icuSources/test/intltest/itspoof.cpp

   1 /*
   2 **********************************************************************
   3 * Copyright (C) 2011-2015, International Business Machines Corporation
   4 * and others.  All Rights Reserved.
   5 **********************************************************************
   6 */
   7 /**
   8  * IntlTestSpoof tests for USpoofDetector
   9  */
  10
  11 #include "unicode/utypes.h"
  12
  13 #if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO
  14
  15 #include "itspoof.h"
  16
  17 #include "unicode/normlzr.h"
  18 #include "unicode/regex.h"
  19 #include "unicode/unistr.h"
  20 #include "unicode/uscript.h"
  21 #include "unicode/uspoof.h"
  22
  23 #include "cstring.h"
  24 #include "identifier_info.h"
  25 #include "scriptset.h"
  26 #include "uhash.h"
  27
  28 #include <stdlib.h>
  29 #include <stdio.h>
  30
  31 #define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \
  32     errcheckln(status, "Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}}
  33
  34 #define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
  35     errln("Test Failure at file %s, line %d: \"%s\" is false.", __FILE__, __LINE__, #expr);};}
  36
  37 #define TEST_ASSERT_MSG(expr, msg) {if ((expr)==FALSE) { \
  38     dataerrln("Test Failure at file %s, line %d, %s: \"%s\" is false.", __FILE__, __LINE__, msg, #expr);};}
  39
  40 #define TEST_ASSERT_EQ(a, b) { if ((a) != (b)) { \
  41     errln("Test Failure at file %s, line %d: \"%s\" (%d) != \"%s\" (%d)", \
  42              __FILE__, __LINE__, #a, (a), #b, (b)); }}
  43
  44 #define TEST_ASSERT_NE(a, b) { if ((a) == (b)) { \
  45     errln("Test Failure at file %s, line %d: \"%s\" (%d) == \"%s\" (%d)", \
  46              __FILE__, __LINE__, #a, (a), #b, (b)); }}
  47
  48 /*
  49  *   TEST_SETUP and TEST_TEARDOWN
  50  *         macros to handle the boilerplate around setting up test case.
  51  *         Put arbitrary test code between SETUP and TEARDOWN.
  52  *         "sc" is the ready-to-go  SpoofChecker for use in the tests.
  53  */
  54 #define TEST_SETUP {  \
  55     UErrorCode status = U_ZERO_ERROR; \
  56     USpoofChecker *sc;     \
  57     sc = uspoof_open(&status);  \
  58     TEST_ASSERT_SUCCESS(status);   \
  59     if (U_SUCCESS(status)){
  60
  61 #define TEST_TEARDOWN  \
  62     }  \
  63     TEST_ASSERT_SUCCESS(status);  \
  64     uspoof_close(sc);  \
  65 }
  66
  67
  68
  69
  70 void IntlTestSpoof::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
  71 {
  72     if (exec) logln("TestSuite spoof: ");
  73     switch (index) {
  74         case 0:
  75             name = "TestSpoofAPI";
  76             if (exec) {
  77                 testSpoofAPI();
  78             }
  79             break;
  80         case 1:
  81             name = "TestSkeleton";
  82             if (exec) {
  83                 testSkeleton();
  84             }
  85             break;
  86         case 2:
  87             name = "TestAreConfusable";
  88             if (exec) {
  89                 testAreConfusable();
  90             }
  91             break;
  92         case 3:
  93             name = "TestInvisible";
  94             if (exec) {
  95                 testInvisible();
  96             }
  97             break;
  98         case 4:
  99             name = "testConfData";
 100             if (exec) {
 101                 testConfData();
 102             }
 103             break;
 104         case 5:
 105             name = "testBug8654";
 106             if (exec) {
 107                 testBug8654();
 108             }
 109             break;
 110         case 6:
 111             name = "testIdentifierInfo";
 112             if (exec) {
 113                 testIdentifierInfo();
 114             }
 115             break;
 116         case 7:
 117             name = "testScriptSet";
 118             if (exec) {
 119                 testScriptSet();
 120             }
 121             break;
 122         case 8:
 123             name = "testRestrictionLevel";
 124             if (exec) {
 125                 testRestrictionLevel();
 126             }
 127             break;
 128        case 9:
 129             name = "testMixedNumbers";
 130             if (exec) {
 131                 testMixedNumbers();
 132             }
 133             break;
 134
 135
 136         default: name=""; break;
 137     }
 138 }
 139
 140 void IntlTestSpoof::testSpoofAPI() {
 141
 142     TEST_SETUP
 143         UnicodeString s("xyz");  // Many latin ranges are whole-script confusable with other scripts.
 144                                  // If this test starts failing, consult confusablesWholeScript.txt
 145         int32_t position = 666;
 146         int32_t checkResults = uspoof_checkUnicodeString(sc, s, &position, &status);
 147         TEST_ASSERT_SUCCESS(status);
 148         TEST_ASSERT_EQ(0, checkResults);
 149         TEST_ASSERT_EQ(0, position);
 150     TEST_TEARDOWN;
 151
 152     TEST_SETUP
 153         UnicodeString s1("cxs");
 154         UnicodeString s2 = UnicodeString("\\u0441\\u0445\\u0455").unescape();  // Cyrillic "cxs"
 155         int32_t checkResults = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
 156         TEST_ASSERT_EQ(USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE, checkResults);
 157
 158     TEST_TEARDOWN;
 159
 160     TEST_SETUP
 161         UnicodeString s("I1l0O");
 162         UnicodeString dest;
 163         UnicodeString &retStr = uspoof_getSkeletonUnicodeString(sc, USPOOF_ANY_CASE, s, dest, &status);
 164         TEST_ASSERT_SUCCESS(status);
 165         TEST_ASSERT(UnicodeString("lllOO") == dest);
 166         TEST_ASSERT(&dest == &retStr);
 167     TEST_TEARDOWN;
 168 }
 169
 170
 171 #define CHECK_SKELETON(type, input, expected) { \
 172     checkSkeleton(sc, type, input, expected, __LINE__); \
 173     }
 174
 175
 176 // testSkeleton.   Spot check a number of confusable skeleton substitutions from the
 177 //                 Unicode data file confusables.txt
 178 //                 Test cases chosen for substitutions of various lengths, and
 179 //                 membership in different mapping tables.
 180 //          Note: for ICU 55, all tables collapsed to the MA table data.
 181 //          TODO: for ICU 56 with Unicode 8, revisit this test.
 182 //
 183 void IntlTestSpoof::testSkeleton() {
 184     const uint32_t ML = 0;
 185     const uint32_t SL = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
 186     const uint32_t MA = USPOOF_ANY_CASE;
 187     const uint32_t SA = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
 188
 189     TEST_SETUP
 190         CHECK_SKELETON(SL, "nochange", "nochange");
 191         CHECK_SKELETON(SA, "nochange", "nochange");
 192         CHECK_SKELETON(ML, "nochange", "nochange");
 193         CHECK_SKELETON(MA, "nochange", "nochange");
 194         CHECK_SKELETON(MA, "love", "love");
 195         CHECK_SKELETON(MA, "1ove", "love");   // Digit 1 to letter l
 196         CHECK_SKELETON(ML, "OOPS", "OOPS");
 197         CHECK_SKELETON(ML, "00PS", "OOPS");
 198         CHECK_SKELETON(MA, "OOPS", "OOPS");
 199         CHECK_SKELETON(MA, "00PS", "OOPS");   // Digit 0 to letter O in any case mode only
 200         CHECK_SKELETON(SL, "\\u059c", "\\u0301");
 201         CHECK_SKELETON(SL, "\\u2A74", "\\u003A\\u003A\\u003D");
 202         CHECK_SKELETON(SL, "\\u247E", "\\u0028\\u006C\\u006C\\u0029");  // "(ll)"
 203         CHECK_SKELETON(SL, "\\uFDFB", "\\u062C\\u0644\\u0020\\u062C\\u0644\\u006c\\u0644\\u006f");
 204
 205         // This mapping exists in the ML and MA tables, does not exist in SL, SA
 206         // 0C83 ;       0983 ;  ML
 207         // 0C83 ;       0983 ;  MA
 208         //
 209
 210         CHECK_SKELETON(SL, "\\u0C83", "\\u0983");
 211         CHECK_SKELETON(SA, "\\u0C83", "\\u0983");
 212         CHECK_SKELETON(ML, "\\u0C83", "\\u0983");
 213         CHECK_SKELETON(MA, "\\u0C83", "\\u0983");
 214
 215         // 0391 mappings exist only in MA and SA tables.
 216         CHECK_SKELETON(MA, "\\u0391", "A");
 217         CHECK_SKELETON(SA, "\\u0391", "A");
 218         CHECK_SKELETON(ML, "\\u0391", "A");
 219         CHECK_SKELETON(SL, "\\u0391", "A");
 220
 221         // 13CF Mappings in all four tables, different in MA.
 222         CHECK_SKELETON(ML, "\\u13CF", "b");
 223         CHECK_SKELETON(MA, "\\u13CF", "b");
 224         CHECK_SKELETON(SL, "\\u13CF", "b");
 225         CHECK_SKELETON(SA, "\\u13CF", "b");
 226
 227         // 0022 ;  0027 0027 ;
 228         // all tables.
 229         CHECK_SKELETON(SL, "\\u0022", "\\u0027\\u0027");
 230         CHECK_SKELETON(SA, "\\u0022", "\\u0027\\u0027");
 231         CHECK_SKELETON(ML, "\\u0022", "\\u0027\\u0027");
 232         CHECK_SKELETON(MA, "\\u0022", "\\u0027\\u0027");
 233
 234         // 017F mappings exist only in MA and SA tables.
 235         CHECK_SKELETON(MA, "\\u017F", "f");
 236         CHECK_SKELETON(SA, "\\u017F", "f");
 237         CHECK_SKELETON(ML, "\\u017F", "f");
 238         CHECK_SKELETON(SL, "\\u017F", "f");
 239
 240     TEST_TEARDOWN;
 241 }
 242
 243
 244 //
 245 //  Run a single confusable skeleton transformation test case.
 246 //
 247 void IntlTestSpoof::checkSkeleton(const USpoofChecker *sc, uint32_t type,
 248                                   const char *input, const char *expected, int32_t lineNum) {
 249     UnicodeString uInput = UnicodeString(input).unescape();
 250     UnicodeString uExpected = UnicodeString(expected).unescape();
 251
 252     UErrorCode status = U_ZERO_ERROR;
 253     UnicodeString actual;
 254     uspoof_getSkeletonUnicodeString(sc, type, uInput, actual, &status);
 255     if (U_FAILURE(status)) {
 256         errln("File %s, Line %d, Test case from line %d, status is %s", __FILE__, __LINE__, lineNum,
 257               u_errorName(status));
 258         return;
 259     }
 260     if (uExpected != actual) {
 261         errln("File %s, Line %d, Test case from line %d, Actual and Expected skeletons differ.",
 262                __FILE__, __LINE__, lineNum);
 263         errln(UnicodeString(" Actual   Skeleton: \"") + actual + UnicodeString("\"\n") +
 264               UnicodeString(" Expected Skeleton: \"") + uExpected + UnicodeString("\""));
 265     }
 266 }
 267
 268 void IntlTestSpoof::testAreConfusable() {
 269     TEST_SETUP
 270         UnicodeString s1("A long string that will overflow stack buffers.  A long string that will overflow stack buffers. "
 271                          "A long string that will overflow stack buffers.  A long string that will overflow stack buffers. ");
 272         UnicodeString s2("A long string that wi11 overflow stack buffers.  A long string that will overflow stack buffers. "
 273                          "A long string that wi11 overflow stack buffers.  A long string that will overflow stack buffers. ");
 274         TEST_ASSERT_EQ(USPOOF_SINGLE_SCRIPT_CONFUSABLE, uspoof_areConfusableUnicodeString(sc, s1, s2, &status));
 275         TEST_ASSERT_SUCCESS(status);
 276
 277     TEST_TEARDOWN;
 278 }
 279
 280 void IntlTestSpoof::testInvisible() {
 281     TEST_SETUP
 282         UnicodeString  s = UnicodeString("abcd\\u0301ef").unescape();
 283         int32_t position = -42;
 284         TEST_ASSERT_EQ(0, uspoof_checkUnicodeString(sc, s, &position, &status));
 285         TEST_ASSERT_SUCCESS(status);
 286         TEST_ASSERT(0 == position);
 287
 288         UnicodeString  s2 = UnicodeString("abcd\\u0301\\u0302\\u0301ef").unescape();
 289         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s2, &position, &status));
 290         TEST_ASSERT_SUCCESS(status);
 291         TEST_ASSERT_EQ(0, position);
 292
 293         // Two acute accents, one from the composed a with acute accent, \u00e1,
 294         // and one separate.
 295         position = -42;
 296         UnicodeString  s3 = UnicodeString("abcd\\u00e1\\u0301xyz").unescape();
 297         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s3, &position, &status));
 298         TEST_ASSERT_SUCCESS(status);
 299         TEST_ASSERT_EQ(0, position);
 300     TEST_TEARDOWN;
 301 }
 302
 303 void IntlTestSpoof::testBug8654() {
 304     TEST_SETUP
 305         UnicodeString s = UnicodeString("B\\u00c1\\u0301").unescape();
 306         int32_t position = -42;
 307         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s, &position, &status) & USPOOF_INVISIBLE );
 308         TEST_ASSERT_SUCCESS(status);
 309         TEST_ASSERT_EQ(0, position);
 310     TEST_TEARDOWN;
 311 }
 312
 313 static UnicodeString parseHex(const UnicodeString &in) {
 314     // Convert a series of hex numbers in a Unicode String to a string with the
 315     // corresponding characters.
 316     // The conversion is _really_ annoying.  There must be some function to just do it.
 317     UnicodeString result;
 318     UChar32 cc = 0;
 319     for (int32_t i=0; i<in.length(); i++) {
 320         UChar c = in.charAt(i);
 321         if (c == 0x20) {   // Space
 322             if (cc > 0) {
 323                result.append(cc);
 324                cc = 0;
 325             }
 326         } else if (c>=0x30 && c<=0x39) {
 327             cc = (cc<<4) + (c - 0x30);
 328         } else if ((c>=0x41 && c<=0x46) || (c>=0x61 && c<=0x66)) {
 329             cc = (cc<<4) + (c & 0x0f)+9;
 330         }
 331         // else do something with bad input.
 332     }
 333     if (cc > 0) {
 334         result.append(cc);
 335     }
 336     return result;
 337 }
 338
 339
 340 //
 341 // Append the hex form of a UChar32 to a UnicodeString.
 342 // Used in formatting error messages.
 343 // Match the formatting of numbers in confusables.txt
 344 // Minimum of 4 digits, no leading zeroes for positions 5 and up.
 345 //
 346 static void appendHexUChar(UnicodeString &dest, UChar32 c) {
 347     UBool   doZeroes = FALSE;
 348     for (int bitNum=28; bitNum>=0; bitNum-=4) {
 349         if (bitNum <= 12) {
 350             doZeroes = TRUE;
 351         }
 352         int hexDigit = (c>>bitNum) & 0x0f;
 353         if (hexDigit != 0 || doZeroes) {
 354             doZeroes = TRUE;
 355             dest.append((UChar)(hexDigit<=9? hexDigit + 0x30: hexDigit -10 + 0x41));
 356         }
 357     }
 358     dest.append((UChar)0x20);
 359 }
 360
 361 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
 362
 363 //  testConfData - Check each data item from the Unicode confusables.txt file,
 364 //                 verify that it transforms correctly in a skeleton.
 365 //
 366 void IntlTestSpoof::testConfData() {
 367     char buffer[2000];
 368     if (getUnidataPath(buffer) == NULL) {
 369         errln("Skipping test spoof/testConfData. Unable to find path to source/data/unidata/.");
 370         return;
 371     }
 372     uprv_strcat(buffer, "confusables.txt");
 373
 374     LocalStdioFilePointer f(fopen(buffer, "rb"));
 375     if (f.isNull()) {
 376         errln("Skipping test spoof/testConfData.  File confusables.txt not accessible.");
 377         return;
 378     }
 379     fseek(f.getAlias(), 0, SEEK_END);
 380     int32_t  fileSize = ftell(f.getAlias());
 381     LocalArray<char> fileBuf(new char[fileSize]);
 382     fseek(f.getAlias(), 0, SEEK_SET);
 383     int32_t amt_read = fread(fileBuf.getAlias(), 1, fileSize, f.getAlias());
 384     TEST_ASSERT_EQ(amt_read, fileSize);
 385     TEST_ASSERT(fileSize>0);
 386     if (amt_read != fileSize || fileSize <=0) {
 387         return;
 388     }
 389     UnicodeString confusablesTxt = UnicodeString::fromUTF8(StringPiece(fileBuf.getAlias(), fileSize));
 390
 391     UErrorCode status = U_ZERO_ERROR;
 392     LocalUSpoofCheckerPointer sc(uspoof_open(&status));
 393     TEST_ASSERT_SUCCESS(status);
 394
 395     // Parse lines from the confusables.txt file.  Example Line:
 396     // FF44 ;   0064 ;  SL      # ( d -> d ) FULLWIDTH ....
 397     // Three fields.  The hex fields can contain more than one character,
 398     //                and each character may be more than 4 digits (for supplemntals)
 399     // This regular expression matches lines and splits the fields into capture groups.
 400     RegexMatcher parseLine("(?m)^([0-9A-F]{4}[^#;]*?);([^#;]*?);([^#]*)", confusablesTxt, 0, status);
 401     TEST_ASSERT_SUCCESS(status);
 402     while (parseLine.find()) {
 403         UnicodeString from = parseHex(parseLine.group(1, status));
 404         if (!Normalizer::isNormalized(from, UNORM_NFD, status)) {
 405             // The source character was not NFD.
 406             // Skip this case; the first step in obtaining a skeleton is to NFD the input,
 407             //  so the mapping in this line of confusables.txt will never be applied.
 408             continue;
 409         }
 410
 411         UnicodeString rawExpected = parseHex(parseLine.group(2, status));
 412         UnicodeString expected;
 413         Normalizer::decompose(rawExpected, FALSE /*NFD*/, 0, expected, status);
 414         TEST_ASSERT_SUCCESS(status);
 415
 416         int32_t skeletonType = 0;
 417         UnicodeString tableType = parseLine.group(3, status);
 418         TEST_ASSERT_SUCCESS(status);
 419         if (tableType.indexOf("SL") >= 0) {
 420             skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
 421         } else if (tableType.indexOf("SA") >= 0) {
 422             skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
 423         } else if (tableType.indexOf("ML") >= 0) {
 424             skeletonType = 0;
 425         } else if (tableType.indexOf("MA") >= 0) {
 426             skeletonType = USPOOF_ANY_CASE;
 427         }
 428
 429         UnicodeString actual;
 430         uspoof_getSkeletonUnicodeString(sc.getAlias(), skeletonType, from, actual, &status);
 431         TEST_ASSERT_SUCCESS(status);
 432         TEST_ASSERT(actual == expected);
 433         if (actual != expected) {
 434             errln(parseLine.group(0, status));
 435             UnicodeString line = "Actual: ";
 436             int i = 0;
 437             while (i < actual.length()) {
 438                 appendHexUChar(line, actual.char32At(i));
 439                 i = actual.moveIndex32(i, 1);
 440             }
 441             errln(line);
 442         }
 443         if (U_FAILURE(status)) {
 444             break;
 445         }
 446     }
 447 }
 448
 449 // testIdentifierInfo. Note that IdentifierInfo is not public ICU API at this time
 450 void IntlTestSpoof::testIdentifierInfo() {
 451     UErrorCode status = U_ZERO_ERROR;
 452     ScriptSet bitset12; bitset12.set(USCRIPT_LATIN, status).set(USCRIPT_HANGUL, status);
 453     ScriptSet bitset2;  bitset2.set(USCRIPT_HANGUL, status);
 454     TEST_ASSERT(bitset12.contains(bitset2));
 455     TEST_ASSERT(bitset12.contains(bitset12));
 456     TEST_ASSERT(!bitset2.contains(bitset12));
 457
 458     ScriptSet arabSet;  arabSet.set(USCRIPT_ARABIC, status);
 459     ScriptSet latinSet; latinSet.set(USCRIPT_LATIN, status);
 460     UElement arabEl;  arabEl.pointer = &arabSet;
 461     UElement latinEl; latinEl.pointer = &latinSet;
 462     TEST_ASSERT(uhash_compareScriptSet(arabEl, latinEl) < 0);
 463     TEST_ASSERT(uhash_compareScriptSet(latinEl, arabEl) > 0);
 464
 465     UnicodeString scriptString;
 466     bitset12.displayScripts(scriptString);
 467     TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang Latn") == scriptString);
 468
 469     status = U_ZERO_ERROR;
 470     UHashtable *alternates = uhash_open(uhash_hashScriptSet ,uhash_compareScriptSet, NULL, &status);
 471     uhash_puti(alternates, &bitset12, 1, &status);
 472     uhash_puti(alternates, &bitset2, 1, &status);
 473     UnicodeString alternatesString;
 474     IdentifierInfo::displayAlternates(alternatesString, alternates, status);
 475     TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang; Hang Latn") == alternatesString);
 476     TEST_ASSERT_SUCCESS(status);
 477
 478     status = U_ZERO_ERROR;
 479     ScriptSet tScriptSet;
 480     tScriptSet.parseScripts(scriptString, status);
 481     TEST_ASSERT_SUCCESS(status);
 482     TEST_ASSERT(bitset12 == tScriptSet);
 483     UnicodeString ss;
 484     ss.remove();
 485     uhash_close(alternates);
 486
 487     struct Test {
 488         const char         *fTestString;
 489         URestrictionLevel   fRestrictionLevel;
 490         const char         *fNumerics;
 491         const char         *fScripts;
 492         const char         *fAlternates;
 493         const char         *fCommonAlternates;
 494     } tests[] = {
 495             {"\\u0061\\u2665",                USPOOF_UNRESTRICTIVE,      "[]", "Latn", "", ""},
 496             {"\\u0061\\u3006",                USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hani Hira Kana", "Hani Hira Kana"},
 497             {"\\u0061\\u30FC\\u3006",         USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hira Kana", "Hira Kana"},
 498             {"\\u0061\\u30FC\\u3006\\u30A2",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
 499             {"\\u30A2\\u0061\\u30FC\\u3006",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
 500             {"\\u0061\\u0031\\u0661",         USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660]", "Latn", "Arab Thaa", "Arab Thaa"},
 501             {"\\u0061\\u0031\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660\\u06F0]", "Latn Arab", "", ""},
 502             {"\\u0661\\u30FC\\u3006\\u0061\\u30A2\\u0031\\u0967\\u06F1",  USPOOF_UNRESTRICTIVE,
 503                   "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi Mahj", "Deva Kthi Mahj"},
 504             {"\\u0061\\u30A2\\u30FC\\u3006\\u0031\\u0967\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE,
 505                   "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi Mahj", "Deva Kthi Mahj"}
 506     };
 507
 508     int testNum;
 509     for (testNum = 0; testNum < UPRV_LENGTHOF(tests); testNum++) {
 510         char testNumStr[40];
 511         sprintf(testNumStr, "testNum = %d", testNum);
 512         Test &test = tests[testNum];
 513         status = U_ZERO_ERROR;
 514         UnicodeString testString(test.fTestString);  // Note: may do charset conversion.
 515         testString = testString.unescape();
 516         IdentifierInfo idInfo(status);
 517         TEST_ASSERT_SUCCESS(status);
 518         idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
 519         idInfo.setIdentifier(testString, status);
 520         TEST_ASSERT_MSG(*idInfo.getIdentifier() == testString, testNumStr);
 521
 522         URestrictionLevel restrictionLevel = test.fRestrictionLevel;
 523         TEST_ASSERT_MSG(restrictionLevel == idInfo.getRestrictionLevel(status), testNumStr);
 524
 525         status = U_ZERO_ERROR;
 526         UnicodeSet numerics(UnicodeString(test.fNumerics).unescape(), status);
 527         TEST_ASSERT_SUCCESS(status);
 528         TEST_ASSERT_MSG(numerics == *idInfo.getNumerics(), testNumStr);
 529
 530         ScriptSet scripts;
 531         scripts.parseScripts(UnicodeString(test.fScripts), status);
 532         TEST_ASSERT_MSG(scripts == *idInfo.getScripts(), testNumStr);
 533
 534         UnicodeString alternatesStr;
 535         IdentifierInfo::displayAlternates(alternatesStr, idInfo.getAlternates(), status);
 536         TEST_ASSERT_MSG(UnicodeString(test.fAlternates) == alternatesStr, testNumStr);
 537
 538         ScriptSet commonAlternates;
 539         commonAlternates.parseScripts(UnicodeString(test.fCommonAlternates), status);
 540         TEST_ASSERT_MSG(commonAlternates == *idInfo.getCommonAmongAlternates(), testNumStr);
 541     }
 542
 543     // Test of getScriptCount()
 544     //   Script and or Script Extension for chars used in the tests
 545     //     \\u3013  ; Bopo Hang Hani Hira Kana # So       GETA MARK
 546     //     \\uA838  ; Deva Gujr Guru Kthi Takr # Sc       NORTH INDIC RUPEE MARK
 547     //     \\u0951  ; Deva Latn                # Mn       DEVANAGARI STRESS SIGN UDATTA
 548     //
 549     //     \\u0370  ; Greek                    # L        GREEK CAPITAL LETTER HETA
 550     //     \\u0481  ; Cyrillic                 # L&       CYRILLIC SMALL LETTER KOPPA
 551     //     \\u0904  ; Devanagari               # Lo       DEVANAGARI LETTER SHORT A
 552     //     \\u3041  ; Hiragana                 # Lo       HIRAGANA LETTER SMALL A
 553     //     1234     ; Common                   #          ascii digits
 554     //     \\u0300  ; Inherited                # Mn       COMBINING GRAVE ACCENT
 555
 556     struct ScriptTest {
 557         const char *fTestString;
 558         int32_t     fScriptCount;
 559     } scriptTests[] = {
 560         {"Hello", 1},
 561         {"Hello\\u0370", 2},
 562         {"1234", 0},
 563         {"Hello1234\\u0300", 1},   // Common and Inherited are ignored.
 564         {"\\u0030", 0},
 565         {"abc\\u0951", 1},
 566         {"abc\\u3013", 2},
 567         {"\\uA838\\u0951", 1},     // Triggers commonAmongAlternates path.
 568         {"\\u3013\\uA838", 2}
 569     };
 570
 571     status = U_ZERO_ERROR;
 572     IdentifierInfo identifierInfo(status);
 573     for (testNum=0; testNum<UPRV_LENGTHOF(scriptTests); testNum++) {
 574         ScriptTest &test = scriptTests[testNum];
 575         char msgBuf[100];
 576         sprintf(msgBuf, "testNum = %d ", testNum);
 577         UnicodeString testString = UnicodeString(test.fTestString).unescape();
 578
 579         status = U_ZERO_ERROR;
 580         identifierInfo.setIdentifier(testString, status);
 581         int32_t scriptCount = identifierInfo.getScriptCount();
 582         TEST_ASSERT_MSG(test.fScriptCount == scriptCount, msgBuf);
 583     }
 584 }
 585
 586 void IntlTestSpoof::testScriptSet() {
 587     ScriptSet s1;
 588     ScriptSet s2;
 589     UErrorCode status = U_ZERO_ERROR;
 590
 591     TEST_ASSERT(s1 == s2);
 592     s1.set(USCRIPT_ARABIC,status);
 593     TEST_ASSERT_SUCCESS(status);
 594     TEST_ASSERT(!(s1 == s2));
 595     TEST_ASSERT(s1.test(USCRIPT_ARABIC, status));
 596     TEST_ASSERT(s1.test(USCRIPT_GREEK, status) == FALSE);
 597
 598     status = U_ZERO_ERROR;
 599     s1.reset(USCRIPT_ARABIC, status);
 600     TEST_ASSERT(s1 == s2);
 601
 602     status = U_ZERO_ERROR;
 603     s1.setAll();
 604     TEST_ASSERT(s1.test(USCRIPT_COMMON, status));
 605     TEST_ASSERT(s1.test(USCRIPT_ETHIOPIC, status));
 606     TEST_ASSERT(s1.test(USCRIPT_CODE_LIMIT, status));
 607     s1.resetAll();
 608     TEST_ASSERT(!s1.test(USCRIPT_COMMON, status));
 609     TEST_ASSERT(!s1.test(USCRIPT_ETHIOPIC, status));
 610     TEST_ASSERT(!s1.test(USCRIPT_CODE_LIMIT, status));
 611
 612     status = U_ZERO_ERROR;
 613     s1.set(USCRIPT_TAKRI, status);
 614     s1.set(USCRIPT_BLISSYMBOLS, status);
 615     s2.setAll();
 616     TEST_ASSERT(s2.contains(s1));
 617     TEST_ASSERT(!s1.contains(s2));
 618     TEST_ASSERT(s2.intersects(s1));
 619     TEST_ASSERT(s1.intersects(s2));
 620     s2.reset(USCRIPT_TAKRI, status);
 621     TEST_ASSERT(!s2.contains(s1));
 622     TEST_ASSERT(!s1.contains(s2));
 623     TEST_ASSERT(s1.intersects(s2));
 624     TEST_ASSERT(s2.intersects(s1));
 625     TEST_ASSERT_SUCCESS(status);
 626
 627     status = U_ZERO_ERROR;
 628     s1.resetAll();
 629     s1.set(USCRIPT_NKO, status);
 630     s1.set(USCRIPT_COMMON, status);
 631     s2 = s1;
 632     TEST_ASSERT(s2 == s1);
 633     TEST_ASSERT_EQ(2, s2.countMembers());
 634     s2.intersect(s1);
 635     TEST_ASSERT(s2 == s1);
 636     s2.setAll();
 637     TEST_ASSERT(!(s2 == s1));
 638     TEST_ASSERT(s2.countMembers() >= USCRIPT_CODE_LIMIT);
 639     s2.intersect(s1);
 640     TEST_ASSERT(s2 == s1);
 641
 642     s2.setAll();
 643     s2.reset(USCRIPT_COMMON, status);
 644     s2.intersect(s1);
 645     TEST_ASSERT(s2.countMembers() == 1);
 646
 647     s1.resetAll();
 648     s1.set(USCRIPT_AFAKA, status);
 649     s1.set(USCRIPT_VAI, status);
 650     s1.set(USCRIPT_INHERITED, status);
 651     int32_t n = -1;
 652     for (int32_t i=0; i<4; i++) {
 653         n = s1.nextSetBit(n+1);
 654         switch (i) {
 655           case 0: TEST_ASSERT_EQ(USCRIPT_INHERITED, n); break;
 656           case 1: TEST_ASSERT_EQ(USCRIPT_VAI, n); break;
 657           case 2: TEST_ASSERT_EQ(USCRIPT_AFAKA, n); break;
 658           case 3: TEST_ASSERT_EQ(-1, (int32_t)n); break;
 659           default: TEST_ASSERT(FALSE);
 660         }
 661     }
 662     TEST_ASSERT_SUCCESS(status);
 663 }
 664
 665
 666 void IntlTestSpoof::testRestrictionLevel() {
 667     struct Test {
 668         const char         *fId;
 669         URestrictionLevel   fExpectedRestrictionLevel;
 670     } tests[] = {
 671         {"\\u0061\\u03B3\\u2665", USPOOF_UNRESTRICTIVE},
 672         {"a",                     USPOOF_ASCII},
 673         {"\\u03B3",               USPOOF_SINGLE_SCRIPT_RESTRICTIVE},
 674         {"\\u0061\\u30A2\\u30FC", USPOOF_HIGHLY_RESTRICTIVE},
 675         {"\\u0061\\u0904",        USPOOF_MODERATELY_RESTRICTIVE},
 676         {"\\u0061\\u03B3",        USPOOF_MINIMALLY_RESTRICTIVE}
 677     };
 678     char msgBuffer[100];
 679
 680     URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_SINGLE_SCRIPT_RESTRICTIVE,
 681          USPOOF_HIGHLY_RESTRICTIVE, USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE,
 682          USPOOF_UNRESTRICTIVE};
 683
 684     UErrorCode status = U_ZERO_ERROR;
 685     IdentifierInfo idInfo(status);
 686     TEST_ASSERT_SUCCESS(status);
 687     idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
 688     TEST_ASSERT_SUCCESS(status);
 689     for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
 690         status = U_ZERO_ERROR;
 691         const Test &test = tests[testNum];
 692         UnicodeString testString = UnicodeString(test.fId).unescape();
 693         URestrictionLevel expectedLevel = test.fExpectedRestrictionLevel;
 694         idInfo.setIdentifier(testString, status);
 695         sprintf(msgBuffer, "testNum = %d ", testNum);
 696         TEST_ASSERT_SUCCESS(status);
 697         TEST_ASSERT_MSG(expectedLevel == idInfo.getRestrictionLevel(status), msgBuffer);
 698         for (int levelIndex=0; levelIndex<UPRV_LENGTHOF(restrictionLevels); levelIndex++) {
 699             status = U_ZERO_ERROR;
 700             URestrictionLevel levelSetInSpoofChecker = restrictionLevels[levelIndex];
 701             USpoofChecker *sc = uspoof_open(&status);
 702             uspoof_setChecks(sc, USPOOF_RESTRICTION_LEVEL, &status);
 703             uspoof_setAllowedChars(sc, uspoof_getRecommendedSet(&status), &status);
 704             uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
 705             int32_t actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status);
 706
 707             // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
 708             int32_t expectedValue = 0;
 709             if (expectedLevel > levelSetInSpoofChecker) {
 710                 expectedValue |= USPOOF_RESTRICTION_LEVEL;
 711             }
 712             if (!uspoof_getRecommendedUnicodeSet(&status)->containsAll(testString)) {
 713                 expectedValue |= USPOOF_CHAR_LIMIT;
 714             }
 715             sprintf(msgBuffer, "testNum = %d, levelIndex = %d, expected = %#x, actual = %#x",
 716                     testNum, levelIndex, expectedValue, actualValue);
 717             TEST_ASSERT_MSG(expectedValue == actualValue, msgBuffer);
 718             TEST_ASSERT_SUCCESS(status);
 719
 720             // Run the same check again, with the Spoof Checker configured to return
 721             // the actual restriction level.
 722             uspoof_setChecks(sc, USPOOF_AUX_INFO | USPOOF_RESTRICTION_LEVEL, &status);
 723             uspoof_setAllowedChars(sc, uspoof_getRecommendedSet(&status), &status);
 724             uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
 725             int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
 726             TEST_ASSERT_SUCCESS(status);
 727             if (U_SUCCESS(status)) {
 728                 TEST_ASSERT_EQ(expectedLevel, result & USPOOF_RESTRICTION_LEVEL_MASK);
 729                 TEST_ASSERT_EQ(expectedValue, result & USPOOF_ALL_CHECKS);
 730             }
 731             uspoof_close(sc);
 732         }
 733     }
 734 }
 735
 736
 737 void IntlTestSpoof::testMixedNumbers() {
 738     struct Test {
 739         const char *fTestString;
 740         const char *fExpectedSet;
 741     } tests[] = {
 742         {"1",              "[0]"},
 743         {"\\u0967",        "[\\u0966]"},
 744         {"1\\u0967",       "[0\\u0966]"},
 745         {"\\u0661\\u06F1", "[\\u0660\\u06F0]"}
 746     };
 747     UErrorCode status = U_ZERO_ERROR;
 748     IdentifierInfo idInfo(status);
 749     for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
 750         char msgBuf[100];
 751         sprintf(msgBuf, "testNum = %d ", testNum);
 752         Test &test = tests[testNum];
 753
 754         status = U_ZERO_ERROR;
 755         UnicodeString testString = UnicodeString(test.fTestString).unescape();
 756         UnicodeSet expectedSet(UnicodeString(test.fExpectedSet).unescape(), status);
 757         idInfo.setIdentifier(testString, status);
 758         TEST_ASSERT_SUCCESS(status);
 759         TEST_ASSERT_MSG(expectedSet == *idInfo.getNumerics(), msgBuf);
 760
 761         status = U_ZERO_ERROR;
 762         USpoofChecker *sc = uspoof_open(&status);
 763         uspoof_setChecks(sc, USPOOF_MIXED_NUMBERS, &status); // only check this
 764         int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
 765         UBool mixedNumberFailure = ((result & USPOOF_MIXED_NUMBERS) != 0);
 766         TEST_ASSERT_MSG((expectedSet.size() > 1) == mixedNumberFailure, msgBuf);
 767         uspoof_close(sc);
 768     }
 769 }
 770
 771 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO */