icuSources/test/intltest/itspoof.cpp

   1 /*
   2 **********************************************************************
   3 * Copyright (C) 2011-2013, 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 #define LENGTHOF(array) ((int32_t)(sizeof(array)/sizeof((array)[0])))
  49
  50 /*
  51  *   TEST_SETUP and TEST_TEARDOWN
  52  *         macros to handle the boilerplate around setting up test case.
  53  *         Put arbitrary test code between SETUP and TEARDOWN.
  54  *         "sc" is the ready-to-go  SpoofChecker for use in the tests.
  55  */
  56 #define TEST_SETUP {  \
  57     UErrorCode status = U_ZERO_ERROR; \
  58     USpoofChecker *sc;     \
  59     sc = uspoof_open(&status);  \
  60     TEST_ASSERT_SUCCESS(status);   \
  61     if (U_SUCCESS(status)){
  62
  63 #define TEST_TEARDOWN  \
  64     }  \
  65     TEST_ASSERT_SUCCESS(status);  \
  66     uspoof_close(sc);  \
  67 }
  68
  69
  70
  71
  72 void IntlTestSpoof::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
  73 {
  74     if (exec) logln("TestSuite spoof: ");
  75     switch (index) {
  76         case 0:
  77             name = "TestSpoofAPI";
  78             if (exec) {
  79                 testSpoofAPI();
  80             }
  81             break;
  82         case 1:
  83             name = "TestSkeleton";
  84             if (exec) {
  85                 testSkeleton();
  86             }
  87             break;
  88         case 2:
  89             name = "TestAreConfusable";
  90             if (exec) {
  91                 testAreConfusable();
  92             }
  93             break;
  94         case 3:
  95             name = "TestInvisible";
  96             if (exec) {
  97                 testInvisible();
  98             }
  99             break;
 100         case 4:
 101             name = "testConfData";
 102             if (exec) {
 103                 testConfData();
 104             }
 105             break;
 106         case 5:
 107             name = "testBug8654";
 108             if (exec) {
 109                 testBug8654();
 110             }
 111             break;
 112         case 6:
 113             name = "testIdentifierInfo";
 114             if (exec) {
 115                 testIdentifierInfo();
 116             }
 117             break;
 118         case 7:
 119             name = "testScriptSet";
 120             if (exec) {
 121                 testScriptSet();
 122             }
 123             break;
 124         case 8:
 125             name = "testRestrictionLevel";
 126             if (exec) {
 127                 testRestrictionLevel();
 128             }
 129             break;
 130        case 9:
 131             name = "testMixedNumbers";
 132             if (exec) {
 133                 testMixedNumbers();
 134             }
 135             break;
 136
 137
 138         default: name=""; break;
 139     }
 140 }
 141
 142 void IntlTestSpoof::testSpoofAPI() {
 143
 144     TEST_SETUP
 145         UnicodeString s("xyz");  // Many latin ranges are whole-script confusable with other scripts.
 146                                  // If this test starts failing, consult confusablesWholeScript.txt
 147         int32_t position = 666;
 148         int32_t checkResults = uspoof_checkUnicodeString(sc, s, &position, &status);
 149         TEST_ASSERT_SUCCESS(status);
 150         TEST_ASSERT_EQ(0, checkResults);
 151         TEST_ASSERT_EQ(0, position);
 152     TEST_TEARDOWN;
 153
 154     TEST_SETUP
 155         UnicodeString s1("cxs");
 156         UnicodeString s2 = UnicodeString("\\u0441\\u0445\\u0455").unescape();  // Cyrillic "cxs"
 157         int32_t checkResults = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
 158         TEST_ASSERT_EQ(USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE, checkResults);
 159
 160     TEST_TEARDOWN;
 161
 162     TEST_SETUP
 163         UnicodeString s("I1l0O");
 164         UnicodeString dest;
 165         UnicodeString &retStr = uspoof_getSkeletonUnicodeString(sc, USPOOF_ANY_CASE, s, dest, &status);
 166         TEST_ASSERT_SUCCESS(status);
 167         TEST_ASSERT(UnicodeString("lllOO") == dest);
 168         TEST_ASSERT(&dest == &retStr);
 169     TEST_TEARDOWN;
 170 }
 171
 172
 173 #define CHECK_SKELETON(type, input, expected) { \
 174     checkSkeleton(sc, type, input, expected, __LINE__); \
 175     }
 176
 177
 178 // testSkeleton.   Spot check a number of confusable skeleton substitutions from the
 179 //                 Unicode data file confusables.txt
 180 //                 Test cases chosen for substitutions of various lengths, and
 181 //                 membership in different mapping tables.
 182 void IntlTestSpoof::testSkeleton() {
 183     const uint32_t ML = 0;
 184     const uint32_t SL = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
 185     const uint32_t MA = USPOOF_ANY_CASE;
 186     const uint32_t SA = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
 187
 188     TEST_SETUP
 189         // A long "identifier" that will overflow implementation stack buffers, forcing heap allocations.
 190         CHECK_SKELETON(SL, " A 1ong \\u02b9identifier' that will overflow implementation stack buffers, forcing heap allocations."
 191                            " A 1ong 'identifier' that will overflow implementation stack buffers, forcing heap allocations."
 192                            " A 1ong 'identifier' that will overflow implementation stack buffers, forcing heap allocations."
 193                            " A 1ong 'identifier' that will overflow implementation stack buffers, forcing heap allocations.",
 194
 195                " A long 'identifier' that vvill overflovv irnplernentation stack buffers, forcing heap allocations."
 196                " A long 'identifier' that vvill overflovv irnplernentation stack buffers, forcing heap allocations."
 197                " A long 'identifier' that vvill overflovv irnplernentation stack buffers, forcing heap allocations."
 198                " A long 'identifier' that vvill overflovv irnplernentation stack buffers, forcing heap allocations.")
 199
 200         CHECK_SKELETON(SL, "nochange", "nochange");
 201         CHECK_SKELETON(MA, "love", "love");
 202         CHECK_SKELETON(MA, "1ove", "love");   // Digit 1 to letter l
 203         CHECK_SKELETON(ML, "OOPS", "OOPS");
 204         CHECK_SKELETON(ML, "00PS", "00PS");   // Digit 0 unchanged in lower case mode.
 205         CHECK_SKELETON(MA, "OOPS", "OOPS");
 206         CHECK_SKELETON(MA, "00PS", "OOPS");   // Digit 0 to letter O in any case mode only
 207         CHECK_SKELETON(SL, "\\u059c", "\\u0301");
 208         CHECK_SKELETON(SL, "\\u2A74", "\\u003A\\u003A\\u003D");
 209         CHECK_SKELETON(SL, "\\u247E", "\\u0028\\u006C\\u006C\\u0029");  // "(ll)"
 210         CHECK_SKELETON(SL, "\\uFDFB", "\\u062C\\u0644\\u0020\\u062C\\u0644\\u0627\\u0644\\u0647");
 211
 212         // This mapping exists in the ML and MA tables, does not exist in SL, SA
 213         //0C83 ;        0C03 ;
 214         CHECK_SKELETON(SL, "\\u0C83", "\\u0C83");
 215         CHECK_SKELETON(SA, "\\u0C83", "\\u0C83");
 216         CHECK_SKELETON(ML, "\\u0C83", "\\u0983");
 217         CHECK_SKELETON(MA, "\\u0C83", "\\u0983");
 218
 219         // 0391 ; 0041 ;
 220         // This mapping exists only in the MA table.
 221         CHECK_SKELETON(MA, "\\u0391", "A");
 222         CHECK_SKELETON(SA, "\\u0391", "\\u0391");
 223         CHECK_SKELETON(ML, "\\u0391", "\\u0391");
 224         CHECK_SKELETON(SL, "\\u0391", "\\u0391");
 225
 226         // 13CF ;  0062 ;
 227         // This mapping exists in the ML and MA tables
 228         CHECK_SKELETON(ML, "\\u13CF", "b");
 229         CHECK_SKELETON(MA, "\\u13CF", "b");
 230         CHECK_SKELETON(SL, "\\u13CF", "\\u13CF");
 231         CHECK_SKELETON(SA, "\\u13CF", "\\u13CF");
 232
 233         // 0022 ;  0027 0027 ;
 234         // all tables.
 235         CHECK_SKELETON(SL, "\\u0022", "\\u0027\\u0027");
 236         CHECK_SKELETON(SA, "\\u0022", "\\u0027\\u0027");
 237         CHECK_SKELETON(ML, "\\u0022", "\\u0027\\u0027");
 238         CHECK_SKELETON(MA, "\\u0022", "\\u0027\\u0027");
 239
 240         // 017F ;  0066 ;
 241         // This mapping exists in the SA and MA tables
 242         CHECK_SKELETON(MA, "\\u017F", "f");
 243         CHECK_SKELETON(SA, "\\u017F", "f");
 244
 245     TEST_TEARDOWN;
 246 }
 247
 248
 249 //
 250 //  Run a single confusable skeleton transformation test case.
 251 //
 252 void IntlTestSpoof::checkSkeleton(const USpoofChecker *sc, uint32_t type,
 253                                   const char *input, const char *expected, int32_t lineNum) {
 254     UnicodeString uInput = UnicodeString(input).unescape();
 255     UnicodeString uExpected = UnicodeString(expected).unescape();
 256
 257     UErrorCode status = U_ZERO_ERROR;
 258     UnicodeString actual;
 259     uspoof_getSkeletonUnicodeString(sc, type, uInput, actual, &status);
 260     if (U_FAILURE(status)) {
 261         errln("File %s, Line %d, Test case from line %d, status is %s", __FILE__, __LINE__, lineNum,
 262               u_errorName(status));
 263         return;
 264     }
 265     if (uExpected != actual) {
 266         errln("File %s, Line %d, Test case from line %d, Actual and Expected skeletons differ.",
 267                __FILE__, __LINE__, lineNum);
 268         errln(UnicodeString(" Actual   Skeleton: \"") + actual + UnicodeString("\"\n") +
 269               UnicodeString(" Expected Skeleton: \"") + uExpected + UnicodeString("\""));
 270     }
 271 }
 272
 273 void IntlTestSpoof::testAreConfusable() {
 274     TEST_SETUP
 275         UnicodeString s1("A long string that will overflow stack buffers.  A long string that will overflow stack buffers. "
 276                          "A long string that will overflow stack buffers.  A long string that will overflow stack buffers. ");
 277         UnicodeString s2("A long string that wi11 overflow stack buffers.  A long string that will overflow stack buffers. "
 278                          "A long string that wi11 overflow stack buffers.  A long string that will overflow stack buffers. ");
 279         TEST_ASSERT_EQ(USPOOF_SINGLE_SCRIPT_CONFUSABLE, uspoof_areConfusableUnicodeString(sc, s1, s2, &status));
 280         TEST_ASSERT_SUCCESS(status);
 281
 282     TEST_TEARDOWN;
 283 }
 284
 285 void IntlTestSpoof::testInvisible() {
 286     TEST_SETUP
 287         UnicodeString  s = UnicodeString("abcd\\u0301ef").unescape();
 288         int32_t position = -42;
 289         TEST_ASSERT_EQ(0, uspoof_checkUnicodeString(sc, s, &position, &status));
 290         TEST_ASSERT_SUCCESS(status);
 291         TEST_ASSERT(0 == position);
 292
 293         UnicodeString  s2 = UnicodeString("abcd\\u0301\\u0302\\u0301ef").unescape();
 294         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s2, &position, &status));
 295         TEST_ASSERT_SUCCESS(status);
 296         TEST_ASSERT_EQ(0, position);
 297
 298         // Two acute accents, one from the composed a with acute accent, \u00e1,
 299         // and one separate.
 300         position = -42;
 301         UnicodeString  s3 = UnicodeString("abcd\\u00e1\\u0301xyz").unescape();
 302         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s3, &position, &status));
 303         TEST_ASSERT_SUCCESS(status);
 304         TEST_ASSERT_EQ(0, position);
 305     TEST_TEARDOWN;
 306 }
 307
 308 void IntlTestSpoof::testBug8654() {
 309     TEST_SETUP
 310         UnicodeString s = UnicodeString("B\\u00c1\\u0301").unescape();
 311         int32_t position = -42;
 312         TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s, &position, &status) & USPOOF_INVISIBLE );
 313         TEST_ASSERT_SUCCESS(status);
 314         TEST_ASSERT_EQ(0, position);
 315     TEST_TEARDOWN;
 316 }
 317
 318 static UnicodeString parseHex(const UnicodeString &in) {
 319     // Convert a series of hex numbers in a Unicode String to a string with the
 320     // corresponding characters.
 321     // The conversion is _really_ annoying.  There must be some function to just do it.
 322     UnicodeString result;
 323     UChar32 cc = 0;
 324     for (int32_t i=0; i<in.length(); i++) {
 325         UChar c = in.charAt(i);
 326         if (c == 0x20) {   // Space
 327             if (cc > 0) {
 328                result.append(cc);
 329                cc = 0;
 330             }
 331         } else if (c>=0x30 && c<=0x39) {
 332             cc = (cc<<4) + (c - 0x30);
 333         } else if ((c>=0x41 && c<=0x46) || (c>=0x61 && c<=0x66)) {
 334             cc = (cc<<4) + (c & 0x0f)+9;
 335         }
 336         // else do something with bad input.
 337     }
 338     if (cc > 0) {
 339         result.append(cc);
 340     }
 341     return result;
 342 }
 343
 344
 345 //
 346 // Append the hex form of a UChar32 to a UnicodeString.
 347 // Used in formatting error messages.
 348 // Match the formatting of numbers in confusables.txt
 349 // Minimum of 4 digits, no leading zeroes for positions 5 and up.
 350 //
 351 static void appendHexUChar(UnicodeString &dest, UChar32 c) {
 352     UBool   doZeroes = FALSE;
 353     for (int bitNum=28; bitNum>=0; bitNum-=4) {
 354         if (bitNum <= 12) {
 355             doZeroes = TRUE;
 356         }
 357         int hexDigit = (c>>bitNum) & 0x0f;
 358         if (hexDigit != 0 || doZeroes) {
 359             doZeroes = TRUE;
 360             dest.append((UChar)(hexDigit<=9? hexDigit + 0x30: hexDigit -10 + 0x41));
 361         }
 362     }
 363     dest.append((UChar)0x20);
 364 }
 365
 366 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
 367
 368 //  testConfData - Check each data item from the Unicode confusables.txt file,
 369 //                 verify that it transforms correctly in a skeleton.
 370 //
 371 void IntlTestSpoof::testConfData() {
 372     UErrorCode status = U_ZERO_ERROR;
 373
 374     const char *testDataDir = IntlTest::getSourceTestData(status);
 375     TEST_ASSERT_SUCCESS(status);
 376     char buffer[2000];
 377     uprv_strcpy(buffer, testDataDir);
 378     uprv_strcat(buffer, "confusables.txt");
 379
 380     LocalStdioFilePointer f(fopen(buffer, "rb"));
 381     if (f.isNull()) {
 382         errln("Skipping test spoof/testConfData.  File confusables.txt not accessible.");
 383         return;
 384     }
 385     fseek(f.getAlias(), 0, SEEK_END);
 386     int32_t  fileSize = ftell(f.getAlias());
 387     LocalArray<char> fileBuf(new char[fileSize]);
 388     fseek(f.getAlias(), 0, SEEK_SET);
 389     int32_t amt_read = fread(fileBuf.getAlias(), 1, fileSize, f.getAlias());
 390     TEST_ASSERT_EQ(amt_read, fileSize);
 391     TEST_ASSERT(fileSize>0);
 392     if (amt_read != fileSize || fileSize <=0) {
 393         return;
 394     }
 395     UnicodeString confusablesTxt = UnicodeString::fromUTF8(StringPiece(fileBuf.getAlias(), fileSize));
 396
 397     LocalUSpoofCheckerPointer sc(uspoof_open(&status));
 398     TEST_ASSERT_SUCCESS(status);
 399
 400     // Parse lines from the confusables.txt file.  Example Line:
 401     // FF44 ;   0064 ;  SL      # ( d -> d ) FULLWIDTH ....
 402     // Three fields.  The hex fields can contain more than one character,
 403     //                and each character may be more than 4 digits (for supplemntals)
 404     // This regular expression matches lines and splits the fields into capture groups.
 405     RegexMatcher parseLine("(?m)^([0-9A-F]{4}[^#;]*?);([^#;]*?);([^#]*)", confusablesTxt, 0, status);
 406     TEST_ASSERT_SUCCESS(status);
 407     while (parseLine.find()) {
 408         UnicodeString from = parseHex(parseLine.group(1, status));
 409         if (!Normalizer::isNormalized(from, UNORM_NFD, status)) {
 410             // The source character was not NFD.
 411             // Skip this case; the first step in obtaining a skeleton is to NFD the input,
 412             //  so the mapping in this line of confusables.txt will never be applied.
 413             continue;
 414         }
 415
 416         UnicodeString rawExpected = parseHex(parseLine.group(2, status));
 417         UnicodeString expected;
 418         Normalizer::decompose(rawExpected, FALSE /*NFD*/, 0, expected, status);
 419         TEST_ASSERT_SUCCESS(status);
 420
 421         int32_t skeletonType = 0;
 422         UnicodeString tableType = parseLine.group(3, status);
 423         TEST_ASSERT_SUCCESS(status);
 424         if (tableType.indexOf("SL") >= 0) {
 425             skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
 426         } else if (tableType.indexOf("SA") >= 0) {
 427             skeletonType = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
 428         } else if (tableType.indexOf("ML") >= 0) {
 429             skeletonType = 0;
 430         } else if (tableType.indexOf("MA") >= 0) {
 431             skeletonType = USPOOF_ANY_CASE;
 432         }
 433
 434         UnicodeString actual;
 435         uspoof_getSkeletonUnicodeString(sc.getAlias(), skeletonType, from, actual, &status);
 436         TEST_ASSERT_SUCCESS(status);
 437         TEST_ASSERT(actual == expected);
 438         if (actual != expected) {
 439             errln(parseLine.group(0, status));
 440             UnicodeString line = "Actual: ";
 441             int i = 0;
 442             while (i < actual.length()) {
 443                 appendHexUChar(line, actual.char32At(i));
 444                 i = actual.moveIndex32(i, 1);
 445             }
 446             errln(line);
 447         }
 448         if (U_FAILURE(status)) {
 449             break;
 450         }
 451     }
 452 }
 453
 454 // testIdentifierInfo. Note that IdentifierInfo is not public ICU API at this time
 455 void IntlTestSpoof::testIdentifierInfo() {
 456     UErrorCode status = U_ZERO_ERROR;
 457     ScriptSet bitset12; bitset12.set(USCRIPT_LATIN, status).set(USCRIPT_HANGUL, status);
 458     ScriptSet bitset2;  bitset2.set(USCRIPT_HANGUL, status);
 459     TEST_ASSERT(bitset12.contains(bitset2));
 460     TEST_ASSERT(bitset12.contains(bitset12));
 461     TEST_ASSERT(!bitset2.contains(bitset12));
 462
 463     ScriptSet arabSet;  arabSet.set(USCRIPT_ARABIC, status);
 464     ScriptSet latinSet; latinSet.set(USCRIPT_LATIN, status);
 465     UElement arabEl;  arabEl.pointer = &arabSet;
 466     UElement latinEl; latinEl.pointer = &latinSet;
 467     TEST_ASSERT(uhash_compareScriptSet(arabEl, latinEl) < 0);
 468     TEST_ASSERT(uhash_compareScriptSet(latinEl, arabEl) > 0);
 469
 470     UnicodeString scriptString;
 471     bitset12.displayScripts(scriptString);
 472     TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang Latn") == scriptString);
 473
 474     status = U_ZERO_ERROR;
 475     UHashtable *alternates = uhash_open(uhash_hashScriptSet ,uhash_compareScriptSet, NULL, &status);
 476     uhash_puti(alternates, &bitset12, 1, &status);
 477     uhash_puti(alternates, &bitset2, 1, &status);
 478     UnicodeString alternatesString;
 479     IdentifierInfo::displayAlternates(alternatesString, alternates, status);
 480     TEST_ASSERT(UNICODE_STRING_SIMPLE("Hang; Hang Latn") == alternatesString);
 481     TEST_ASSERT_SUCCESS(status);
 482
 483     status = U_ZERO_ERROR;
 484     ScriptSet tScriptSet;
 485     tScriptSet.parseScripts(scriptString, status);
 486     TEST_ASSERT_SUCCESS(status);
 487     TEST_ASSERT(bitset12 == tScriptSet);
 488     UnicodeString ss;
 489     ss.remove();
 490     uhash_close(alternates);
 491
 492     struct Test {
 493         const char         *fTestString;
 494         URestrictionLevel   fRestrictionLevel;
 495         const char         *fNumerics;
 496         const char         *fScripts;
 497         const char         *fAlternates;
 498         const char         *fCommonAlternates;
 499     } tests[] = {
 500             {"\\u0061\\u2665",                USPOOF_UNRESTRICTIVE,      "[]", "Latn", "", ""},
 501             {"\\u0061\\u3006",                USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hani Hira Kana", "Hani Hira Kana"},
 502             {"\\u0061\\u30FC\\u3006",         USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hira Kana", "Hira Kana"},
 503             {"\\u0061\\u30FC\\u3006\\u30A2",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
 504             {"\\u30A2\\u0061\\u30FC\\u3006",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
 505             {"\\u0061\\u0031\\u0661",         USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660]", "Latn", "Arab Thaa", "Arab Thaa"},
 506             {"\\u0061\\u0031\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660\\u06F0]", "Latn Arab", "", ""},
 507             {"\\u0661\\u30FC\\u3006\\u0061\\u30A2\\u0031\\u0967\\u06F1",  USPOOF_UNRESTRICTIVE,
 508                   "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab Deva", "", ""},
 509             {"\\u0061\\u30A2\\u30FC\\u3006\\u0031\\u0967\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE,
 510                   "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab Deva", "", ""}
 511     };
 512
 513     int testNum;
 514     for (testNum = 0; testNum < LENGTHOF(tests); testNum++) {
 515         char testNumStr[40];
 516         sprintf(testNumStr, "testNum = %d", testNum);
 517         Test &test = tests[testNum];
 518         status = U_ZERO_ERROR;
 519         UnicodeString testString(test.fTestString);  // Note: may do charset conversion.
 520         testString = testString.unescape();
 521         IdentifierInfo idInfo(status);
 522         TEST_ASSERT_SUCCESS(status);
 523         idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
 524         idInfo.setIdentifier(testString, status);
 525         TEST_ASSERT_MSG(*idInfo.getIdentifier() == testString, testNumStr);
 526
 527         URestrictionLevel restrictionLevel = test.fRestrictionLevel;
 528         TEST_ASSERT_MSG(restrictionLevel == idInfo.getRestrictionLevel(status), testNumStr);
 529
 530         status = U_ZERO_ERROR;
 531         UnicodeSet numerics(UnicodeString(test.fNumerics).unescape(), status);
 532         TEST_ASSERT_SUCCESS(status);
 533         TEST_ASSERT_MSG(numerics == *idInfo.getNumerics(), testNumStr);
 534
 535         ScriptSet scripts;
 536         scripts.parseScripts(UnicodeString(test.fScripts), status);
 537         TEST_ASSERT_MSG(scripts == *idInfo.getScripts(), testNumStr);
 538
 539         UnicodeString alternatesStr;
 540         IdentifierInfo::displayAlternates(alternatesStr, idInfo.getAlternates(), status);
 541         TEST_ASSERT_MSG(UnicodeString(test.fAlternates) == alternatesStr, testNumStr);
 542
 543         ScriptSet commonAlternates;
 544         commonAlternates.parseScripts(UnicodeString(test.fCommonAlternates), status);
 545         TEST_ASSERT_MSG(commonAlternates == *idInfo.getCommonAmongAlternates(), testNumStr);
 546     }
 547
 548     // Test of getScriptCount()
 549     //   Script and or Script Extension for chars used in the tests
 550     //     \\u3013  ; Bopo Hang Hani Hira Kana # So       GETA MARK
 551     //     \\uA838  ; Deva Gujr Guru Kthi Takr # Sc       NORTH INDIC RUPEE MARK
 552     //     \\u0951  ; Deva Latn                # Mn       DEVANAGARI STRESS SIGN UDATTA
 553     //
 554     //     \\u0370  ; Greek                    # L        GREEK CAPITAL LETTER HETA
 555     //     \\u0481  ; Cyrillic                 # L&       CYRILLIC SMALL LETTER KOPPA
 556     //     \\u0904  ; Devanagari               # Lo       DEVANAGARI LETTER SHORT A
 557     //     \\u3041  ; Hiragana                 # Lo       HIRAGANA LETTER SMALL A
 558     //     1234     ; Common                   #          ascii digits
 559     //     \\u0300  ; Inherited                # Mn       COMBINING GRAVE ACCENT
 560
 561     struct ScriptTest {
 562         const char *fTestString;
 563         int32_t     fScriptCount;
 564     } scriptTests[] = {
 565         {"Hello", 1},
 566         {"Hello\\u0370", 2},
 567         {"1234", 0},
 568         {"Hello1234\\u0300", 1},   // Common and Inherited are ignored.
 569         {"\\u0030", 0},
 570         {"abc\\u0951", 1},
 571         {"abc\\u3013", 2},
 572         {"\\uA838\\u0951", 1},     // Triggers commonAmongAlternates path.
 573         {"\\u3013\\uA838", 2}
 574     };
 575
 576     status = U_ZERO_ERROR;
 577     IdentifierInfo identifierInfo(status);
 578     for (testNum=0; testNum<LENGTHOF(scriptTests); testNum++) {
 579         ScriptTest &test = scriptTests[testNum];
 580         char msgBuf[100];
 581         sprintf(msgBuf, "testNum = %d ", testNum);
 582         UnicodeString testString = UnicodeString(test.fTestString).unescape();
 583
 584         status = U_ZERO_ERROR;
 585         identifierInfo.setIdentifier(testString, status);
 586         int32_t scriptCount = identifierInfo.getScriptCount();
 587         TEST_ASSERT_MSG(test.fScriptCount == scriptCount, msgBuf);
 588     }
 589 }
 590
 591 void IntlTestSpoof::testScriptSet() {
 592     ScriptSet s1;
 593     ScriptSet s2;
 594     UErrorCode status = U_ZERO_ERROR;
 595
 596     TEST_ASSERT(s1 == s2);
 597     s1.set(USCRIPT_ARABIC,status);
 598     TEST_ASSERT_SUCCESS(status);
 599     TEST_ASSERT(!(s1 == s2));
 600     TEST_ASSERT(s1.test(USCRIPT_ARABIC, status));
 601     TEST_ASSERT(s1.test(USCRIPT_GREEK, status) == FALSE);
 602
 603     status = U_ZERO_ERROR;
 604     s1.reset(USCRIPT_ARABIC, status);
 605     TEST_ASSERT(s1 == s2);
 606
 607     status = U_ZERO_ERROR;
 608     s1.setAll();
 609     TEST_ASSERT(s1.test(USCRIPT_COMMON, status));
 610     TEST_ASSERT(s1.test(USCRIPT_ETHIOPIC, status));
 611     TEST_ASSERT(s1.test(USCRIPT_CODE_LIMIT, status));
 612     s1.resetAll();
 613     TEST_ASSERT(!s1.test(USCRIPT_COMMON, status));
 614     TEST_ASSERT(!s1.test(USCRIPT_ETHIOPIC, status));
 615     TEST_ASSERT(!s1.test(USCRIPT_CODE_LIMIT, status));
 616
 617     status = U_ZERO_ERROR;
 618     s1.set(USCRIPT_TAKRI, status);
 619     s1.set(USCRIPT_BLISSYMBOLS, status);
 620     s2.setAll();
 621     TEST_ASSERT(s2.contains(s1));
 622     TEST_ASSERT(!s1.contains(s2));
 623     TEST_ASSERT(s2.intersects(s1));
 624     TEST_ASSERT(s1.intersects(s2));
 625     s2.reset(USCRIPT_TAKRI, status);
 626     TEST_ASSERT(!s2.contains(s1));
 627     TEST_ASSERT(!s1.contains(s2));
 628     TEST_ASSERT(s1.intersects(s2));
 629     TEST_ASSERT(s2.intersects(s1));
 630     TEST_ASSERT_SUCCESS(status);
 631
 632     status = U_ZERO_ERROR;
 633     s1.resetAll();
 634     s1.set(USCRIPT_NKO, status);
 635     s1.set(USCRIPT_COMMON, status);
 636     s2 = s1;
 637     TEST_ASSERT(s2 == s1);
 638     TEST_ASSERT_EQ(2, s2.countMembers());
 639     s2.intersect(s1);
 640     TEST_ASSERT(s2 == s1);
 641     s2.setAll();
 642     TEST_ASSERT(!(s2 == s1));
 643     TEST_ASSERT(s2.countMembers() >= USCRIPT_CODE_LIMIT);
 644     s2.intersect(s1);
 645     TEST_ASSERT(s2 == s1);
 646
 647     s2.setAll();
 648     s2.reset(USCRIPT_COMMON, status);
 649     s2.intersect(s1);
 650     TEST_ASSERT(s2.countMembers() == 1);
 651
 652     s1.resetAll();
 653     s1.set(USCRIPT_AFAKA, status);
 654     s1.set(USCRIPT_VAI, status);
 655     s1.set(USCRIPT_INHERITED, status);
 656     int32_t n = -1;
 657     for (int32_t i=0; i<4; i++) {
 658         n = s1.nextSetBit(n+1);
 659         switch (i) {
 660           case 0: TEST_ASSERT_EQ(USCRIPT_INHERITED, n); break;
 661           case 1: TEST_ASSERT_EQ(USCRIPT_VAI, n); break;
 662           case 2: TEST_ASSERT_EQ(USCRIPT_AFAKA, n); break;
 663           case 3: TEST_ASSERT_EQ(-1, (int32_t)n); break;
 664           default: TEST_ASSERT(FALSE);
 665         }
 666     }
 667     TEST_ASSERT_SUCCESS(status);
 668 }
 669
 670
 671 void IntlTestSpoof::testRestrictionLevel() {
 672     struct Test {
 673         const char         *fId;
 674         URestrictionLevel   fExpectedRestrictionLevel;
 675     } tests[] = {
 676         {"\\u0061\\u03B3\\u2665", USPOOF_UNRESTRICTIVE},
 677         {"a",                     USPOOF_ASCII},
 678         {"\\u03B3",               USPOOF_HIGHLY_RESTRICTIVE},
 679         {"\\u0061\\u30A2\\u30FC", USPOOF_HIGHLY_RESTRICTIVE},
 680         {"\\u0061\\u0904",        USPOOF_MODERATELY_RESTRICTIVE},
 681         {"\\u0061\\u03B3",        USPOOF_MINIMALLY_RESTRICTIVE}
 682     };
 683     char msgBuffer[100];
 684
 685     URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_HIGHLY_RESTRICTIVE,
 686          USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE, USPOOF_UNRESTRICTIVE};
 687
 688     UErrorCode status = U_ZERO_ERROR;
 689     IdentifierInfo idInfo(status);
 690     TEST_ASSERT_SUCCESS(status);
 691     idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
 692     TEST_ASSERT_SUCCESS(status);
 693     for (int32_t testNum=0; testNum < LENGTHOF(tests); testNum++) {
 694         status = U_ZERO_ERROR;
 695         const Test &test = tests[testNum];
 696         UnicodeString testString = UnicodeString(test.fId).unescape();
 697         URestrictionLevel expectedLevel = test.fExpectedRestrictionLevel;
 698         idInfo.setIdentifier(testString, status);
 699         sprintf(msgBuffer, "testNum = %d ", testNum);
 700         TEST_ASSERT_SUCCESS(status);
 701         TEST_ASSERT_MSG(expectedLevel == idInfo.getRestrictionLevel(status), msgBuffer);
 702         for (int levelIndex=0; levelIndex<LENGTHOF(restrictionLevels); levelIndex++) {
 703             status = U_ZERO_ERROR;
 704             URestrictionLevel levelSetInSpoofChecker = restrictionLevels[levelIndex];
 705             USpoofChecker *sc = uspoof_open(&status);
 706             uspoof_setChecks(sc, USPOOF_RESTRICTION_LEVEL, &status);
 707             uspoof_setAllowedChars(sc, uspoof_getRecommendedSet(&status), &status);
 708             uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
 709             UBool actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status) != 0;
 710
 711             // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
 712             UBool expectedFailure = expectedLevel > levelSetInSpoofChecker ||
 713                                     !uspoof_getRecommendedUnicodeSet(&status)->containsAll(testString);
 714             sprintf(msgBuffer, "testNum = %d, levelIndex = %d", testNum, levelIndex);
 715             TEST_ASSERT_MSG(expectedFailure == actualValue, msgBuffer);
 716             TEST_ASSERT_SUCCESS(status);
 717             uspoof_close(sc);
 718         }
 719     }
 720 }
 721
 722
 723 void IntlTestSpoof::testMixedNumbers() {
 724     struct Test {
 725         const char *fTestString;
 726         const char *fExpectedSet;
 727     } tests[] = {
 728         {"1",              "[0]"},
 729         {"\\u0967",        "[\\u0966]"},
 730         {"1\\u0967",       "[0\\u0966]"},
 731         {"\\u0661\\u06F1", "[\\u0660\\u06F0]"}
 732     };
 733     UErrorCode status = U_ZERO_ERROR;
 734     IdentifierInfo idInfo(status);
 735     for (int32_t testNum=0; testNum < LENGTHOF(tests); testNum++) {
 736         char msgBuf[100];
 737         sprintf(msgBuf, "testNum = %d ", testNum);
 738         Test &test = tests[testNum];
 739
 740         status = U_ZERO_ERROR;
 741         UnicodeString testString = UnicodeString(test.fTestString).unescape();
 742         UnicodeSet expectedSet(UnicodeString(test.fExpectedSet).unescape(), status);
 743         idInfo.setIdentifier(testString, status);
 744         TEST_ASSERT_SUCCESS(status);
 745         TEST_ASSERT_MSG(expectedSet == *idInfo.getNumerics(), msgBuf);
 746
 747         status = U_ZERO_ERROR;
 748         USpoofChecker *sc = uspoof_open(&status);
 749         uspoof_setChecks(sc, USPOOF_MIXED_NUMBERS, &status); // only check this
 750         int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
 751         UBool mixedNumberFailure = ((result & USPOOF_MIXED_NUMBERS) != 0);
 752         TEST_ASSERT_MSG((expectedSet.size() > 1) == mixedNumberFailure, msgBuf);
 753         uspoof_close(sc);
 754     }
 755 }
 756
 757 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO */