1 | // © 2016 and later: Unicode, Inc. and others. |
2 | // License & terms of use: http://www.unicode.org/copyright.html |
3 | /* |
4 | *************************************************************************** |
5 | * Copyright (C) 2008-2015, International Business Machines Corporation |
6 | * and others. All Rights Reserved. |
7 | *************************************************************************** |
8 | * file name: uspoof.cpp |
9 | * encoding: UTF-8 |
10 | * tab size: 8 (not used) |
11 | * indentation:4 |
12 | * |
13 | * created on: 2008Feb13 |
14 | * created by: Andy Heninger |
15 | * |
16 | * Unicode Spoof Detection |
17 | */ |
18 | #include "unicode/utypes.h" |
19 | #include "unicode/normalizer2.h" |
20 | #include "unicode/uspoof.h" |
21 | #include "unicode/ustring.h" |
22 | #include "unicode/utf16.h" |
23 | #include "cmemory.h" |
24 | #include "cstring.h" |
25 | #include "mutex.h" |
26 | #include "scriptset.h" |
27 | #include "uassert.h" |
28 | #include "ucln_in.h" |
29 | #include "uspoof_impl.h" |
30 | #include "umutex.h" |
31 | |
32 | |
33 | #if !UCONFIG_NO_NORMALIZATION |
34 | |
35 | U_NAMESPACE_USE |
36 | |
37 | |
38 | // |
39 | // Static Objects used by the spoof impl, their thread safe initialization and their cleanup. |
40 | // |
41 | static UnicodeSet *gInclusionSet = nullptr; |
42 | static UnicodeSet *gRecommendedSet = nullptr; |
43 | static const Normalizer2 *gNfdNormalizer = nullptr; |
44 | static UInitOnce gSpoofInitStaticsOnce {}; |
45 | |
46 | namespace { |
47 | |
48 | UBool U_CALLCONV |
49 | uspoof_cleanup() { |
50 | delete gInclusionSet; |
51 | gInclusionSet = nullptr; |
52 | delete gRecommendedSet; |
53 | gRecommendedSet = nullptr; |
54 | gNfdNormalizer = nullptr; |
55 | gSpoofInitStaticsOnce.reset(); |
56 | return true; |
57 | } |
58 | |
59 | void U_CALLCONV initializeStatics(UErrorCode &status) { |
60 | static const char16_t *inclusionPat = |
61 | u"['\\-.\\:\\u00B7\\u0375\\u058A\\u05F3\\u05F4\\u06FD\\u06FE\\u0F0B\\u2010" |
62 | u"\\u2019\\u2027\\u30A0\\u30FB]" ; |
63 | gInclusionSet = new UnicodeSet(UnicodeString(inclusionPat), status); |
64 | if (gInclusionSet == nullptr) { |
65 | status = U_MEMORY_ALLOCATION_ERROR; |
66 | return; |
67 | } |
68 | gInclusionSet->freeze(); |
69 | |
70 | // Note: data from IdentifierStatus.txt & IdentifierType.txt |
71 | // There is tooling to generate this constant in the unicodetools project: |
72 | // org.unicode.text.tools.RecommendedSetGenerator |
73 | // It will print the Java and C++ code to the console for easy copy-paste into this file. |
74 | static const char16_t *recommendedPat = |
75 | u"[0-9A-Z_a-z\\u00C0-\\u00D6\\u00D8-\\u00F6\\u00F8-\\u0131\\u0134-\\u013E" |
76 | u"\\u0141-\\u0148\\u014A-\\u017E\\u018F\\u01A0\\u01A1\\u01AF\\u01B0\\u01CD-" |
77 | u"\\u01DC\\u01DE-\\u01E3\\u01E6-\\u01F0\\u01F4\\u01F5\\u01F8-\\u021B\\u021E" |
78 | u"\\u021F\\u0226-\\u0233\\u0259\\u02BB\\u02BC\\u02EC\\u0300-\\u0304\\u0306-" |
79 | u"\\u030C\\u030F-\\u0311\\u0313\\u0314\\u031B\\u0323-\\u0328\\u032D\\u032E" |
80 | u"\\u0330\\u0331\\u0335\\u0338\\u0339\\u0342\\u0345\\u037B-\\u037D\\u0386" |
81 | u"\\u0388-\\u038A\\u038C\\u038E-\\u03A1\\u03A3-\\u03CE\\u03FC-\\u045F\\u048A-" |
82 | u"\\u04FF\\u0510-\\u0529\\u052E\\u052F\\u0531-\\u0556\\u0559\\u0561-\\u0586" |
83 | u"\\u05B4\\u05D0-\\u05EA\\u05EF-\\u05F2\\u0620-\\u063F\\u0641-\\u0655\\u0660-" |
84 | u"\\u0669\\u0670-\\u0672\\u0674\\u0679-\\u068D\\u068F-\\u06A0\\u06A2-\\u06D3" |
85 | u"\\u06D5\\u06E5\\u06E6\\u06EE-\\u06FC\\u06FF\\u0750-\\u07B1\\u0870-\\u0887" |
86 | u"\\u0889-\\u088E\\u08A0-\\u08AC\\u08B2\\u08B5-\\u08C9\\u0901-\\u094D\\u094F" |
87 | u"\\u0950\\u0956\\u0957\\u0960-\\u0963\\u0966-\\u096F\\u0971-\\u0977\\u0979-" |
88 | u"\\u097F\\u0981-\\u0983\\u0985-\\u098C\\u098F\\u0990\\u0993-\\u09A8\\u09AA-" |
89 | u"\\u09B0\\u09B2\\u09B6-\\u09B9\\u09BC-\\u09C4\\u09C7\\u09C8\\u09CB-\\u09CE" |
90 | u"\\u09D7\\u09E0-\\u09E3\\u09E6-\\u09F1\\u09FE\\u0A01-\\u0A03\\u0A05-\\u0A0A" |
91 | u"\\u0A0F\\u0A10\\u0A13-\\u0A28\\u0A2A-\\u0A30\\u0A32\\u0A35\\u0A38\\u0A39" |
92 | u"\\u0A3C\\u0A3E-\\u0A42\\u0A47\\u0A48\\u0A4B-\\u0A4D\\u0A5C\\u0A66-\\u0A74" |
93 | u"\\u0A81-\\u0A83\\u0A85-\\u0A8D\\u0A8F-\\u0A91\\u0A93-\\u0AA8\\u0AAA-\\u0AB0" |
94 | u"\\u0AB2\\u0AB3\\u0AB5-\\u0AB9\\u0ABC-\\u0AC5\\u0AC7-\\u0AC9\\u0ACB-\\u0ACD" |
95 | u"\\u0AD0\\u0AE0-\\u0AE3\\u0AE6-\\u0AEF\\u0AFA-\\u0AFF\\u0B01-\\u0B03\\u0B05-" |
96 | u"\\u0B0C\\u0B0F\\u0B10\\u0B13-\\u0B28\\u0B2A-\\u0B30\\u0B32\\u0B33\\u0B35-" |
97 | u"\\u0B39\\u0B3C-\\u0B43\\u0B47\\u0B48\\u0B4B-\\u0B4D\\u0B55-\\u0B57\\u0B5F-" |
98 | u"\\u0B61\\u0B66-\\u0B6F\\u0B71\\u0B82\\u0B83\\u0B85-\\u0B8A\\u0B8E-\\u0B90" |
99 | u"\\u0B92-\\u0B95\\u0B99\\u0B9A\\u0B9C\\u0B9E\\u0B9F\\u0BA3\\u0BA4\\u0BA8-" |
100 | u"\\u0BAA\\u0BAE-\\u0BB9\\u0BBE-\\u0BC2\\u0BC6-\\u0BC8\\u0BCA-\\u0BCD\\u0BD0" |
101 | u"\\u0BD7\\u0BE6-\\u0BEF\\u0C01-\\u0C0C\\u0C0E-\\u0C10\\u0C12-\\u0C28\\u0C2A-" |
102 | u"\\u0C33\\u0C35-\\u0C39\\u0C3C-\\u0C44\\u0C46-\\u0C48\\u0C4A-\\u0C4D\\u0C55" |
103 | u"\\u0C56\\u0C5D\\u0C60\\u0C61\\u0C66-\\u0C6F\\u0C80\\u0C82\\u0C83\\u0C85-" |
104 | u"\\u0C8C\\u0C8E-\\u0C90\\u0C92-\\u0CA8\\u0CAA-\\u0CB3\\u0CB5-\\u0CB9\\u0CBC-" |
105 | u"\\u0CC4\\u0CC6-\\u0CC8\\u0CCA-\\u0CCD\\u0CD5\\u0CD6\\u0CDD\\u0CE0-\\u0CE3" |
106 | u"\\u0CE6-\\u0CEF\\u0CF1-\\u0CF3\\u0D00\\u0D02\\u0D03\\u0D05-\\u0D0C\\u0D0E-" |
107 | u"\\u0D10\\u0D12-\\u0D3A\\u0D3D-\\u0D43\\u0D46-\\u0D48\\u0D4A-\\u0D4E\\u0D54-" |
108 | u"\\u0D57\\u0D60\\u0D61\\u0D66-\\u0D6F\\u0D7A-\\u0D7F\\u0D82\\u0D83\\u0D85-" |
109 | u"\\u0D8E\\u0D91-\\u0D96\\u0D9A-\\u0DA5\\u0DA7-\\u0DB1\\u0DB3-\\u0DBB\\u0DBD" |
110 | u"\\u0DC0-\\u0DC6\\u0DCA\\u0DCF-\\u0DD4\\u0DD6\\u0DD8-\\u0DDE\\u0DF2\\u0E01-" |
111 | u"\\u0E32\\u0E34-\\u0E3A\\u0E40-\\u0E4E\\u0E50-\\u0E59\\u0E81\\u0E82\\u0E84" |
112 | u"\\u0E86-\\u0E8A\\u0E8C-\\u0EA3\\u0EA5\\u0EA7-\\u0EB2\\u0EB4-\\u0EBD\\u0EC0-" |
113 | u"\\u0EC4\\u0EC6\\u0EC8-\\u0ECE\\u0ED0-\\u0ED9\\u0EDE\\u0EDF\\u0F00\\u0F20-" |
114 | u"\\u0F29\\u0F35\\u0F37\\u0F3E-\\u0F42\\u0F44-\\u0F47\\u0F49-\\u0F4C\\u0F4E-" |
115 | u"\\u0F51\\u0F53-\\u0F56\\u0F58-\\u0F5B\\u0F5D-\\u0F68\\u0F6A-\\u0F6C\\u0F71" |
116 | u"\\u0F72\\u0F74\\u0F7A-\\u0F80\\u0F82-\\u0F84\\u0F86-\\u0F92\\u0F94-\\u0F97" |
117 | u"\\u0F99-\\u0F9C\\u0F9E-\\u0FA1\\u0FA3-\\u0FA6\\u0FA8-\\u0FAB\\u0FAD-\\u0FB8" |
118 | u"\\u0FBA-\\u0FBC\\u0FC6\\u1000-\\u1049\\u1050-\\u109D\\u10C7\\u10CD\\u10D0-" |
119 | u"\\u10F0\\u10F7-\\u10FA\\u10FD-\\u10FF\\u1200-\\u1248\\u124A-\\u124D\\u1250-" |
120 | u"\\u1256\\u1258\\u125A-\\u125D\\u1260-\\u1288\\u128A-\\u128D\\u1290-\\u12B0" |
121 | u"\\u12B2-\\u12B5\\u12B8-\\u12BE\\u12C0\\u12C2-\\u12C5\\u12C8-\\u12D6\\u12D8-" |
122 | u"\\u1310\\u1312-\\u1315\\u1318-\\u135A\\u135D-\\u135F\\u1380-\\u138F\\u1780-" |
123 | u"\\u17A2\\u17A5-\\u17A7\\u17A9-\\u17B3\\u17B6-\\u17CD\\u17D0\\u17D2\\u17D7" |
124 | u"\\u17DC\\u17E0-\\u17E9\\u1C90-\\u1CBA\\u1CBD-\\u1CBF\\u1E00-\\u1E99\\u1E9E" |
125 | u"\\u1EA0-\\u1EF9\\u1F00-\\u1F15\\u1F18-\\u1F1D\\u1F20-\\u1F45\\u1F48-\\u1F4D" |
126 | u"\\u1F50-\\u1F57\\u1F59\\u1F5B\\u1F5D\\u1F5F-\\u1F70\\u1F72\\u1F74\\u1F76" |
127 | u"\\u1F78\\u1F7A\\u1F7C\\u1F80-\\u1FB4\\u1FB6-\\u1FBA\\u1FBC\\u1FC2-\\u1FC4" |
128 | u"\\u1FC6-\\u1FC8\\u1FCA\\u1FCC\\u1FD0-\\u1FD2\\u1FD6-\\u1FDA\\u1FE0-\\u1FE2" |
129 | u"\\u1FE4-\\u1FEA\\u1FEC\\u1FF2-\\u1FF4\\u1FF6-\\u1FF8\\u1FFA\\u1FFC\\u2D27" |
130 | u"\\u2D2D\\u2D80-\\u2D96\\u2DA0-\\u2DA6\\u2DA8-\\u2DAE\\u2DB0-\\u2DB6\\u2DB8-" |
131 | u"\\u2DBE\\u2DC0-\\u2DC6\\u2DC8-\\u2DCE\\u2DD0-\\u2DD6\\u2DD8-\\u2DDE\\u3005-" |
132 | u"\\u3007\\u3041-\\u3096\\u3099\\u309A\\u309D\\u309E\\u30A1-\\u30FA\\u30FC-" |
133 | u"\\u30FE\\u3105-\\u312D\\u312F\\u31A0-\\u31BF\\u3400-\\u4DBF\\u4E00-\\u9FFF" |
134 | u"\\uA67F\\uA717-\\uA71F\\uA788\\uA78D\\uA792\\uA793\\uA7AA\\uA7C0-\\uA7CA" |
135 | u"\\uA7D0\\uA7D1\\uA7D3\\uA7D5-\\uA7D9\\uA9E7-\\uA9FE\\uAA60-\\uAA76\\uAA7A-" |
136 | u"\\uAA7F\\uAB01-\\uAB06\\uAB09-\\uAB0E\\uAB11-\\uAB16\\uAB20-\\uAB26\\uAB28-" |
137 | u"\\uAB2E\\uAB66\\uAB67\\uAC00-\\uD7A3\\uFA0E\\uFA0F\\uFA11\\uFA13\\uFA14" |
138 | u"\\uFA1F\\uFA21\\uFA23\\uFA24\\uFA27-\\uFA29\\U00011301\\U00011303" |
139 | u"\\U0001133B\\U0001133C\\U00016FF0\\U00016FF1\\U0001B11F-\\U0001B122" |
140 | u"\\U0001B132\\U0001B150-\\U0001B152\\U0001B155\\U0001B164-\\U0001B167" |
141 | u"\\U0001DF00-\\U0001DF1E\\U0001DF25-\\U0001DF2A\\U0001E08F\\U0001E7E0-" |
142 | u"\\U0001E7E6\\U0001E7E8-\\U0001E7EB\\U0001E7ED\\U0001E7EE\\U0001E7F0-" |
143 | u"\\U0001E7FE\\U00020000-\\U0002A6DF\\U0002A700-\\U0002B739\\U0002B740-" |
144 | u"\\U0002B81D\\U0002B820-\\U0002CEA1\\U0002CEB0-\\U0002EBE0\\U00030000-" |
145 | u"\\U0003134A\\U00031350-\\U000323AF]" ; |
146 | |
147 | gRecommendedSet = new UnicodeSet(UnicodeString(recommendedPat), status); |
148 | if (gRecommendedSet == nullptr) { |
149 | status = U_MEMORY_ALLOCATION_ERROR; |
150 | delete gInclusionSet; |
151 | return; |
152 | } |
153 | gRecommendedSet->freeze(); |
154 | gNfdNormalizer = Normalizer2::getNFDInstance(status); |
155 | ucln_i18n_registerCleanup(UCLN_I18N_SPOOF, uspoof_cleanup); |
156 | } |
157 | |
158 | } // namespace |
159 | |
160 | U_CFUNC void uspoof_internalInitStatics(UErrorCode *status) { |
161 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
162 | } |
163 | |
164 | U_CAPI USpoofChecker * U_EXPORT2 |
165 | uspoof_open(UErrorCode *status) { |
166 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
167 | if (U_FAILURE(*status)) { |
168 | return nullptr; |
169 | } |
170 | SpoofImpl *si = new SpoofImpl(*status); |
171 | if (si == nullptr) { |
172 | *status = U_MEMORY_ALLOCATION_ERROR; |
173 | return nullptr; |
174 | } |
175 | if (U_FAILURE(*status)) { |
176 | delete si; |
177 | return nullptr; |
178 | } |
179 | return si->asUSpoofChecker(); |
180 | } |
181 | |
182 | |
183 | U_CAPI USpoofChecker * U_EXPORT2 |
184 | uspoof_openFromSerialized(const void *data, int32_t length, int32_t *pActualLength, |
185 | UErrorCode *status) { |
186 | if (U_FAILURE(*status)) { |
187 | return nullptr; |
188 | } |
189 | |
190 | if (data == nullptr) { |
191 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
192 | return nullptr; |
193 | } |
194 | |
195 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
196 | if (U_FAILURE(*status)) |
197 | { |
198 | return nullptr; |
199 | } |
200 | |
201 | SpoofData *sd = new SpoofData(data, length, *status); |
202 | if (sd == nullptr) { |
203 | *status = U_MEMORY_ALLOCATION_ERROR; |
204 | return nullptr; |
205 | } |
206 | |
207 | if (U_FAILURE(*status)) { |
208 | delete sd; |
209 | return nullptr; |
210 | } |
211 | |
212 | SpoofImpl *si = new SpoofImpl(sd, *status); |
213 | if (si == nullptr) { |
214 | *status = U_MEMORY_ALLOCATION_ERROR; |
215 | delete sd; // explicit delete as the destructor for si won't be called. |
216 | return nullptr; |
217 | } |
218 | |
219 | if (U_FAILURE(*status)) { |
220 | delete si; // no delete for sd, as the si destructor will delete it. |
221 | return nullptr; |
222 | } |
223 | |
224 | if (pActualLength != nullptr) { |
225 | *pActualLength = sd->size(); |
226 | } |
227 | return si->asUSpoofChecker(); |
228 | } |
229 | |
230 | |
231 | U_CAPI USpoofChecker * U_EXPORT2 |
232 | uspoof_clone(const USpoofChecker *sc, UErrorCode *status) { |
233 | const SpoofImpl *src = SpoofImpl::validateThis(sc, *status); |
234 | if (src == nullptr) { |
235 | return nullptr; |
236 | } |
237 | SpoofImpl *result = new SpoofImpl(*src, *status); // copy constructor |
238 | if (result == nullptr) { |
239 | *status = U_MEMORY_ALLOCATION_ERROR; |
240 | return nullptr; |
241 | } |
242 | if (U_FAILURE(*status)) { |
243 | delete result; |
244 | result = nullptr; |
245 | } |
246 | return result->asUSpoofChecker(); |
247 | } |
248 | |
249 | |
250 | U_CAPI void U_EXPORT2 |
251 | uspoof_close(USpoofChecker *sc) { |
252 | UErrorCode status = U_ZERO_ERROR; |
253 | SpoofImpl *This = SpoofImpl::validateThis(sc, status); |
254 | delete This; |
255 | } |
256 | |
257 | |
258 | U_CAPI void U_EXPORT2 |
259 | uspoof_setChecks(USpoofChecker *sc, int32_t checks, UErrorCode *status) { |
260 | SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
261 | if (This == nullptr) { |
262 | return; |
263 | } |
264 | |
265 | // Verify that the requested checks are all ones (bits) that |
266 | // are acceptable, known values. |
267 | if (checks & ~(USPOOF_ALL_CHECKS | USPOOF_AUX_INFO)) { |
268 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
269 | return; |
270 | } |
271 | |
272 | This->fChecks = checks; |
273 | } |
274 | |
275 | |
276 | U_CAPI int32_t U_EXPORT2 |
277 | uspoof_getChecks(const USpoofChecker *sc, UErrorCode *status) { |
278 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
279 | if (This == nullptr) { |
280 | return 0; |
281 | } |
282 | return This->fChecks; |
283 | } |
284 | |
285 | U_CAPI void U_EXPORT2 |
286 | uspoof_setRestrictionLevel(USpoofChecker *sc, URestrictionLevel restrictionLevel) { |
287 | UErrorCode status = U_ZERO_ERROR; |
288 | SpoofImpl *This = SpoofImpl::validateThis(sc, status); |
289 | if (This != nullptr) { |
290 | This->fRestrictionLevel = restrictionLevel; |
291 | This->fChecks |= USPOOF_RESTRICTION_LEVEL; |
292 | } |
293 | } |
294 | |
295 | U_CAPI URestrictionLevel U_EXPORT2 |
296 | uspoof_getRestrictionLevel(const USpoofChecker *sc) { |
297 | UErrorCode status = U_ZERO_ERROR; |
298 | const SpoofImpl *This = SpoofImpl::validateThis(sc, status); |
299 | if (This == nullptr) { |
300 | return USPOOF_UNRESTRICTIVE; |
301 | } |
302 | return This->fRestrictionLevel; |
303 | } |
304 | |
305 | U_CAPI void U_EXPORT2 |
306 | uspoof_setAllowedLocales(USpoofChecker *sc, const char *localesList, UErrorCode *status) { |
307 | SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
308 | if (This == nullptr) { |
309 | return; |
310 | } |
311 | This->setAllowedLocales(localesList, *status); |
312 | } |
313 | |
314 | U_CAPI const char * U_EXPORT2 |
315 | uspoof_getAllowedLocales(USpoofChecker *sc, UErrorCode *status) { |
316 | SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
317 | if (This == nullptr) { |
318 | return nullptr; |
319 | } |
320 | return This->getAllowedLocales(*status); |
321 | } |
322 | |
323 | |
324 | U_CAPI const USet * U_EXPORT2 |
325 | uspoof_getAllowedChars(const USpoofChecker *sc, UErrorCode *status) { |
326 | const UnicodeSet *result = uspoof_getAllowedUnicodeSet(sc, status); |
327 | return result->toUSet(); |
328 | } |
329 | |
330 | U_CAPI const UnicodeSet * U_EXPORT2 |
331 | uspoof_getAllowedUnicodeSet(const USpoofChecker *sc, UErrorCode *status) { |
332 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
333 | if (This == nullptr) { |
334 | return nullptr; |
335 | } |
336 | return This->fAllowedCharsSet; |
337 | } |
338 | |
339 | |
340 | U_CAPI void U_EXPORT2 |
341 | uspoof_setAllowedChars(USpoofChecker *sc, const USet *chars, UErrorCode *status) { |
342 | const UnicodeSet *set = UnicodeSet::fromUSet(chars); |
343 | uspoof_setAllowedUnicodeSet(sc, set, status); |
344 | } |
345 | |
346 | |
347 | U_CAPI void U_EXPORT2 |
348 | uspoof_setAllowedUnicodeSet(USpoofChecker *sc, const UnicodeSet *chars, UErrorCode *status) { |
349 | SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
350 | if (This == nullptr) { |
351 | return; |
352 | } |
353 | if (chars->isBogus()) { |
354 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
355 | return; |
356 | } |
357 | UnicodeSet *clonedSet = chars->clone(); |
358 | if (clonedSet == nullptr || clonedSet->isBogus()) { |
359 | *status = U_MEMORY_ALLOCATION_ERROR; |
360 | return; |
361 | } |
362 | clonedSet->freeze(); |
363 | delete This->fAllowedCharsSet; |
364 | This->fAllowedCharsSet = clonedSet; |
365 | This->fChecks |= USPOOF_CHAR_LIMIT; |
366 | } |
367 | |
368 | |
369 | U_CAPI int32_t U_EXPORT2 |
370 | uspoof_check(const USpoofChecker *sc, |
371 | const char16_t *id, int32_t length, |
372 | int32_t *position, |
373 | UErrorCode *status) { |
374 | |
375 | // Backwards compatibility: |
376 | if (position != nullptr) { |
377 | *position = 0; |
378 | } |
379 | |
380 | // Delegate to uspoof_check2 |
381 | return uspoof_check2(sc, id, length, nullptr, status); |
382 | } |
383 | |
384 | |
385 | U_CAPI int32_t U_EXPORT2 |
386 | uspoof_check2(const USpoofChecker *sc, |
387 | const char16_t* id, int32_t length, |
388 | USpoofCheckResult* checkResult, |
389 | UErrorCode *status) { |
390 | |
391 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
392 | if (This == nullptr) { |
393 | return 0; |
394 | } |
395 | if (length < -1) { |
396 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
397 | return 0; |
398 | } |
399 | UnicodeString idStr((length == -1), id, length); // Aliasing constructor. |
400 | int32_t result = uspoof_check2UnicodeString(sc, idStr, checkResult, status); |
401 | return result; |
402 | } |
403 | |
404 | |
405 | U_CAPI int32_t U_EXPORT2 |
406 | uspoof_checkUTF8(const USpoofChecker *sc, |
407 | const char *id, int32_t length, |
408 | int32_t *position, |
409 | UErrorCode *status) { |
410 | |
411 | // Backwards compatibility: |
412 | if (position != nullptr) { |
413 | *position = 0; |
414 | } |
415 | |
416 | // Delegate to uspoof_check2 |
417 | return uspoof_check2UTF8(sc, id, length, nullptr, status); |
418 | } |
419 | |
420 | |
421 | U_CAPI int32_t U_EXPORT2 |
422 | uspoof_check2UTF8(const USpoofChecker *sc, |
423 | const char *id, int32_t length, |
424 | USpoofCheckResult* checkResult, |
425 | UErrorCode *status) { |
426 | |
427 | if (U_FAILURE(*status)) { |
428 | return 0; |
429 | } |
430 | UnicodeString idStr = UnicodeString::fromUTF8(StringPiece(id, length>=0 ? length : static_cast<int32_t>(uprv_strlen(id)))); |
431 | int32_t result = uspoof_check2UnicodeString(sc, idStr, checkResult, status); |
432 | return result; |
433 | } |
434 | |
435 | |
436 | U_CAPI int32_t U_EXPORT2 |
437 | uspoof_areConfusable(const USpoofChecker *sc, |
438 | const char16_t *id1, int32_t length1, |
439 | const char16_t *id2, int32_t length2, |
440 | UErrorCode *status) { |
441 | SpoofImpl::validateThis(sc, *status); |
442 | if (U_FAILURE(*status)) { |
443 | return 0; |
444 | } |
445 | if (length1 < -1 || length2 < -1) { |
446 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
447 | return 0; |
448 | } |
449 | |
450 | UnicodeString id1Str((length1==-1), id1, length1); // Aliasing constructor |
451 | UnicodeString id2Str((length2==-1), id2, length2); // Aliasing constructor |
452 | return uspoof_areConfusableUnicodeString(sc, id1Str, id2Str, status); |
453 | } |
454 | |
455 | |
456 | U_CAPI int32_t U_EXPORT2 |
457 | uspoof_areConfusableUTF8(const USpoofChecker *sc, |
458 | const char *id1, int32_t length1, |
459 | const char *id2, int32_t length2, |
460 | UErrorCode *status) { |
461 | SpoofImpl::validateThis(sc, *status); |
462 | if (U_FAILURE(*status)) { |
463 | return 0; |
464 | } |
465 | if (length1 < -1 || length2 < -1) { |
466 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
467 | return 0; |
468 | } |
469 | UnicodeString id1Str = UnicodeString::fromUTF8(StringPiece(id1, length1>=0? length1 : static_cast<int32_t>(uprv_strlen(id1)))); |
470 | UnicodeString id2Str = UnicodeString::fromUTF8(StringPiece(id2, length2>=0? length2 : static_cast<int32_t>(uprv_strlen(id2)))); |
471 | int32_t results = uspoof_areConfusableUnicodeString(sc, id1Str, id2Str, status); |
472 | return results; |
473 | } |
474 | |
475 | |
476 | U_CAPI int32_t U_EXPORT2 |
477 | uspoof_areConfusableUnicodeString(const USpoofChecker *sc, |
478 | const icu::UnicodeString &id1, |
479 | const icu::UnicodeString &id2, |
480 | UErrorCode *status) { |
481 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
482 | if (U_FAILURE(*status)) { |
483 | return 0; |
484 | } |
485 | // |
486 | // See section 4 of UAX 39 for the algorithm for checking whether two strings are confusable, |
487 | // and for definitions of the types (single, whole, mixed-script) of confusables. |
488 | |
489 | // We only care about a few of the check flags. Ignore the others. |
490 | // If no tests relevant to this function have been specified, return an error. |
491 | // TODO: is this really the right thing to do? It's probably an error on the caller's part, |
492 | // but logically we would just return 0 (no error). |
493 | if ((This->fChecks & USPOOF_CONFUSABLE) == 0) { |
494 | *status = U_INVALID_STATE_ERROR; |
495 | return 0; |
496 | } |
497 | |
498 | // Compute the skeletons and check for confusability. |
499 | UnicodeString id1Skeleton; |
500 | uspoof_getSkeletonUnicodeString(sc, 0 /* deprecated */, id1, id1Skeleton, status); |
501 | UnicodeString id2Skeleton; |
502 | uspoof_getSkeletonUnicodeString(sc, 0 /* deprecated */, id2, id2Skeleton, status); |
503 | if (U_FAILURE(*status)) { return 0; } |
504 | if (id1Skeleton != id2Skeleton) { |
505 | return 0; |
506 | } |
507 | |
508 | // If we get here, the strings are confusable. Now we just need to set the flags for the appropriate classes |
509 | // of confusables according to UTS 39 section 4. |
510 | // Start by computing the resolved script sets of id1 and id2. |
511 | ScriptSet ; |
512 | This->getResolvedScriptSet(id1, id1RSS, *status); |
513 | ScriptSet ; |
514 | This->getResolvedScriptSet(id2, id2RSS, *status); |
515 | |
516 | // Turn on all applicable flags |
517 | int32_t result = 0; |
518 | if (id1RSS.intersects(id2RSS)) { |
519 | result |= USPOOF_SINGLE_SCRIPT_CONFUSABLE; |
520 | } else { |
521 | result |= USPOOF_MIXED_SCRIPT_CONFUSABLE; |
522 | if (!id1RSS.isEmpty() && !id2RSS.isEmpty()) { |
523 | result |= USPOOF_WHOLE_SCRIPT_CONFUSABLE; |
524 | } |
525 | } |
526 | |
527 | // Turn off flags that the user doesn't want |
528 | if ((This->fChecks & USPOOF_SINGLE_SCRIPT_CONFUSABLE) == 0) { |
529 | result &= ~USPOOF_SINGLE_SCRIPT_CONFUSABLE; |
530 | } |
531 | if ((This->fChecks & USPOOF_MIXED_SCRIPT_CONFUSABLE) == 0) { |
532 | result &= ~USPOOF_MIXED_SCRIPT_CONFUSABLE; |
533 | } |
534 | if ((This->fChecks & USPOOF_WHOLE_SCRIPT_CONFUSABLE) == 0) { |
535 | result &= ~USPOOF_WHOLE_SCRIPT_CONFUSABLE; |
536 | } |
537 | |
538 | return result; |
539 | } |
540 | |
541 | |
542 | U_CAPI int32_t U_EXPORT2 |
543 | uspoof_checkUnicodeString(const USpoofChecker *sc, |
544 | const icu::UnicodeString &id, |
545 | int32_t *position, |
546 | UErrorCode *status) { |
547 | |
548 | // Backwards compatibility: |
549 | if (position != nullptr) { |
550 | *position = 0; |
551 | } |
552 | |
553 | // Delegate to uspoof_check2 |
554 | return uspoof_check2UnicodeString(sc, id, nullptr, status); |
555 | } |
556 | |
557 | namespace { |
558 | |
559 | int32_t checkImpl(const SpoofImpl* This, const UnicodeString& id, CheckResult* checkResult, UErrorCode* status) { |
560 | U_ASSERT(This != nullptr); |
561 | U_ASSERT(checkResult != nullptr); |
562 | checkResult->clear(); |
563 | int32_t result = 0; |
564 | |
565 | if (0 != (This->fChecks & USPOOF_RESTRICTION_LEVEL)) { |
566 | URestrictionLevel idRestrictionLevel = This->getRestrictionLevel(id, *status); |
567 | if (idRestrictionLevel > This->fRestrictionLevel) { |
568 | result |= USPOOF_RESTRICTION_LEVEL; |
569 | } |
570 | checkResult->fRestrictionLevel = idRestrictionLevel; |
571 | } |
572 | |
573 | if (0 != (This->fChecks & USPOOF_MIXED_NUMBERS)) { |
574 | UnicodeSet numerics; |
575 | This->getNumerics(id, numerics, *status); |
576 | if (numerics.size() > 1) { |
577 | result |= USPOOF_MIXED_NUMBERS; |
578 | } |
579 | checkResult->fNumerics = numerics; // UnicodeSet::operator= |
580 | } |
581 | |
582 | if (0 != (This->fChecks & USPOOF_HIDDEN_OVERLAY)) { |
583 | int32_t index = This->findHiddenOverlay(id, *status); |
584 | if (index != -1) { |
585 | result |= USPOOF_HIDDEN_OVERLAY; |
586 | } |
587 | } |
588 | |
589 | |
590 | if (0 != (This->fChecks & USPOOF_CHAR_LIMIT)) { |
591 | int32_t i; |
592 | UChar32 c; |
593 | int32_t length = id.length(); |
594 | for (i=0; i<length ;) { |
595 | c = id.char32At(i); |
596 | i += U16_LENGTH(c); |
597 | if (!This->fAllowedCharsSet->contains(c)) { |
598 | result |= USPOOF_CHAR_LIMIT; |
599 | break; |
600 | } |
601 | } |
602 | } |
603 | |
604 | if (0 != (This->fChecks & USPOOF_INVISIBLE)) { |
605 | // This check needs to be done on NFD input |
606 | UnicodeString nfdText; |
607 | gNfdNormalizer->normalize(id, nfdText, *status); |
608 | int32_t nfdLength = nfdText.length(); |
609 | |
610 | // scan for more than one occurrence of the same non-spacing mark |
611 | // in a sequence of non-spacing marks. |
612 | int32_t i; |
613 | UChar32 c; |
614 | UChar32 firstNonspacingMark = 0; |
615 | UBool haveMultipleMarks = false; |
616 | UnicodeSet marksSeenSoFar; // Set of combining marks in a single combining sequence. |
617 | |
618 | for (i=0; i<nfdLength ;) { |
619 | c = nfdText.char32At(i); |
620 | i += U16_LENGTH(c); |
621 | if (u_charType(c) != U_NON_SPACING_MARK) { |
622 | firstNonspacingMark = 0; |
623 | if (haveMultipleMarks) { |
624 | marksSeenSoFar.clear(); |
625 | haveMultipleMarks = false; |
626 | } |
627 | continue; |
628 | } |
629 | if (firstNonspacingMark == 0) { |
630 | firstNonspacingMark = c; |
631 | continue; |
632 | } |
633 | if (!haveMultipleMarks) { |
634 | marksSeenSoFar.add(firstNonspacingMark); |
635 | haveMultipleMarks = true; |
636 | } |
637 | if (marksSeenSoFar.contains(c)) { |
638 | // report the error, and stop scanning. |
639 | // No need to find more than the first failure. |
640 | result |= USPOOF_INVISIBLE; |
641 | break; |
642 | } |
643 | marksSeenSoFar.add(c); |
644 | } |
645 | } |
646 | |
647 | checkResult->fChecks = result; |
648 | return checkResult->toCombinedBitmask(This->fChecks); |
649 | } |
650 | |
651 | } // namespace |
652 | |
653 | U_CAPI int32_t U_EXPORT2 |
654 | uspoof_check2UnicodeString(const USpoofChecker *sc, |
655 | const icu::UnicodeString &id, |
656 | USpoofCheckResult* checkResult, |
657 | UErrorCode *status) { |
658 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
659 | if (This == nullptr) { |
660 | return false; |
661 | } |
662 | |
663 | if (checkResult != nullptr) { |
664 | CheckResult* ThisCheckResult = CheckResult::validateThis(checkResult, *status); |
665 | if (ThisCheckResult == nullptr) { |
666 | return false; |
667 | } |
668 | return checkImpl(This, id, ThisCheckResult, status); |
669 | } else { |
670 | // Stack-allocate the checkResult since this method doesn't return it |
671 | CheckResult stackCheckResult; |
672 | return checkImpl(This, id, &stackCheckResult, status); |
673 | } |
674 | } |
675 | |
676 | |
677 | U_CAPI int32_t U_EXPORT2 |
678 | uspoof_getSkeleton(const USpoofChecker *sc, |
679 | uint32_t type, |
680 | const char16_t *id, int32_t length, |
681 | char16_t *dest, int32_t destCapacity, |
682 | UErrorCode *status) { |
683 | |
684 | SpoofImpl::validateThis(sc, *status); |
685 | if (U_FAILURE(*status)) { |
686 | return 0; |
687 | } |
688 | if (length<-1 || destCapacity<0 || (destCapacity==0 && dest!=nullptr)) { |
689 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
690 | return 0; |
691 | } |
692 | |
693 | UnicodeString idStr((length==-1), id, length); // Aliasing constructor |
694 | UnicodeString destStr; |
695 | uspoof_getSkeletonUnicodeString(sc, type, idStr, destStr, status); |
696 | destStr.extract(dest, destCapacity, *status); |
697 | return destStr.length(); |
698 | } |
699 | |
700 | |
701 | |
702 | U_I18N_API UnicodeString & U_EXPORT2 |
703 | uspoof_getSkeletonUnicodeString(const USpoofChecker *sc, |
704 | uint32_t /*type*/, |
705 | const UnicodeString &id, |
706 | UnicodeString &dest, |
707 | UErrorCode *status) { |
708 | const SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
709 | if (U_FAILURE(*status)) { |
710 | return dest; |
711 | } |
712 | |
713 | UnicodeString nfdId; |
714 | gNfdNormalizer->normalize(id, nfdId, *status); |
715 | |
716 | // Apply the skeleton mapping to the NFD normalized input string |
717 | // Accumulate the skeleton, possibly unnormalized, in a UnicodeString. |
718 | int32_t inputIndex = 0; |
719 | UnicodeString skelStr; |
720 | int32_t normalizedLen = nfdId.length(); |
721 | for (inputIndex=0; inputIndex < normalizedLen; ) { |
722 | UChar32 c = nfdId.char32At(inputIndex); |
723 | inputIndex += U16_LENGTH(c); |
724 | This->fSpoofData->confusableLookup(c, skelStr); |
725 | } |
726 | |
727 | gNfdNormalizer->normalize(skelStr, dest, *status); |
728 | return dest; |
729 | } |
730 | |
731 | |
732 | U_CAPI int32_t U_EXPORT2 |
733 | uspoof_getSkeletonUTF8(const USpoofChecker *sc, |
734 | uint32_t type, |
735 | const char *id, int32_t length, |
736 | char *dest, int32_t destCapacity, |
737 | UErrorCode *status) { |
738 | SpoofImpl::validateThis(sc, *status); |
739 | if (U_FAILURE(*status)) { |
740 | return 0; |
741 | } |
742 | if (length<-1 || destCapacity<0 || (destCapacity==0 && dest!=nullptr)) { |
743 | *status = U_ILLEGAL_ARGUMENT_ERROR; |
744 | return 0; |
745 | } |
746 | |
747 | UnicodeString srcStr = UnicodeString::fromUTF8(StringPiece(id, length>=0 ? length : static_cast<int32_t>(uprv_strlen(id)))); |
748 | UnicodeString destStr; |
749 | uspoof_getSkeletonUnicodeString(sc, type, srcStr, destStr, status); |
750 | if (U_FAILURE(*status)) { |
751 | return 0; |
752 | } |
753 | |
754 | int32_t lengthInUTF8 = 0; |
755 | u_strToUTF8(dest, destCapacity, &lengthInUTF8, |
756 | destStr.getBuffer(), destStr.length(), status); |
757 | return lengthInUTF8; |
758 | } |
759 | |
760 | |
761 | U_CAPI int32_t U_EXPORT2 |
762 | uspoof_serialize(USpoofChecker *sc,void *buf, int32_t capacity, UErrorCode *status) { |
763 | SpoofImpl *This = SpoofImpl::validateThis(sc, *status); |
764 | if (This == nullptr) { |
765 | U_ASSERT(U_FAILURE(*status)); |
766 | return 0; |
767 | } |
768 | |
769 | return This->fSpoofData->serialize(buf, capacity, *status); |
770 | } |
771 | |
772 | U_CAPI const USet * U_EXPORT2 |
773 | uspoof_getInclusionSet(UErrorCode *status) { |
774 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
775 | return gInclusionSet->toUSet(); |
776 | } |
777 | |
778 | U_CAPI const USet * U_EXPORT2 |
779 | uspoof_getRecommendedSet(UErrorCode *status) { |
780 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
781 | return gRecommendedSet->toUSet(); |
782 | } |
783 | |
784 | U_I18N_API const UnicodeSet * U_EXPORT2 |
785 | uspoof_getInclusionUnicodeSet(UErrorCode *status) { |
786 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
787 | return gInclusionSet; |
788 | } |
789 | |
790 | U_I18N_API const UnicodeSet * U_EXPORT2 |
791 | uspoof_getRecommendedUnicodeSet(UErrorCode *status) { |
792 | umtx_initOnce(gSpoofInitStaticsOnce, &initializeStatics, *status); |
793 | return gRecommendedSet; |
794 | } |
795 | |
796 | //------------------ |
797 | // CheckResult APIs |
798 | //------------------ |
799 | |
800 | U_CAPI USpoofCheckResult* U_EXPORT2 |
801 | uspoof_openCheckResult(UErrorCode *status) { |
802 | CheckResult* checkResult = new CheckResult(); |
803 | if (checkResult == nullptr) { |
804 | *status = U_MEMORY_ALLOCATION_ERROR; |
805 | return nullptr; |
806 | } |
807 | return checkResult->asUSpoofCheckResult(); |
808 | } |
809 | |
810 | U_CAPI void U_EXPORT2 |
811 | uspoof_closeCheckResult(USpoofCheckResult* checkResult) { |
812 | UErrorCode status = U_ZERO_ERROR; |
813 | CheckResult* This = CheckResult::validateThis(checkResult, status); |
814 | delete This; |
815 | } |
816 | |
817 | U_CAPI int32_t U_EXPORT2 |
818 | uspoof_getCheckResultChecks(const USpoofCheckResult *checkResult, UErrorCode *status) { |
819 | const CheckResult* This = CheckResult::validateThis(checkResult, *status); |
820 | if (U_FAILURE(*status)) { return 0; } |
821 | return This->fChecks; |
822 | } |
823 | |
824 | U_CAPI URestrictionLevel U_EXPORT2 |
825 | uspoof_getCheckResultRestrictionLevel(const USpoofCheckResult *checkResult, UErrorCode *status) { |
826 | const CheckResult* This = CheckResult::validateThis(checkResult, *status); |
827 | if (U_FAILURE(*status)) { return USPOOF_UNRESTRICTIVE; } |
828 | return This->fRestrictionLevel; |
829 | } |
830 | |
831 | U_CAPI const USet* U_EXPORT2 |
832 | uspoof_getCheckResultNumerics(const USpoofCheckResult *checkResult, UErrorCode *status) { |
833 | const CheckResult* This = CheckResult::validateThis(checkResult, *status); |
834 | if (U_FAILURE(*status)) { return nullptr; } |
835 | return This->fNumerics.toUSet(); |
836 | } |
837 | |
838 | |
839 | |
840 | #endif // !UCONFIG_NO_NORMALIZATION |
841 | |