Update ICU to 54.1 step 1

source/test/intltest/itspoof.cpp

 /*
 **********************************************************************
-* Copyright (C) 2011-2013, International Business Machines Corporation 
+* Copyright (C) 2011-2014, International Business Machines Corporation 
 * and others.  All Rights Reserved.
 **********************************************************************
 */
 /**
  * IntlTestSpoof tests for USpoofDetector
  */
 
 #include "unicode/utypes.h"
 
 #if !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO
@@ skipping 24 matching lines @@
     dataerrln("Test Failure at file %s, line %d, %s: \"%s\" is false.", __FILE__, __LINE__, msg, #expr);};}
 
 #define TEST_ASSERT_EQ(a, b) { if ((a) != (b)) { \
     errln("Test Failure at file %s, line %d: \"%s\" (%d) != \"%s\" (%d)", \
              __FILE__, __LINE__, #a, (a), #b, (b)); }}
 
 #define TEST_ASSERT_NE(a, b) { if ((a) == (b)) { \
     errln("Test Failure at file %s, line %d: \"%s\" (%d) == \"%s\" (%d)", \
              __FILE__, __LINE__, #a, (a), #b, (b)); }}
 
-#define LENGTHOF(array) ((int32_t)(sizeof(array)/sizeof((array)[0])))
-
 /*
  *   TEST_SETUP and TEST_TEARDOWN
  *         macros to handle the boilerplate around setting up test case.
  *         Put arbitrary test code between SETUP and TEARDOWN.
  *         "sc" is the ready-to-go  SpoofChecker for use in the tests.
  */
 #define TEST_SETUP {  \
     UErrorCode status = U_ZERO_ERROR; \
     USpoofChecker *sc;     \
     sc = uspoof_open(&status);  \
@@ skipping 302 matching lines @@
     }
     dest.append((UChar)0x20);
 }
 
 U_DEFINE_LOCAL_OPEN_POINTER(LocalStdioFilePointer, FILE, fclose);
 
 //  testConfData - Check each data item from the Unicode confusables.txt file,
 //                 verify that it transforms correctly in a skeleton.
 //
 void IntlTestSpoof::testConfData() {
-    UErrorCode status = U_ZERO_ERROR;
-
-    const char *testDataDir = IntlTest::getSourceTestData(status);
-    TEST_ASSERT_SUCCESS(status);
     char buffer[2000];
-    uprv_strcpy(buffer, testDataDir);
+    if (getUnidataPath(buffer) == NULL) {
+        errln("Skipping test spoof/testConfData. Unable to find path to source/data/unidata/.");
+        return;
+    }
     uprv_strcat(buffer, "confusables.txt");
 
     LocalStdioFilePointer f(fopen(buffer, "rb"));
     if (f.isNull()) {
         errln("Skipping test spoof/testConfData.  File confusables.txt not accessible.");
         return;
     }
     fseek(f.getAlias(), 0, SEEK_END);
     int32_t  fileSize = ftell(f.getAlias());
     LocalArray<char> fileBuf(new char[fileSize]);
     fseek(f.getAlias(), 0, SEEK_SET);
     int32_t amt_read = fread(fileBuf.getAlias(), 1, fileSize, f.getAlias());
     TEST_ASSERT_EQ(amt_read, fileSize);
     TEST_ASSERT(fileSize>0);
     if (amt_read != fileSize || fileSize <=0) {
         return;
     }
     UnicodeString confusablesTxt = UnicodeString::fromUTF8(StringPiece(fileBuf.getAlias(), fileSize));
 
+    UErrorCode status = U_ZERO_ERROR;
     LocalUSpoofCheckerPointer sc(uspoof_open(&status));
     TEST_ASSERT_SUCCESS(status);
 
     // Parse lines from the confusables.txt file.  Example Line:
     // FF44 ;   0064 ;  SL      # ( d -> d ) FULLWIDTH ....
     // Three fields.  The hex fields can contain more than one character,
     //                and each character may be more than 4 digits (for supplemntals)
     // This regular expression matches lines and splits the fields into capture groups.
     RegexMatcher parseLine("(?m)^([0-9A-F]{4}[^#;]*?);([^#;]*?);([^#]*)", confusablesTxt, 0, status);
     TEST_ASSERT_SUCCESS(status);
@@ skipping 91 matching lines @@
         const char         *fCommonAlternates;
     } tests[] = {
             {"\\u0061\\u2665",                USPOOF_UNRESTRICTIVE,      "[]", "Latn", "", ""},
             {"\\u0061\\u3006",                USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hani Hira Kana", "Hani Hira Kana"},
             {"\\u0061\\u30FC\\u3006",         USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn", "Hira Kana", "Hira Kana"},
             {"\\u0061\\u30FC\\u3006\\u30A2",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
             {"\\u30A2\\u0061\\u30FC\\u3006",  USPOOF_HIGHLY_RESTRICTIVE, "[]", "Latn Kana", "", ""},
             {"\\u0061\\u0031\\u0661",         USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660]", "Latn", "Arab Thaa", "Arab Thaa"},
             {"\\u0061\\u0031\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE,      "[\\u0030\\u0660\\u06F0]", "Latn Arab", "", ""},
             {"\\u0661\\u30FC\\u3006\\u0061\\u30A2\\u0031\\u0967\\u06F1",  USPOOF_UNRESTRICTIVE, 
-                  "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi", "Deva Kthi"},
+                  "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi Mahj", "Deva Kthi Mahj"},
             {"\\u0061\\u30A2\\u30FC\\u3006\\u0031\\u0967\\u0661\\u06F1",  USPOOF_UNRESTRICTIVE, 
-                  "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi", "Deva Kthi"}
+                  "[\\u0030\\u0660\\u06F0\\u0966]", "Latn Kana Arab", "Deva Kthi Mahj", "Deva Kthi Mahj"}
     };
 
     int testNum;
-    for (testNum = 0; testNum < LENGTHOF(tests); testNum++) {
+    for (testNum = 0; testNum < UPRV_LENGTHOF(tests); testNum++) {
         char testNumStr[40];
         sprintf(testNumStr, "testNum = %d", testNum);
         Test &test = tests[testNum];
         status = U_ZERO_ERROR;
         UnicodeString testString(test.fTestString);  // Note: may do charset conversion.
         testString = testString.unescape();
         IdentifierInfo idInfo(status);
         TEST_ASSERT_SUCCESS(status);
         idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
         idInfo.setIdentifier(testString, status);
@@ skipping 43 matching lines @@
         {"Hello1234\\u0300", 1},   // Common and Inherited are ignored.
         {"\\u0030", 0},
         {"abc\\u0951", 1},
         {"abc\\u3013", 2},
         {"\\uA838\\u0951", 1},     // Triggers commonAmongAlternates path.
         {"\\u3013\\uA838", 2}
     };
 
     status = U_ZERO_ERROR;
     IdentifierInfo identifierInfo(status);
-    for (testNum=0; testNum<LENGTHOF(scriptTests); testNum++) {
+    for (testNum=0; testNum<UPRV_LENGTHOF(scriptTests); testNum++) {
         ScriptTest &test = scriptTests[testNum];
         char msgBuf[100];
         sprintf(msgBuf, "testNum = %d ", testNum);
         UnicodeString testString = UnicodeString(test.fTestString).unescape();
         
         status = U_ZERO_ERROR;
         identifierInfo.setIdentifier(testString, status);
         int32_t scriptCount = identifierInfo.getScriptCount();
         TEST_ASSERT_MSG(test.fScriptCount == scriptCount, msgBuf);
     }
@@ skipping 79 matching lines @@
 }
 
 
 void IntlTestSpoof::testRestrictionLevel() {
     struct Test {
         const char         *fId;
         URestrictionLevel   fExpectedRestrictionLevel;
     } tests[] = {
         {"\\u0061\\u03B3\\u2665", USPOOF_UNRESTRICTIVE},
         {"a",                     USPOOF_ASCII},
-        {"\\u03B3",               USPOOF_HIGHLY_RESTRICTIVE},
+        {"\\u03B3",               USPOOF_SINGLE_SCRIPT_RESTRICTIVE},
         {"\\u0061\\u30A2\\u30FC", USPOOF_HIGHLY_RESTRICTIVE},
         {"\\u0061\\u0904",        USPOOF_MODERATELY_RESTRICTIVE},
         {"\\u0061\\u03B3",        USPOOF_MINIMALLY_RESTRICTIVE}
     };
     char msgBuffer[100];
 
-    URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_HIGHLY_RESTRICTIVE, 
-         USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE, USPOOF_UNRESTRICTIVE};
+    URestrictionLevel restrictionLevels[] = { USPOOF_ASCII, USPOOF_SINGLE_SCRIPT_RESTRICTIVE, 
+         USPOOF_HIGHLY_RESTRICTIVE, USPOOF_MODERATELY_RESTRICTIVE, USPOOF_MINIMALLY_RESTRICTIVE, 
+         USPOOF_UNRESTRICTIVE};
     
     UErrorCode status = U_ZERO_ERROR;
     IdentifierInfo idInfo(status);
     TEST_ASSERT_SUCCESS(status);
     idInfo.setIdentifierProfile(*uspoof_getRecommendedUnicodeSet(&status));
     TEST_ASSERT_SUCCESS(status);
-    for (int32_t testNum=0; testNum < LENGTHOF(tests); testNum++) {
+    for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
         status = U_ZERO_ERROR;
         const Test &test = tests[testNum];
         UnicodeString testString = UnicodeString(test.fId).unescape();
         URestrictionLevel expectedLevel = test.fExpectedRestrictionLevel;
         idInfo.setIdentifier(testString, status);
         sprintf(msgBuffer, "testNum = %d ", testNum);
         TEST_ASSERT_SUCCESS(status);
         TEST_ASSERT_MSG(expectedLevel == idInfo.getRestrictionLevel(status), msgBuffer);
-        for (int levelIndex=0; levelIndex<LENGTHOF(restrictionLevels); levelIndex++) {
+        for (int levelIndex=0; levelIndex<UPRV_LENGTHOF(restrictionLevels); levelIndex++) {
             status = U_ZERO_ERROR;
             URestrictionLevel levelSetInSpoofChecker = restrictionLevels[levelIndex];
             USpoofChecker *sc = uspoof_open(&status);
             uspoof_setChecks(sc, USPOOF_RESTRICTION_LEVEL, &status);
             uspoof_setAllowedChars(sc, uspoof_getRecommendedSet(&status), &status);
             uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
-            UBool actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status) != 0;
+            int32_t actualValue = uspoof_checkUnicodeString(sc, testString, NULL, &status);
+            
+            // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
+            int32_t expectedValue = 0;
+            if (expectedLevel > levelSetInSpoofChecker) {
+                expectedValue |= USPOOF_RESTRICTION_LEVEL;
+            }
+            if (!uspoof_getRecommendedUnicodeSet(&status)->containsAll(testString)) {
+                expectedValue |= USPOOF_CHAR_LIMIT;
+            }
+            sprintf(msgBuffer, "testNum = %d, levelIndex = %d, expected = %#x, actual = %#x",
+                    testNum, levelIndex, expectedValue, actualValue);
+            TEST_ASSERT_MSG(expectedValue == actualValue, msgBuffer);
+            TEST_ASSERT_SUCCESS(status);
 
-            // we want to fail if the text is (say) MODERATE and the testLevel is ASCII
-            UBool expectedFailure = expectedLevel > levelSetInSpoofChecker ||
-                                    !uspoof_getRecommendedUnicodeSet(&status)->containsAll(testString);
-            sprintf(msgBuffer, "testNum = %d, levelIndex = %d", testNum, levelIndex);
-            TEST_ASSERT_MSG(expectedFailure == actualValue, msgBuffer);
+            // Run the same check again, with the Spoof Checker configured to return
+            // the actual restriction level.
+            uspoof_setChecks(sc, USPOOF_AUX_INFO | USPOOF_RESTRICTION_LEVEL, &status);
+            uspoof_setAllowedChars(sc, uspoof_getRecommendedSet(&status), &status);
+            uspoof_setRestrictionLevel(sc, levelSetInSpoofChecker);
+            int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
             TEST_ASSERT_SUCCESS(status);
+            if (U_SUCCESS(status)) {
+                TEST_ASSERT_EQ(expectedLevel, result & USPOOF_RESTRICTION_LEVEL_MASK);
+                TEST_ASSERT_EQ(expectedValue, result & USPOOF_ALL_CHECKS);
+            }
             uspoof_close(sc);
         }
     }
 }
 
 
 void IntlTestSpoof::testMixedNumbers() {
     struct Test {
         const char *fTestString;
         const char *fExpectedSet;
     } tests[] = {
         {"1",              "[0]"},
         {"\\u0967",        "[\\u0966]"},
         {"1\\u0967",       "[0\\u0966]"},
         {"\\u0661\\u06F1", "[\\u0660\\u06F0]"}
     };
     UErrorCode status = U_ZERO_ERROR;
     IdentifierInfo idInfo(status);
-    for (int32_t testNum=0; testNum < LENGTHOF(tests); testNum++) {
+    for (int32_t testNum=0; testNum < UPRV_LENGTHOF(tests); testNum++) {
         char msgBuf[100];
         sprintf(msgBuf, "testNum = %d ", testNum);
         Test &test = tests[testNum];
 
         status = U_ZERO_ERROR;
         UnicodeString testString = UnicodeString(test.fTestString).unescape();
         UnicodeSet expectedSet(UnicodeString(test.fExpectedSet).unescape(), status);
         idInfo.setIdentifier(testString, status);
         TEST_ASSERT_SUCCESS(status);
         TEST_ASSERT_MSG(expectedSet == *idInfo.getNumerics(), msgBuf);
 
         status = U_ZERO_ERROR;
         USpoofChecker *sc = uspoof_open(&status);
         uspoof_setChecks(sc, USPOOF_MIXED_NUMBERS, &status); // only check this
         int32_t result = uspoof_checkUnicodeString(sc, testString, NULL, &status);
         UBool mixedNumberFailure = ((result & USPOOF_MIXED_NUMBERS) != 0);
         TEST_ASSERT_MSG((expectedSet.size() > 1) == mixedNumberFailure, msgBuf);
         uspoof_close(sc);
     }
 }
 
 #endif /* !UCONFIG_NO_REGULAR_EXPRESSIONS && !UCONFIG_NO_NORMALIZATION && !UCONFIG_NO_FILE_IO */