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