1 /*
2 **********************************************************************
3 * Copyright (C) 2009, 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
14
15 #include "itspoof.h"
16 #include "unicode/uspoof.h"
17
18 #define TEST_ASSERT_SUCCESS(status) {if (U_FAILURE(status)) { \
19 errcheckln(status, "Failure at file %s, line %d, error = %s", __FILE__, __LINE__, u_errorName(status));}}
20
21 #define TEST_ASSERT(expr) {if ((expr)==FALSE) { \
22 errln("Test Failure at file %s, line %d: \"%s\" is false.\n", __FILE__, __LINE__, #expr);};}
23
24 #define TEST_ASSERT_EQ(a, b) { if ((a) != (b)) { \
25 errln("Test Failure at file %s, line %d: \"%s\" (%d) != \"%s\" (%d) \n", \
26 __FILE__, __LINE__, #a, (a), #b, (b)); }}
27
28 #define TEST_ASSERT_NE(a, b) { if ((a) == (b)) { \
29 errln("Test Failure at file %s, line %d: \"%s\" (%d) == \"%s\" (%d) \n", \
30 __FILE__, __LINE__, #a, (a), #b, (b)); }}
31
32 /*
33 * TEST_SETUP and TEST_TEARDOWN
34 * macros to handle the boilerplate around setting up test case.
35 * Put arbitrary test code between SETUP and TEARDOWN.
36 * "sc" is the ready-to-go SpoofChecker for use in the tests.
37 */
38 #define TEST_SETUP { \
39 UErrorCode status = U_ZERO_ERROR; \
40 USpoofChecker *sc; \
41 sc = uspoof_open(&status); \
42 TEST_ASSERT_SUCCESS(status); \
43 if (U_SUCCESS(status)){
44
45 #define TEST_TEARDOWN \
46 } \
47 TEST_ASSERT_SUCCESS(status); \
48 uspoof_close(sc); \
49 }
50
51
52
53
runIndexedTest(int32_t index,UBool exec,const char * & name,char *)54 void IntlTestSpoof::runIndexedTest( int32_t index, UBool exec, const char* &name, char* /*par*/ )
55 {
56 if (exec) logln("TestSuite spoof: ");
57 switch (index) {
58 case 0:
59 name = "TestSpoofAPI";
60 if (exec) {
61 testSpoofAPI();
62 }
63 break;
64 case 1:
65 name = "TestSkeleton";
66 if (exec) {
67 testSkeleton();
68 }
69 break;
70 case 2:
71 name = "TestAreConfusable";
72 if (exec) {
73 testAreConfusable();
74 }
75 break;
76 case 3:
77 name = "TestInvisible";
78 if (exec) {
79 testInvisible();
80 }
81 break;
82 default: name=""; break;
83 }
84 }
85
testSpoofAPI()86 void IntlTestSpoof::testSpoofAPI() {
87
88 TEST_SETUP
89 UnicodeString s("uvw");
90 int32_t position = 666;
91 int32_t checkResults = uspoof_checkUnicodeString(sc, s, &position, &status);
92 TEST_ASSERT_SUCCESS(status);
93 TEST_ASSERT_EQ(0, checkResults);
94 TEST_ASSERT_EQ(666, position);
95 TEST_TEARDOWN;
96
97 TEST_SETUP
98 UnicodeString s1("cxs");
99 UnicodeString s2 = UnicodeString("\\u0441\\u0445\\u0455").unescape(); // Cyrillic "cxs"
100 int32_t checkResults = uspoof_areConfusableUnicodeString(sc, s1, s2, &status);
101 TEST_ASSERT_EQ(USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE, checkResults);
102
103 TEST_TEARDOWN;
104
105 TEST_SETUP
106 UnicodeString s("I1l0O");
107 UnicodeString dest;
108 UnicodeString &retStr = uspoof_getSkeletonUnicodeString(sc, USPOOF_ANY_CASE, s, dest, &status);
109 TEST_ASSERT_SUCCESS(status);
110 TEST_ASSERT(UnicodeString("11100") == dest);
111 TEST_ASSERT(&dest == &retStr);
112 TEST_TEARDOWN;
113 }
114
115
116 #define CHECK_SKELETON(type, input, expected) { \
117 checkSkeleton(sc, type, input, expected, __LINE__); \
118 }
119
120
121 // testSkeleton. Spot check a number of confusable skeleton substitutions from the
122 // Unicode data file confusables.txt
123 // Test cases chosen for substitutions of various lengths, and
124 // membership in different mapping tables.
testSkeleton()125 void IntlTestSpoof::testSkeleton() {
126 const uint32_t ML = 0;
127 const uint32_t SL = USPOOF_SINGLE_SCRIPT_CONFUSABLE;
128 const uint32_t MA = USPOOF_ANY_CASE;
129 const uint32_t SA = USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_ANY_CASE;
130
131 TEST_SETUP
132 // A long "identifier" that will overflow implementation stack buffers, forcing heap allocations.
133 CHECK_SKELETON(SL, " A long 'identifier' that will overflow implementation stack buffers, forcing heap allocations."
134 " A long 'identifier' that will overflow implementation stack buffers, forcing heap allocations."
135 " A long 'identifier' that will overflow implementation stack buffers, forcing heap allocations."
136 " A long 'identifier' that will overflow implementation stack buffers, forcing heap allocations.",
137
138 " A 1ong \\u02b9identifier\\u02b9 that wi11 overf1ow imp1ementation stack buffers, forcing heap a11ocations."
139 " A 1ong \\u02b9identifier\\u02b9 that wi11 overf1ow imp1ementation stack buffers, forcing heap a11ocations."
140 " A 1ong \\u02b9identifier\\u02b9 that wi11 overf1ow imp1ementation stack buffers, forcing heap a11ocations."
141 " A 1ong \\u02b9identifier\\u02b9 that wi11 overf1ow imp1ementation stack buffers, forcing heap a11ocations.")
142
143 // FC5F ; FE74 0651 ; ML #* ARABIC LIGATURE SHADDA WITH KASRATAN ISOLATED FORM to
144 // ARABIC KASRATAN ISOLATED FORM, ARABIC SHADDA
145 // This character NFKD normalizes to \u0020 \u064d \u0651, so its confusable mapping
146 // is never used in creating a skeleton.
147 CHECK_SKELETON(SL, "\\uFC5F", " \\u064d\\u0651");
148
149 CHECK_SKELETON(SL, "nochange", "nochange");
150 CHECK_SKELETON(MA, "love", "1ove"); // lower case l to digit 1
151 CHECK_SKELETON(ML, "OOPS", "OOPS");
152 CHECK_SKELETON(MA, "OOPS", "00PS"); // Letter O to digit 0 in any case mode only
153 CHECK_SKELETON(SL, "\\u059c", "\\u0301");
154 CHECK_SKELETON(SL, "\\u2A74", "\\u003A\\u003A\\u003D");
155 CHECK_SKELETON(SL, "\\u247E", "\\u0028\\u0031\\u0031\\u0029");
156 CHECK_SKELETON(SL, "\\uFDFB", "\\u062C\\u0644\\u0020\\u062C\\u0644\\u0627\\u0644\\u0647");
157
158 // This mapping exists in the ML and MA tables, does not exist in SL, SA
159 //0C83 ; 0C03 ; ML # ( ಃ → ః ) KANNADA SIGN VISARGA → TELUGU SIGN VISARGA # {source:513}
160 CHECK_SKELETON(SL, "\\u0C83", "\\u0C83");
161 CHECK_SKELETON(SA, "\\u0C83", "\\u0C83");
162 CHECK_SKELETON(ML, "\\u0C83", "\\u0C03");
163 CHECK_SKELETON(MA, "\\u0C83", "\\u0C03");
164
165 // 0391 ; 0041 ; MA # ( Α → A ) GREEK CAPITAL LETTER ALPHA to LATIN CAPITAL LETTER A
166 // This mapping exists only in the MA table.
167 CHECK_SKELETON(MA, "\\u0391", "A");
168 CHECK_SKELETON(SA, "\\u0391", "\\u0391");
169 CHECK_SKELETON(ML, "\\u0391", "\\u0391");
170 CHECK_SKELETON(SL, "\\u0391", "\\u0391");
171
172 // 13CF ; 0062 ; MA # CHEROKEE LETTER SI to LATIN SMALL LETTER B
173 // This mapping exists in the ML and MA tables
174 CHECK_SKELETON(ML, "\\u13CF", "b");
175 CHECK_SKELETON(MA, "\\u13CF", "b");
176 CHECK_SKELETON(SL, "\\u13CF", "\\u13CF");
177 CHECK_SKELETON(SA, "\\u13CF", "\\u13CF");
178
179 // 0022 ; 02B9 02B9 ; SA #* QUOTATION MARK to MODIFIER LETTER PRIME, MODIFIER LETTER PRIME
180 // all tables.
181 CHECK_SKELETON(SL, "\\u0022", "\\u02B9\\u02B9");
182 CHECK_SKELETON(SA, "\\u0022", "\\u02B9\\u02B9");
183 CHECK_SKELETON(ML, "\\u0022", "\\u02B9\\u02B9");
184 CHECK_SKELETON(MA, "\\u0022", "\\u02B9\\u02B9");
185
186 TEST_TEARDOWN;
187 }
188
189
190 //
191 // Run a single confusable skeleton transformation test case.
192 //
checkSkeleton(const USpoofChecker * sc,uint32_t type,const char * input,const char * expected,int32_t lineNum)193 void IntlTestSpoof::checkSkeleton(const USpoofChecker *sc, uint32_t type,
194 const char *input, const char *expected, int32_t lineNum) {
195 UnicodeString uInput = UnicodeString(input).unescape();
196 UnicodeString uExpected = UnicodeString(expected).unescape();
197
198 UErrorCode status = U_ZERO_ERROR;
199 UnicodeString actual;
200 uspoof_getSkeletonUnicodeString(sc, type, uInput, actual, &status);
201 if (U_FAILURE(status)) {
202 errln("File %s, Line %d, Test case from line %d, status is %s", __FILE__, __LINE__, lineNum,
203 u_errorName(status));
204 return;
205 }
206 if (uExpected != actual) {
207 errln("File %s, Line %d, Test case from line %d, Actual and Expected skeletons differ.",
208 __FILE__, __LINE__, lineNum);
209 errln(UnicodeString(" Actual Skeleton: \"") + actual + UnicodeString("\"\n") +
210 UnicodeString(" Expected Skeleton: \"") + uExpected + UnicodeString("\""));
211 }
212 }
213
testAreConfusable()214 void IntlTestSpoof::testAreConfusable() {
215 TEST_SETUP
216 UnicodeString s1("A long string that will overflow stack buffers. A long string that will overflow stack buffers. "
217 "A long string that will overflow stack buffers. A long string that will overflow stack buffers. ");
218 UnicodeString s2("A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. "
219 "A long string that wi11 overflow stack buffers. A long string that will overflow stack buffers. ");
220 TEST_ASSERT_EQ(USPOOF_SINGLE_SCRIPT_CONFUSABLE, uspoof_areConfusableUnicodeString(sc, s1, s2, &status));
221 TEST_ASSERT_SUCCESS(status);
222
223 TEST_TEARDOWN;
224 }
225
testInvisible()226 void IntlTestSpoof::testInvisible() {
227 TEST_SETUP
228 UnicodeString s = UnicodeString("abcd\\u0301ef").unescape();
229 int32_t position = -42;
230 TEST_ASSERT_EQ(0, uspoof_checkUnicodeString(sc, s, &position, &status));
231 TEST_ASSERT_SUCCESS(status);
232 TEST_ASSERT(position == -42);
233
234 UnicodeString s2 = UnicodeString("abcd\\u0301\\u0302\\u0301ef").unescape();
235 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s2, &position, &status));
236 TEST_ASSERT_SUCCESS(status);
237 TEST_ASSERT_EQ(7, position);
238
239 // Tow acute accents, one from the composed a with acute accent, \u00e1,
240 // and one separate.
241 position = -42;
242 UnicodeString s3 = UnicodeString("abcd\\u00e1\\u0301xyz").unescape();
243 TEST_ASSERT_EQ(USPOOF_INVISIBLE, uspoof_checkUnicodeString(sc, s3, &position, &status));
244 TEST_ASSERT_SUCCESS(status);
245 TEST_ASSERT_EQ(7, position);
246 TEST_TEARDOWN;
247 }
248 #endif // UCONFIG_NO_REGULAR_EXPRESSIONS
249
250