1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 **********************************************************************
5 * Copyright (C) 2011-2015, International Business Machines Corporation
6 * and others. All Rights Reserved.
7 **********************************************************************
8 */
9 /**
10 * IntlTestSpoof tests for USpoofDetector
11 */
12
13 #include "unicode/utypes.h"
14
15 #if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO
16
17 #include "itspoof.h"
18
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"
24
25 #include "cstring.h"
26 #include "scriptset.h"
27 #include "uhash.h"
28
29 #include <stdlib.h>
30 #include <stdio.h>
31
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));}}
34
35 #define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
36 errln("Test Failure at file %s, line %d: \"%s\" is false.", __FILE__, __LINE__, #expr);};}
37
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);};}
40
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)); }}
44
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)); }}
48
49 /*
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.
54 */
55 #define TEST_SETUP { \
56 UErrorCode status = U_ZERO_ERROR; \
57 USpoofChecker *sc; \
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)){
64
65 #define TEST_TEARDOWN \
66 } \
67 TEST_ASSERT_SUCCESS(status); \
68 uspoof_closeCheckResult(checkResult); \
69 uspoof_close(sc); \
70 }
71
72
73
74
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)75 void IntlTestSpoof::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
76 {
77 if (exec) {
78 logln("TestSuite spoof: ");
79 }
80 TESTCASE_AUTO_BEGIN;
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);
91 TESTCASE_AUTO(testBug12825);
92 TESTCASE_AUTO(testBug12815);
93 TESTCASE_AUTO(testBug13314_MixedNumbers);
94 TESTCASE_AUTO(testBug13328_MixedCombiningMarks);
95 TESTCASE_AUTO(testCombiningDot);
96 TESTCASE_AUTO_END;
97 }
98
testSpoofAPI()99 void IntlTestSpoof::testSpoofAPI() {
100
101 TEST_SETUP
102 UnicodeString s("xyz"); // Many latin ranges are whole-script confusable with other scripts.
103 // If this test starts failing, consult confusablesWholeScript.txt
104 int32_t position = 666;
105 int32_t checkResults = uspoof_checkUnicodeString(sc, s, &position, &status);
106 TEST_ASSERT_SUCCESS(status);
107 TEST_ASSERT_EQ(0, checkResults);
108 TEST_ASSERT_EQ(0, position);
109 TEST_TEARDOWN;
110
111 TEST_SETUP
112 UnicodeString s1("cxs");
113 UnicodeString s2 = UnicodeString("\\u0441\\u0445\\u0455").unescape(); // Cyrillic "cxs"
114 int32_t checkResults = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
115 TEST_ASSERT_SUCCESS(status);
116 TEST_ASSERT_EQ(USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE, checkResults);
117
118 TEST_TEARDOWN;
119
120 TEST_SETUP
121 UnicodeString s("I1l0O");
122 UnicodeString dest;
123 UnicodeString &retStr = uspoof_getSkeletonUnicodeString(sc, USPOOF_ANY_CASE, s, dest, &status);
124 TEST_ASSERT_SUCCESS(status);
125 TEST_ASSERT(UnicodeString("lllOO") == dest);
126 TEST_ASSERT(&dest == &retStr);
127 TEST_TEARDOWN;
128 }
129
130
131 #define CHECK_SKELETON(type, input, expected) { \
132 checkSkeleton(sc, type, input, expected, __LINE__); \
133 }
134
135
136 // testSkeleton. Spot check a number of confusable skeleton substitutions from the
137 // Unicode data file confusables.txt
138 // Test cases chosen for substitutions of various lengths, and
139 // membership in different mapping tables.
140 // Note: for ICU 55, all tables collapsed to the MA table data.
141 // TODO: for ICU 56 with Unicode 8, revisit this test.
142 //
testSkeleton()143 void IntlTestSpoof::testSkeleton() {
144 const uint32_t ML = 0;
145 const uint32_t SL = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
146 const uint32_t MA = USPOOF_ANY_CASE;
147 const uint32_t SA = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
148
149 TEST_SETUP
150 CHECK_SKELETON(SL, "nochange", "nochange");
151 CHECK_SKELETON(SA, "nochange", "nochange");
152 CHECK_SKELETON(ML, "nochange", "nochange");
153 CHECK_SKELETON(MA, "nochange", "nochange");
154 CHECK_SKELETON(MA, "love", "love");
155 CHECK_SKELETON(MA, "1ove", "love"); // Digit 1 to letter l
156 CHECK_SKELETON(ML, "OOPS", "OOPS");
157 CHECK_SKELETON(ML, "00PS", "OOPS");
158 CHECK_SKELETON(MA, "OOPS", "OOPS");
159 CHECK_SKELETON(MA, "00PS", "OOPS"); // Digit 0 to letter O in any case mode only
160 CHECK_SKELETON(SL, "\\u059c", "\\u0301");
161 CHECK_SKELETON(SL, "\\u2A74", "\\u003A\\u003A\\u003D");
162 CHECK_SKELETON(SL, "\\u247E", "\\u0028\\u006C\\u006C\\u0029"); // "(ll)"
163 CHECK_SKELETON(SL, "\\uFDFB", "\\u062C\\u0644\\u0020\\u062C\\u0644\\u006c\\u0644\\u006f");
164
165 // This mapping exists in the ML and MA tables, does not exist in SL, SA
166 // 0C83 ; 0983 ; ML
167 // 0C83 ; 0983 ; MA
168 //
169
170 CHECK_SKELETON(SL, "\\u0C83", "\\u0983");
171 CHECK_SKELETON(SA, "\\u0C83", "\\u0983");
172 CHECK_SKELETON(ML, "\\u0C83", "\\u0983");
173 CHECK_SKELETON(MA, "\\u0C83", "\\u0983");
174
175 // 0391 mappings exist only in MA and SA tables.
176 CHECK_SKELETON(MA, "\\u0391", "A");
177 CHECK_SKELETON(SA, "\\u0391", "A");
178 CHECK_SKELETON(ML, "\\u0391", "A");
179 CHECK_SKELETON(SL, "\\u0391", "A");
180
181 // 13CF Mappings in all four tables, different in MA.
182 CHECK_SKELETON(ML, "\\u13CF", "b");
183 CHECK_SKELETON(MA, "\\u13CF", "b");
184 CHECK_SKELETON(SL, "\\u13CF", "b");
185 CHECK_SKELETON(SA, "\\u13CF", "b");
186
187 // 0022 ; 0027 0027 ;
188 // all tables.
189 CHECK_SKELETON(SL, "\\u0022", "\\u0027\\u0027");
190 CHECK_SKELETON(SA, "\\u0022", "\\u0027\\u0027");
191 CHECK_SKELETON(ML, "\\u0022", "\\u0027\\u0027");
192 CHECK_SKELETON(MA, "\\u0022", "\\u0027\\u0027");
193
194 // 017F mappings exist only in MA and SA tables.
195 CHECK_SKELETON(MA, "\\u017F", "f");
196 CHECK_SKELETON(SA, "\\u017F", "f");
197 CHECK_SKELETON(ML, "\\u017F", "f");
198 CHECK_SKELETON(SL, "\\u017F", "f");
199
200 TEST_TEARDOWN;
201 }
202
203
204 //
205 // Run a single confusable skeleton transformation test case.
206 //
checkSkeleton(const USpoofChecker * sc,uint32_t type,const char * input,const char * expected,int32_t lineNum)207 void IntlTestSpoof::checkSkeleton(const USpoofChecker *sc, uint32_t type,
208 const char *input, const char *expected, int32_t lineNum) {
209 UnicodeString uInput = UnicodeString(input).unescape();
210 UnicodeString uExpected = UnicodeString(expected).unescape();
211
212 UErrorCode status = U_ZERO_ERROR;
213 UnicodeString actual;
214 uspoof_getSkeletonUnicodeString(sc, type, uInput, actual, &status);
215 if (U_FAILURE(status)) {
216 errln("File %s, Line %d, Test case from line %d, status is %s", __FILE__, __LINE__, lineNum,
217 u_errorName(status));
218 return;
219 }
220 if (uExpected != actual) {
221 errln("File %s, Line %d, Test case from line %d, Actual and Expected skeletons differ.",
222 __FILE__, __LINE__, lineNum);
223 errln(UnicodeString(" Actual Skeleton: \"") + actual + UnicodeString("\"\n") +
224 UnicodeString(" Expected Skeleton: \"") + uExpected + UnicodeString("\""));
225 }
226 }
227
testAreConfusable()228 void IntlTestSpoof::testAreConfusable() {
229 TEST_SETUP
230 UnicodeString s1("A long string that will overflow stack buffers. A long string that will overflow stack buffers. "
231 "A long string that will overflow stack buffers. A long string that will overflow stack buffers. ");
232 UnicodeString s2("A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. "
233 "A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. ");
234 int32_t result = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
235 TEST_ASSERT_SUCCESS(status);
236 TEST_ASSERT_EQ(USPOOF_SINGLE_SCRIPT_CONFUSABLE, result);
237
238 TEST_TEARDOWN;
239 }
240
testInvisible()241 void IntlTestSpoof::testInvisible() {
242 TEST_SETUP
243 UnicodeString s = UnicodeString("abcd\\u0301ef").unescape();
244 int32_t position = -42;
245 TEST_ASSERT_EQ(0, uspoof_checkUnicodeString(sc, s, &position, &status));
246 TEST_ASSERT_SUCCESS(status);
247 TEST_ASSERT(0 == position);
248
249 UnicodeString s2 = UnicodeString("abcd\\u0301\\u0302\\u0301ef").unescape();
250 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s2, &position, &status));
251 TEST_ASSERT_SUCCESS(status);
252 TEST_ASSERT_EQ(0, position);
253
254 // Two acute accents, one from the composed a with acute accent, \u00e1,
255 // and one separate.
256 position = -42;
257 UnicodeString s3 = UnicodeString("abcd\\u00e1\\u0301xyz").unescape();
258 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s3, &position, &status));
259 TEST_ASSERT_SUCCESS(status);
260 TEST_ASSERT_EQ(0, position);
261 TEST_TEARDOWN;
262 }
263
testBug8654()264 void IntlTestSpoof::testBug8654() {
265 TEST_SETUP
266 UnicodeString s = UnicodeString("B\\u00c1\\u0301").unescape();
267 int32_t position = -42;
268 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s, &position, &status) & USPOOF_INVISIBLE );
269 TEST_ASSERT_SUCCESS(status);
270 TEST_ASSERT_EQ(0, position);
271 TEST_TEARDOWN;
272 }
273
parseHex(const UnicodeString & in)274 static UnicodeString parseHex(const UnicodeString &in) {
275 // Convert a series of hex numbers in a Unicode String to a string with the
276 // corresponding characters.
277 // The conversion is _really_ annoying. There must be some function to just do it.
278 UnicodeString result;
279 UChar32 cc = 0;
280 for (int32_t i=0; i<in.length(); i++) {
281 UChar c = in.charAt(i);
282 if (c == 0x20) { // Space
283 if (cc > 0) {
284 result.append(cc);
285 cc = 0;
286 }
287 } else if (c>=0x30 && c<=0x39) {
288 cc = (cc<<4) + (c - 0x30);
289 } else if ((c>=0x41 && c<=0x46) || (c>=0x61 && c<=0x66)) {
290 cc = (cc<<4) + (c & 0x0f)+9;
291 }
292 // else do something with bad input.
293 }
294 if (cc > 0) {
295 result.append(cc);
296 }
297 return result;
298 }
299
300
301 //
302 // Append the hex form of a UChar32 to a UnicodeString.
303 // Used in formatting error messages.
304 // Match the formatting of numbers in confusables.txt
305 // Minimum of 4 digits, no leading zeroes for positions 5 and up.
306 //
appendHexUChar(UnicodeString & dest,UChar32 c)307 static void appendHexUChar(UnicodeString &dest, UChar32 c) {
308 UBool doZeroes = FALSE;
309 for (int bitNum=28; bitNum>=0; bitNum-=4) {
310 if (bitNum <= 12) {
311 doZeroes = TRUE;
312 }
313 int hexDigit = (c>>bitNum) & 0x0f;
314 if (hexDigit != 0 || doZeroes) {
315 doZeroes = TRUE;
316 dest.append((UChar)(hexDigit<=9? hexDigit + 0x30: hexDigit -10 + 0x41));
317 }
318 }
319 dest.append((UChar)0x20);
320 }
321
322 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
323
324 // testConfData - Check each data item from the Unicode confusables.txt file,
325 // verify that it transforms correctly in a skeleton.
326 //
testConfData()327 void IntlTestSpoof::testConfData() {
328 char buffer[2000];
329 if (getUnidataPath(buffer) == NULL) {
330 errln("Skipping test spoof/testConfData. Unable to find path to source/data/unidata/.");
331 return;
332 }
333 uprv_strcat(buffer, "confusables.txt");
334
335 LocalStdioFilePointer f(fopen(buffer, "rb"));
336 if (f.isNull()) {
337 errln("Skipping test spoof/testConfData. File confusables.txt not accessible.");
338 return;
339 }
340 fseek(f.getAlias(), 0, SEEK_END);
341 int32_t fileSize = ftell(f.getAlias());
342 LocalArray<char> fileBuf(new char[fileSize]);
343 fseek(f.getAlias(), 0, SEEK_SET);
344 int32_t amt_read = fread(fileBuf.getAlias(), 1, fileSize, f.getAlias());
345 TEST_ASSERT_EQ(amt_read, fileSize);
346 TEST_ASSERT(fileSize>0);
347 if (amt_read != fileSize || fileSize <=0) {
348 return;
349 }
350 UnicodeString confusablesTxt = UnicodeString::fromUTF8(StringPiece(fileBuf.getAlias(), fileSize));
351
352 UErrorCode status = U_ZERO_ERROR;
353 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
354 TEST_ASSERT_SUCCESS(status);
355
356 // Parse lines from the confusables.txt file. Example Line:
357 // FF44 ; 0064 ; SL # ( d -> d ) FULLWIDTH ....
358 // Three fields. The hex fields can contain more than one character,
359 // and each character may be more than 4 digits (for supplemntals)
360 // This regular expression matches lines and splits the fields into capture groups.
361 RegexMatcher parseLine("(?m)^([0-9A-F]{4}[^#;]*?);([^#;]*?);([^#]*)", confusablesTxt, 0, status);
362 TEST_ASSERT_SUCCESS(status);
363 while (parseLine.find()) {
364 UnicodeString from = parseHex(parseLine.group(1, status));
365 if (!Normalizer::isNormalized(from, UNORM_NFD, status)) {
366 // The source character was not NFD.
367 // Skip this case; the first step in obtaining a skeleton is to NFD the input,
368 // so the mapping in this line of confusables.txt will never be applied.
369 continue;
370 }
371
372 UnicodeString rawExpected = parseHex(parseLine.group(2, status));
373 UnicodeString expected;
374 Normalizer::decompose(rawExpected, FALSE /*NFD*/, 0, expected, status);
375 TEST_ASSERT_SUCCESS(status);
376
377 int32_t skeletonType = 0;
378 UnicodeString tableType = parseLine.group(3, status);
379 TEST_ASSERT_SUCCESS(status);
380 if (tableType.indexOf("SL") >= 0) {
381 skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
382 } else if (tableType.indexOf("SA") >= 0) {
383 skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
384 } else if (tableType.indexOf("ML") >= 0) {
385 skeletonType = 0;
386 } else if (tableType.indexOf("MA") >= 0) {
387 skeletonType = USPOOF_ANY_CASE;
388 }
389
390 UnicodeString actual;
391 uspoof_getSkeletonUnicodeString(sc.getAlias(), skeletonType, from, actual, &status);
392 TEST_ASSERT_SUCCESS(status);
393 TEST_ASSERT(actual == expected);
394 if (actual != expected) {
395 errln(parseLine.group(0, status));
396 UnicodeString line = "Actual: ";
397 int i = 0;
398 while (i < actual.length()) {
399 appendHexUChar(line, actual.char32At(i));
400 i = actual.moveIndex32(i, 1);
401 }
402 errln(line);
403 }
404 if (U_FAILURE(status)) {
405 break;
406 }
407 }
408 }
409
410
testScriptSet()411 void IntlTestSpoof::testScriptSet() {
412 ScriptSet s1;
413 ScriptSet s2;
414 UErrorCode status = U_ZERO_ERROR;
415
416 TEST_ASSERT(s1 == s2);
417 s1.set(USCRIPT_ARABIC,status);
418 TEST_ASSERT_SUCCESS(status);
419 TEST_ASSERT(!(s1 == s2));
420 TEST_ASSERT(s1.test(USCRIPT_ARABIC, status));
421 TEST_ASSERT(s1.test(USCRIPT_GREEK, status) == FALSE);
422
423 status = U_ZERO_ERROR;
424 s1.reset(USCRIPT_ARABIC, status);
425 TEST_ASSERT(s1 == s2);
426
427 status = U_ZERO_ERROR;
428 s1.setAll();
429 TEST_ASSERT(s1.test(USCRIPT_COMMON, status));
430 TEST_ASSERT(s1.test(USCRIPT_ETHIOPIC, status));
431 TEST_ASSERT(s1.test(USCRIPT_CODE_LIMIT, status));
432 s1.resetAll();
433 TEST_ASSERT(!s1.test(USCRIPT_COMMON, status));
434 TEST_ASSERT(!s1.test(USCRIPT_ETHIOPIC, status));
435 TEST_ASSERT(!s1.test(USCRIPT_CODE_LIMIT, status));
436
437 status = U_ZERO_ERROR;
438 s1.set(USCRIPT_TAKRI, status);
439 s1.set(USCRIPT_BLISSYMBOLS, status);
440 s2.setAll();
441 TEST_ASSERT(s2.contains(s1));
442 TEST_ASSERT(!s1.contains(s2));
443 TEST_ASSERT(s2.intersects(s1));
444 TEST_ASSERT(s1.intersects(s2));
445 s2.reset(USCRIPT_TAKRI, status);
446 TEST_ASSERT(!s2.contains(s1));
447 TEST_ASSERT(!s1.contains(s2));
448 TEST_ASSERT(s1.intersects(s2));
449 TEST_ASSERT(s2.intersects(s1));
450 TEST_ASSERT_SUCCESS(status);
451
452 status = U_ZERO_ERROR;
453 s1.resetAll();
454 s1.set(USCRIPT_NKO, status);
455 s1.set(USCRIPT_COMMON, status);
456 s2 = s1;
457 TEST_ASSERT(s2 == s1);
458 TEST_ASSERT_EQ(2, s2.countMembers());
459 s2.intersect(s1);
460 TEST_ASSERT(s2 == s1);
461 s2.setAll();
462 TEST_ASSERT(!(s2 == s1));
463 TEST_ASSERT(s2.countMembers() >= USCRIPT_CODE_LIMIT);
464 s2.intersect(s1);
465 TEST_ASSERT(s2 == s1);
466
467 s2.setAll();
468 s2.reset(USCRIPT_COMMON, status);
469 s2.intersect(s1);
470 TEST_ASSERT(s2.countMembers() == 1);
471
472 s1.resetAll();
473 TEST_ASSERT(s1.isEmpty());
474 s1.set(USCRIPT_LATIN, status);
475 TEST_ASSERT(!s1.isEmpty());
476 s1.setAll();
477 TEST_ASSERT(!s1.isEmpty());
478 TEST_ASSERT_SUCCESS(status);
479
480 s1.resetAll();
481 s1.set(USCRIPT_AFAKA, status);
482 s1.set(USCRIPT_VAI, status);
483 s1.set(USCRIPT_INHERITED, status);
484 int32_t n = -1;
485 for (int32_t i=0; i<4; i++) {
486 n = s1.nextSetBit(n+1);
487 switch (i) {
488 case 0: TEST_ASSERT_EQ(USCRIPT_INHERITED, n); break;
489 case 1: TEST_ASSERT_EQ(USCRIPT_VAI, n); break;
490 case 2: TEST_ASSERT_EQ(USCRIPT_AFAKA, n); break;
491 case 3: TEST_ASSERT_EQ(-1, (int32_t)n); break;
492 default: TEST_ASSERT(FALSE);
493 }
494 }
495 TEST_ASSERT_SUCCESS(status);
496
497 // Script extensions. Depends on data.
498 s1.resetAll();
499 s1.setScriptExtensions(0x67, status);
500 TEST_ASSERT(s1.countMembers() == 1);
501 TEST_ASSERT(s1.test(USCRIPT_LATIN, status));
502 TEST_ASSERT_SUCCESS(status);
503
504 s1.resetAll();
505 s1.setScriptExtensions(0x303C, status);
506 TEST_ASSERT(s1.countMembers() == 3);
507 TEST_ASSERT(s1.test(USCRIPT_HAN, status));
508 TEST_ASSERT(s1.test(USCRIPT_HIRAGANA, status));
509 TEST_ASSERT(s1.test(USCRIPT_KATAKANA, status));
510 TEST_ASSERT_SUCCESS(status);
511
512 // Additional tests
513 ScriptSet bitset12; bitset12.set(USCRIPT_LATIN, status).set(USCRIPT_HANGUL, status);
514 ScriptSet bitset2; bitset2.set(USCRIPT_HANGUL, status);
515 TEST_ASSERT(bitset12.contains(bitset2));
516 TEST_ASSERT(bitset12.contains(bitset12));
517 TEST_ASSERT(!bitset2.contains(bitset12));
518
519 ScriptSet arabSet; arabSet.set(USCRIPT_ARABIC, status);
520 ScriptSet latinSet; latinSet.set(USCRIPT_LATIN, status);
521 UElement arabEl; arabEl.pointer = &arabSet;
522 UElement latinEl; latinEl.pointer = &latinSet;
523 TEST_ASSERT(uhash_compareScriptSet(arabEl, latinEl) < 0);
524 TEST_ASSERT(uhash_compareScriptSet(latinEl, arabEl) > 0);
525
526 UnicodeString scriptString;
527 bitset12.displayScripts(scriptString);
528 TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang Latn") == scriptString);
529 }
530
531
testRestrictionLevel()532 void IntlTestSpoof::testRestrictionLevel() {
533 struct Test {
534 const char *fId;
535 URestrictionLevel fExpectedRestrictionLevel;
536 } tests[] = {
537 {"\\u0061\\u03B3\\u2665", USPOOF_UNRESTRICTIVE},
538 {"a", USPOOF_ASCII},
539 {"\\u03B3", USPOOF_SINGLE_SCRIPT_RESTRICTIVE},
540 {"\\u0061\\u30A2\\u30FC", USPOOF_HIGHLY_RESTRICTIVE},
541 {"\\u0061\\u0904", USPOOF_MODERATELY_RESTRICTIVE},
542 {"\\u0061\\u03B3", USPOOF_MINIMALLY_RESTRICTIVE},
543 {"\\u0061\\u2665", USPOOF_UNRESTRICTIVE},
544 {"\\u0061\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
545 {"\\u0061\\u30FC\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
546 {"\\u0061\\u30FC\\u303C\\u30A2", USPOOF_HIGHLY_RESTRICTIVE},
547 {"\\u30A2\\u0061\\u30FC\\u303C", USPOOF_HIGHLY_RESTRICTIVE},
548 {"\\u0061\\u0031\\u0661", USPOOF_MODERATELY_RESTRICTIVE},
549 {"\\u0061\\u0031\\u0661\\u06F1", USPOOF_MODERATELY_RESTRICTIVE},
550 {"\\u0661\\u30FC\\u303C\\u0061\\u30A2\\u0031\\u0967\\u06F1", USPOOF_MINIMALLY_RESTRICTIVE},
551 {"\\u0061\\u30A2\\u30FC\\u303C\\u0031\\u0967\\u0661\\u06F1", USPOOF_MINIMALLY_RESTRICTIVE}
552 };
553 char msgBuffer[100];
554 URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_SINGLE_SCRIPT_RESTRICTIVE,
555 USPOOF_HIGHLY_RESTRICTIVE, USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE,
556 USPOOF_UNRESTRICTIVE};
557
558 UErrorCode status = U_ZERO_ERROR;
559 UnicodeSet allowedChars;
560 // Allowed Identifier Characters. In addition to the Recommended Set,
561 // allow u303c, which has an interesting script extension of Hani Hira Kana.
562 allowedChars.addAll(*uspoof_getRecommendedUnicodeSet(&status)).add(0x303C);
563
564 for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
565 status = U_ZERO_ERROR;
566 const Test &test = tests[testNum];
567 UnicodeString testString = UnicodeString(test.fId).unescape();
568 URestrictionLevel expectedLevel = test.fExpectedRestrictionLevel;
569 for (int levelIndex=0; levelIndex<UPRV_LENGTHOF(restrictionLevels); levelIndex++) {
570 status = U_ZERO_ERROR;
571 URestrictionLevel levelSetInSpoofChecker = restrictionLevels[levelIndex];
572 USpoofChecker *sc = uspoof_open(&status);
573 uspoof_setAllowedChars(sc, allowedChars.toUSet(), &status);
574 uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
575 uspoof_setChecks(sc, USPOOF_RESTRICTION_LEVEL, &status);
576 int32_t actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status);
577
578 // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
579 int32_t expectedValue = 0;
580 if (expectedLevel > levelSetInSpoofChecker) {
581 expectedValue |= USPOOF_RESTRICTION_LEVEL;
582 }
583 sprintf(msgBuffer, "testNum = %d, levelIndex = %d, expected = %#x, actual = %#x",
584 testNum, levelIndex, expectedValue, actualValue);
585 TEST_ASSERT_MSG(expectedValue == actualValue, msgBuffer);
586 TEST_ASSERT_SUCCESS(status);
587
588 // Run the same check again, with the Spoof Checker configured to return
589 // the actual restriction level.
590 uspoof_setAllowedChars(sc, allowedChars.toUSet(), &status);
591 uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
592 uspoof_setChecks(sc, USPOOF_AUX_INFO | USPOOF_RESTRICTION_LEVEL, &status);
593 int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
594 TEST_ASSERT_SUCCESS(status);
595 if (U_SUCCESS(status)) {
596 TEST_ASSERT_EQ(expectedLevel, result & USPOOF_RESTRICTION_LEVEL_MASK);
597 TEST_ASSERT_EQ(expectedValue, result & USPOOF_ALL_CHECKS);
598 }
599 uspoof_close(sc);
600 }
601 }
602
603 }
604
testMixedNumbers()605 void IntlTestSpoof::testMixedNumbers() {
606 struct Test {
607 const char *fTestString;
608 const char *fExpectedSet;
609 } tests[] = {
610 {"1", "[0]"},
611 {"\\u0967", "[\\u0966]"},
612 {"1\\u0967", "[0\\u0966]"},
613 {"\\u0661\\u06F1", "[\\u0660\\u06F0]"},
614 {"\\u0061\\u2665", "[]"},
615 {"\\u0061\\u303C", "[]"},
616 {"\\u0061\\u30FC\\u303C", "[]"},
617 {"\\u0061\\u30FC\\u303C\\u30A2", "[]"},
618 {"\\u30A2\\u0061\\u30FC\\u303C", "[]"},
619 {"\\u0061\\u0031\\u0661", "[\\u0030\\u0660]"},
620 {"\\u0061\\u0031\\u0661\\u06F1", "[\\u0030\\u0660\\u06F0]"},
621 {"\\u0661\\u30FC\\u303C\\u0061\\u30A2\\u0031\\u0967\\u06F1", "[\\u0030\\u0660\\u06F0\\u0966]"},
622 {"\\u0061\\u30A2\\u30FC\\u303C\\u0031\\u0967\\u0661\\u06F1", "[\\u0030\\u0660\\u06F0\\u0966]"}
623 };
624 UErrorCode status = U_ZERO_ERROR;
625 for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
626 char msgBuf[100];
627 sprintf(msgBuf, "testNum = %d ", testNum);
628 Test &test = tests[testNum];
629
630 status = U_ZERO_ERROR;
631 UnicodeString testString = UnicodeString(test.fTestString).unescape();
632 UnicodeSet expectedSet(UnicodeString(test.fExpectedSet).unescape(), status);
633
634 status = U_ZERO_ERROR;
635 TEST_SETUP
636 uspoof_setChecks(sc, USPOOF_MIXED_NUMBERS, &status); // only check this
637 uspoof_check2UnicodeString(sc, testString, checkResult, &status);
638 UBool mixedNumberFailure = ((uspoof_getCheckResultChecks(checkResult, &status) & USPOOF_MIXED_NUMBERS) != 0);
639 TEST_ASSERT_MSG((expectedSet.size() > 1) == mixedNumberFailure, msgBuf);
640 const UnicodeSet* actualSet = UnicodeSet::fromUSet(uspoof_getCheckResultNumerics(checkResult, &status));
641 TEST_ASSERT_MSG(expectedSet == *actualSet, msgBuf);
642 TEST_TEARDOWN
643 }
644 }
645
646 // Bug #12153 - uspoof_setRestrictionLevel() should enable restriction level testing.
647 //
testBug12153()648 void IntlTestSpoof::testBug12153() {
649 UErrorCode status = U_ZERO_ERROR;
650 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
651 if (!assertSuccess("", status, true, __FILE__, __LINE__)) { return; }
652 int32_t checks = uspoof_getChecks(sc.getAlias(), &status);
653 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) != 0);
654 checks &= ~USPOOF_RESTRICTION_LEVEL;
655 uspoof_setChecks(sc.getAlias(), checks, &status);
656 checks = uspoof_getChecks(sc.getAlias(), &status);
657 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) == 0);
658
659 uspoof_setRestrictionLevel(sc.getAlias(), USPOOF_MODERATELY_RESTRICTIVE);
660 checks = uspoof_getChecks(sc.getAlias(), &status);
661 TEST_ASSERT((checks & USPOOF_RESTRICTION_LEVEL) != 0);
662 TEST_ASSERT_SUCCESS(status);
663 }
664
665 // uspoof_checkUnicodeString should NOT have an infinite loop.
testBug12825()666 void IntlTestSpoof::testBug12825() {
667 UErrorCode status = U_ZERO_ERROR;
668 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
669 TEST_ASSERT_SUCCESS(status);
670 uspoof_setChecks(sc.getAlias(), USPOOF_ALL_CHECKS | USPOOF_AUX_INFO, &status);
671 TEST_ASSERT_SUCCESS(status);
672 uspoof_checkUnicodeString(sc.getAlias(), UnicodeString("\\u30FB").unescape(), NULL, &status);
673 TEST_ASSERT_SUCCESS(status);
674 }
675
676 // uspoof_getSkeleton should NOT set an ILLEGAL_ARGUMENT_EXCEPTION.
testBug12815()677 void IntlTestSpoof::testBug12815() {
678 UErrorCode status = U_ZERO_ERROR;
679 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
680 TEST_ASSERT_SUCCESS(status);
681 uspoof_setChecks(sc.getAlias(), USPOOF_RESTRICTION_LEVEL, &status);
682 TEST_ASSERT_SUCCESS(status);
683 UnicodeString result;
684 uspoof_getSkeletonUnicodeString(sc.getAlias(), 0, UnicodeString("hello world"), result, &status);
685 TEST_ASSERT_SUCCESS(status);
686 }
687
testBug13314_MixedNumbers()688 void IntlTestSpoof::testBug13314_MixedNumbers() {
689 UErrorCode status = U_ZERO_ERROR;
690 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
691 if (!assertSuccess("", status, true, __FILE__, __LINE__)) { return; }
692 uspoof_setChecks(sc.getAlias(), USPOOF_ALL_CHECKS, &status);
693 TEST_ASSERT_SUCCESS(status);
694 int32_t failedChecks = uspoof_areConfusableUnicodeString(sc.getAlias(), u"列", u"列", &status);
695 TEST_ASSERT_SUCCESS(status);
696 assertEquals("The CJK strings should be confusable", USPOOF_SINGLE_SCRIPT_CONFUSABLE, failedChecks);
697 failedChecks = uspoof_check2UnicodeString(sc.getAlias(), u"3Ȝ", nullptr, &status);
698 TEST_ASSERT_SUCCESS(status);
699 assertEquals("The '33' string does not fail spoof", 0, failedChecks);
700 }
701
testBug13328_MixedCombiningMarks()702 void IntlTestSpoof::testBug13328_MixedCombiningMarks() {
703 UErrorCode status = U_ZERO_ERROR;
704 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
705 if (!assertSuccess("", status, true, __FILE__, __LINE__)) { return; }
706 int32_t failedChecks = uspoof_check2UnicodeString(sc.getAlias(), u"\u0061\u0F84", nullptr, &status);
707 TEST_ASSERT_SUCCESS(status);
708 assertEquals(
709 "The mismatched combining marks string fails spoof",
710 USPOOF_RESTRICTION_LEVEL,
711 failedChecks);
712 }
713
testCombiningDot()714 void IntlTestSpoof::testCombiningDot() {
715 UErrorCode status = U_ZERO_ERROR;
716 LocalUSpoofCheckerPointer sc(uspoof_open(&status));
717 if (!assertSuccess("", status, true, __FILE__, __LINE__)) { return; }
718 uspoof_setChecks(sc.getAlias(), USPOOF_HIDDEN_OVERLAY, &status);
719 TEST_ASSERT_SUCCESS(status);
720
721 static const struct TestCase {
722 bool shouldFail;
723 const char16_t* input;
724 } cases[] = {
725 {false, u"i"},
726 {false, u"j"},
727 {false, u"l"},
728 {true, u"i\u0307"},
729 {true, u"j\u0307"},
730 {true, u"l\u0307"},
731 {true, u"ı\u0307"},
732 {true, u"ȷ\u0307"},
733 {true, u"��\u0307"},
734 {true, u"��\u0307"},
735 {false, u"m\u0307"},
736 {true, u"1\u0307"},
737 {true, u"ij\u0307"},
738 {true, u"i\u0307\u0307"},
739 {true, u"abci\u0307def"},
740 {false, u"i\u0301\u0307"}, // U+0301 has combining class ABOVE (230)
741 {true, u"i\u0320\u0307"}, // U+0320 has combining class BELOW
742 {true, u"i\u0320\u0321\u0307"}, // U+0321 also has combining class BELOW
743 {false, u"i\u0320\u0301\u0307"},
744 {false, u"iz\u0307"},
745 };
746
747 for (auto& cas : cases) {
748 int32_t failedChecks = uspoof_check2(sc.getAlias(), cas.input, -1, nullptr, &status);
749 TEST_ASSERT_SUCCESS(status);
750 int32_t expected = cas.shouldFail ? USPOOF_HIDDEN_OVERLAY : 0;
751 assertEquals(cas.input, expected, failedChecks);
752 }
753 }
754
755 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO */
756