASSERT_EQ(USCRIPT_AFAKA, n); break;
486 case 3: TEST_ASSERT_EQ(-1, (int32_t)n); break;
487 default: TEST_ASSERT(FALSE);
490 TEST_ASSERT_SUCCESS(status);
492 // Script extensions. Depends on data.
494 s1.setScriptExtensions(0x67, status);
495 TEST_ASSERT(s1.countMembers() == 1);
496 TEST_ASSERT(s1.test(USCRIPT_LATIN, status));
497 TEST_ASSERT_SUCCESS(status);
500 s1.setScriptExtensions(0x303C, status);
501 TEST_ASSERT(s1.countMembers() == 3);
502 TEST_ASSERT(s1.test(USCRIPT_HAN, status));
503 TEST_ASSERT(s1.test(USCRIPT_HIRAGANA, status));
504 TEST_ASSERT(s1.test(USCRIPT_KATAKANA, status));
505 TEST_ASSERT_SUCCESS(status);
508 ScriptSet bitset12; bitset12.set(USCRIPT_LATIN, status).set(USCRIPT_HANGUL, status);
509 ScriptSet bitset2; bitset2.set(USCRIPT_HANGUL, status);
510 TEST_ASSERT(bitset12.contains(bitset2));
511 TEST_ASSERT(bitset12.contains(bitset12));
512 TEST_ASSERT(!bitset2.contains(bitset12));
514 ScriptSet arabSet; arabSet.set(USCRIPT_ARABIC, status);
515 ScriptSet latinSet; latinSet.set(USCRIPT_LATIN, status);
516 UElement arabEl; arabEl.pointer = &arabSet;
517 UElement latinEl; latinEl.pointer = &latinSet;
518 TEST_ASSERT(uhash_compareScriptSet(arabEl, latinEl) < 0);
519 TEST_ASSERT(uhash_compareScriptSet(latinEl, arabEl) > 0);
521 UnicodeString scriptString;
522 bitset12.displayScripts(scriptString);
523 TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang Latn") == scriptString);
527 void IntlTestSpoof::testRestrictionLevel() {
530 URestrictionLevel fExpectedRestrictionLevel;
532 {"\\u0061\\u03B3\\u2665", USPOOF_UNRESTRICTIVE},
534 {"\\u03B3", USPOOF_SINGLE_SCRIPT_RESTRICTIVE},
535 {"\\u0061\\u30A2\\u30FC", USPOOF_HIGHLY_RESTRICTIVE},
536 {"\\u0061\\u0904", USPOOF_MODERATELY_RESTRICTIVE},
537 {"\\u0061\\u03B3", USPOOF_MINIMALLY_RESTRICTIVE},
538 {"\\u0061\\u2665", USPOOF_UNRESTRICTIVE},
539 {"\\u0061\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
540 {"\\u0061\\u30FC\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
541 {"\\u0061\\u30FC\\u303C\\u30A2", USPOOF_HIGHLY_RESTRICTIVE},
542 {"\\u30A2\\u0061\\u30FC\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
543 {"\\u0061\\u0031\\u0661", USPOOF_MODERATELY_RESTRICTIVE},
544 {"\\u0061\\u0031\\u0661\\u06F1", USPOOF_MODERATELY_RESTRICTIVE},
545 {"\\u0661\\u30FC\\u303C\\u0061\\u30A2\\u0031\\u0967\\u06F1", USPOOF_MINIMALLY_RESTRICTIVE},
546 {"\\u0061\\u30A2\\u30FC\\u303C\\u0031\\u0967\\u0661\\u06F1", USPOOF_MINIMALLY_RESTRICTIVE}
549 URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_SINGLE_SCRIPT_RESTRICTIVE,
550 USPOOF_HIGHLY_RESTRICTIVE, USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE,
551 USPOOF_UNRESTRICTIVE};
553 UErrorCode status = U_ZERO_ERROR;
554 UnicodeSet allowedChars;
555 // Allowed Identifier Characters. In addition to the Recommended Set,
556 // allow u303c, which has an interesting script extension of Hani Hira Kana.
557 allowedChars.addAll(*uspoof_getRecommendedUnicodeSet(&status)).add(0x303C);
559 for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
560 status = U_ZERO_ERROR;
561 const Test &test = tests[testNum];
562 UnicodeString testString = UnicodeString(test.fId).unescape();
563 URestrictionLevel expectedLevel = test.fExpectedRestrictionLevel;
564 for (int levelIndex=0; levelIndex<UPRV_LENGTHOF(restrictionLevels); levelIndex++) {
565 status = U_ZERO_ERROR;
566 URestrictionLevel levelSetInSpoofChecker = restrictionLevels[levelIndex];
567 USpoofChecker *sc = uspoof_open(&status);
568 uspoof_setAllowedChars(sc, allowedChars.toUSet(), &status);
569 uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
570 uspoof_setChecks(sc, USPOOF_RESTRICTION_LEVEL, &status);
571 int32_t actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status);
573 // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
574 int32_t expectedValue = 0;
575 if (expectedLevel > levelSetInSpoofChecker) {
576 expectedValue |= USPOOF_RESTRICTION_LEVEL;
578 sprintf(msgBuffer, "testNum = %d, levelIndex = %d, expected = %#x, actual = %#x",
579 testNum, levelIndex, expectedValue, actualValue);
580 TEST_ASSERT_MSG(expectedValue == actualValue, msgBuffer);
581 TEST_ASSERT_SUCCESS(status);
583 // Run the same check again, with the Spoof Checker configured to return
584 // the actual restriction level.
585 uspoof_setAllowedChars(sc, allowedChars.toUSet(), &status);
586 uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
587 uspoof_setChecks(sc, USPOOF_AUX_INFO | USPOOF_RESTRICTION_LEVEL, &status);
588 int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
589 TEST_ASSERT_SUCCESS(status);
590 if (U_SUCCESS(status)) {
591 TEST_ASSERT_EQ(expectedLevel, result & USPOOF_RESTRICTION_LEVEL_MASK);
592 TEST_ASSERT_EQ(expectedValue, result & USPOOF_ALL_CHECKS);
600 void IntlTestSpoof::testMixedNumbers() {
602 const char *fTestString;
603 const char *fExpectedSet;
606 {"\\u0967", "[\\u0966]"},
607 {"1\\u0967", "[0\\u0966]"},
608 {"\\u0661\\u06F1", "[\\u0660\\u06F0]"},
609 {"\\u0061\\u2665", "[]"},
610 {"\\u0061\\u303C", "[]"},
611 {"\\u0061\\u30FC\\u303C", "[]"},
612 {"\\u0061\\u30FC\\u303C\\u30A2", "[]"},
613 {"\\u30A2\\u0061\\u30FC\\u303C", "[]"},
614 {"\\u0061\\u0031\\u0661", "[\\u0030\\u0660]"},
615 {"\\u0061\\u0031\\u0661\\u06F1", "[\\u0030\\u0660\\u06F0]"},
616 {"\\u0661\\u30FC\\u303C\\u0061\\u30A2\\u0031\\u0967\\u06F1", "[\\u0030\\u0660\\u06F0\\u0966]"},
617 {"\\u0061\\u30A2\\u30FC\\u303C\\u0031\\u0967\\u0661\\u06F1", "[\\u0030\\u0660\\u06F0\\u0966]"}
619 UErrorCode status = U_ZERO_ERROR;
620 for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
622 sprintf(msgBuf, "testNum = %d ", testNum);
623 Test &test = tests[testNum];
625 status = U_ZERO_ERROR;
626 UnicodeString testString = UnicodeString(test.fTestString).unescape();
627 UnicodeSet expectedSet(UnicodeString(test.fExpectedSet).unescape(), status);
629 status = U_ZERO_ERROR;
631 uspoof_setChecks(sc, USPOOF_MIXED_NUMBERS, &status); // only check this
632 uspoof_check2UnicodeString(sc, testString, checkResult, &status);
633 UBool mixedNumberFailure = ((uspoof_getCheckResultChecks(checkResult, &status) & USPOOF_MIXED_NUMBERS) != 0);
634 TEST_ASSERT_MSG((expectedSet.size() > 1) == mixedNumberFailure, msgBuf);
635 const UnicodeSet* actualSet = UnicodeSet::fromUSet(uspoof_getCheckResultNumerics(checkResult, &status));
636 TEST_ASSERT_MSG(expectedSet == *actualSet, msgBuf);
641 // Bug #12153 - uspoof_setRestrictionLevel() should enable restriction level testing.
643 void IntlTestSpoof::testBug12153() {
644 UErrorCode status = U_ZERO_ERROR;
645 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
646 TEST_ASSERT_SUCCESS(status);
647 int32_t checks = uspoof_getChecks(sc.getAlias(), &status);
648 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) != 0);
649 checks &= ~USPOOF_RESTRICTION_LEVEL;
650 uspoof_setChecks(sc.getAlias(), checks, &status);
651 checks = uspoof_getChecks(sc.getAlias(), &status);
652 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) == 0);
654 uspoof_setRestrictionLevel(sc.getAlias(), USPOOF_MODERATELY_RESTRICTIVE);
655 checks = uspoof_getChecks(sc.getAlias(), &status);
656 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) != 0);
657 TEST_ASSERT_SUCCESS(status);
660 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO */