• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 ***************************************************************************
5 * Copyright (C) 2008-2016, International Business Machines Corporation
6 * and others. All Rights Reserved.
7 ***************************************************************************
8 *   file name:  uspoof.h
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 
19 #ifndef USPOOF_H
20 #define USPOOF_H
21 
22 #include "unicode/utypes.h"
23 #include "unicode/uset.h"
24 #include "unicode/parseerr.h"
25 #include "unicode/localpointer.h"
26 
27 #if !UCONFIG_NO_NORMALIZATION
28 
29 
30 #if U_SHOW_CPLUSPLUS_API
31 #include "unicode/unistr.h"
32 #include "unicode/uniset.h"
33 #endif
34 
35 
36 /**
37  * \file
38  * \brief Unicode Security and Spoofing Detection, C API.
39  *
40  * <p>
41  * This class, based on <a href="http://unicode.org/reports/tr36">Unicode Technical Report #36</a> and
42  * <a href="http://unicode.org/reports/tr39">Unicode Technical Standard #39</a>, has two main functions:
43  *
44  * <ol>
45  * <li>Checking whether two strings are visually <em>confusable</em> with each other, such as "Harvest" and
46  * &quot;&Eta;arvest&quot;, where the second string starts with the Greek capital letter Eta.</li>
47  * <li>Checking whether an individual string is likely to be an attempt at confusing the reader (<em>spoof
48  * detection</em>), such as "paypal" with some Latin characters substituted with Cyrillic look-alikes.</li>
49  * </ol>
50  *
51  * <p>
52  * Although originally designed as a method for flagging suspicious identifier strings such as URLs,
53  * <code>USpoofChecker</code> has a number of other practical use cases, such as preventing attempts to evade bad-word
54  * content filters.
55  *
56  * <p>
57  * The functions of this class are exposed as C API, with a handful of syntactical conveniences for C++.
58  *
59  * <h2>Confusables</h2>
60  *
61  * <p>
62  * The following example shows how to use <code>USpoofChecker</code> to check for confusability between two strings:
63  *
64  * \code{.c}
65  * UErrorCode status = U_ZERO_ERROR;
66  * UChar* str1 = (UChar*) u"Harvest";
67  * UChar* str2 = (UChar*) u"\u0397arvest";  // with U+0397 GREEK CAPITAL LETTER ETA
68  *
69  * USpoofChecker* sc = uspoof_open(&status);
70  * uspoof_setChecks(sc, USPOOF_CONFUSABLE, &status);
71  *
72  * int32_t bitmask = uspoof_areConfusable(sc, str1, -1, str2, -1, &status);
73  * UBool result = bitmask != 0;
74  * // areConfusable: 1 (status: U_ZERO_ERROR)
75  * printf("areConfusable: %d (status: %s)\n", result, u_errorName(status));
76  * uspoof_close(sc);
77  * \endcode
78  *
79  * <p>
80  * The call to {@link uspoof_open} creates a <code>USpoofChecker</code> object; the call to {@link uspoof_setChecks}
81  * enables confusable checking and disables all other checks; the call to {@link uspoof_areConfusable} performs the
82  * confusability test; and the following line extracts the result out of the return value. For best performance,
83  * the instance should be created once (e.g., upon application startup), and the efficient
84  * {@link uspoof_areConfusable} method can be used at runtime.
85  *
86  * <p>
87  * The type {@link LocalUSpoofCheckerPointer} is exposed for C++ programmers.  It will automatically call
88  * {@link uspoof_close} when the object goes out of scope:
89  *
90  * \code{.cpp}
91  * UErrorCode status = U_ZERO_ERROR;
92  * LocalUSpoofCheckerPointer sc(uspoof_open(&status));
93  * uspoof_setChecks(sc.getAlias(), USPOOF_CONFUSABLE, &status);
94  * // ...
95  * \endcode
96  *
97  * UTS 39 defines two strings to be <em>confusable</em> if they map to the same <em>skeleton string</em>. A skeleton can
98  * be thought of as a "hash code". {@link uspoof_getSkeleton} computes the skeleton for a particular string, so
99  * the following snippet is equivalent to the example above:
100  *
101  * \code{.c}
102  * UErrorCode status = U_ZERO_ERROR;
103  * UChar* str1 = (UChar*) u"Harvest";
104  * UChar* str2 = (UChar*) u"\u0397arvest";  // with U+0397 GREEK CAPITAL LETTER ETA
105  *
106  * USpoofChecker* sc = uspoof_open(&status);
107  * uspoof_setChecks(sc, USPOOF_CONFUSABLE, &status);
108  *
109  * // Get skeleton 1
110  * int32_t skel1Len = uspoof_getSkeleton(sc, 0, str1, -1, NULL, 0, &status);
111  * UChar* skel1 = (UChar*) malloc(++skel1Len * sizeof(UChar));
112  * status = U_ZERO_ERROR;
113  * uspoof_getSkeleton(sc, 0, str1, -1, skel1, skel1Len, &status);
114  *
115  * // Get skeleton 2
116  * int32_t skel2Len = uspoof_getSkeleton(sc, 0, str2, -1, NULL, 0, &status);
117  * UChar* skel2 = (UChar*) malloc(++skel2Len * sizeof(UChar));
118  * status = U_ZERO_ERROR;
119  * uspoof_getSkeleton(sc, 0, str2, -1, skel2, skel2Len, &status);
120  *
121  * // Are the skeletons the same?
122  * UBool result = u_strcmp(skel1, skel2) == 0;
123  * // areConfusable: 1 (status: U_ZERO_ERROR)
124  * printf("areConfusable: %d (status: %s)\n", result, u_errorName(status));
125  * uspoof_close(sc);
126  * free(skel1);
127  * free(skel2);
128  * \endcode
129  *
130  * If you need to check if a string is confusable with any string in a dictionary of many strings, rather than calling
131  * {@link uspoof_areConfusable} many times in a loop, {@link uspoof_getSkeleton} can be used instead, as shown below:
132  *
133  * \code{.c}
134  * UErrorCode status = U_ZERO_ERROR;
135  * #define DICTIONARY_LENGTH 2
136  * UChar* dictionary[DICTIONARY_LENGTH] = { (UChar*) u"lorem", (UChar*) u"ipsum" };
137  * UChar* skeletons[DICTIONARY_LENGTH];
138  * UChar* str = (UChar*) u"1orern";
139  *
140  * // Setup:
141  * USpoofChecker* sc = uspoof_open(&status);
142  * uspoof_setChecks(sc, USPOOF_CONFUSABLE, &status);
143  * for (size_t i=0; i<DICTIONARY_LENGTH; i++) {
144  *     UChar* word = dictionary[i];
145  *     int32_t len = uspoof_getSkeleton(sc, 0, word, -1, NULL, 0, &status);
146  *     skeletons[i] = (UChar*) malloc(++len * sizeof(UChar));
147  *     status = U_ZERO_ERROR;
148  *     uspoof_getSkeleton(sc, 0, word, -1, skeletons[i], len, &status);
149  * }
150  *
151  * // Live Check:
152  * {
153  *     int32_t len = uspoof_getSkeleton(sc, 0, str, -1, NULL, 0, &status);
154  *     UChar* skel = (UChar*) malloc(++len * sizeof(UChar));
155  *     status = U_ZERO_ERROR;
156  *     uspoof_getSkeleton(sc, 0, str, -1, skel, len, &status);
157  *     UBool result = FALSE;
158  *     for (size_t i=0; i<DICTIONARY_LENGTH; i++) {
159  *         result = u_strcmp(skel, skeletons[i]) == 0;
160  *         if (result == TRUE) { break; }
161  *     }
162  *     // Has confusable in dictionary: 1 (status: U_ZERO_ERROR)
163  *     printf("Has confusable in dictionary: %d (status: %s)\n", result, u_errorName(status));
164  *     free(skel);
165  * }
166  *
167  * for (size_t i=0; i<DICTIONARY_LENGTH; i++) {
168  *     free(skeletons[i]);
169  * }
170  * uspoof_close(sc);
171  * \endcode
172  *
173  * <b>Note:</b> Since the Unicode confusables mapping table is frequently updated, confusable skeletons are <em>not</em>
174  * guaranteed to be the same between ICU releases. We therefore recommend that you always compute confusable skeletons
175  * at runtime and do not rely on creating a permanent, or difficult to update, database of skeletons.
176  *
177  * <h2>Spoof Detection</h2>
178  *
179  * The following snippet shows a minimal example of using <code>USpoofChecker</code> to perform spoof detection on a
180  * string:
181  *
182  * \code{.c}
183  * UErrorCode status = U_ZERO_ERROR;
184  * UChar* str = (UChar*) u"p\u0430ypal";  // with U+0430 CYRILLIC SMALL LETTER A
185  *
186  * // Get the default set of allowable characters:
187  * USet* allowed = uset_openEmpty();
188  * uset_addAll(allowed, uspoof_getRecommendedSet(&status));
189  * uset_addAll(allowed, uspoof_getInclusionSet(&status));
190  *
191  * USpoofChecker* sc = uspoof_open(&status);
192  * uspoof_setAllowedChars(sc, allowed, &status);
193  * uspoof_setRestrictionLevel(sc, USPOOF_MODERATELY_RESTRICTIVE);
194  *
195  * int32_t bitmask = uspoof_check(sc, str, -1, NULL, &status);
196  * UBool result = bitmask != 0;
197  * // fails checks: 1 (status: U_ZERO_ERROR)
198  * printf("fails checks: %d (status: %s)\n", result, u_errorName(status));
199  * uspoof_close(sc);
200  * uset_close(allowed);
201  * \endcode
202  *
203  * As in the case for confusability checking, it is good practice to create one <code>USpoofChecker</code> instance at
204  * startup, and call the cheaper {@link uspoof_check} online. We specify the set of
205  * allowed characters to be those with type RECOMMENDED or INCLUSION, according to the recommendation in UTS 39.
206  *
207  * In addition to {@link uspoof_check}, the function {@link uspoof_checkUTF8} is exposed for UTF8-encoded char* strings,
208  * and {@link uspoof_checkUnicodeString} is exposed for C++ programmers.
209  *
210  * If the {@link USPOOF_AUX_INFO} check is enabled, a limited amount of information on why a string failed the checks
211  * is available in the returned bitmask.  For complete information, use the {@link uspoof_check2} class of functions
212  * with a {@link USpoofCheckResult} parameter:
213  *
214  * \code{.c}
215  * UErrorCode status = U_ZERO_ERROR;
216  * UChar* str = (UChar*) u"p\u0430ypal";  // with U+0430 CYRILLIC SMALL LETTER A
217  *
218  * // Get the default set of allowable characters:
219  * USet* allowed = uset_openEmpty();
220  * uset_addAll(allowed, uspoof_getRecommendedSet(&status));
221  * uset_addAll(allowed, uspoof_getInclusionSet(&status));
222  *
223  * USpoofChecker* sc = uspoof_open(&status);
224  * uspoof_setAllowedChars(sc, allowed, &status);
225  * uspoof_setRestrictionLevel(sc, USPOOF_MODERATELY_RESTRICTIVE);
226  *
227  * USpoofCheckResult* checkResult = uspoof_openCheckResult(&status);
228  * int32_t bitmask = uspoof_check2(sc, str, -1, checkResult, &status);
229  *
230  * int32_t failures1 = bitmask;
231  * int32_t failures2 = uspoof_getCheckResultChecks(checkResult, &status);
232  * assert(failures1 == failures2);
233  * // checks that failed: 0x00000010 (status: U_ZERO_ERROR)
234  * printf("checks that failed: %#010x (status: %s)\n", failures1, u_errorName(status));
235  *
236  * // Cleanup:
237  * uspoof_close(sc);
238  * uset_close(allowed);
239  * uspoof_closeCheckResult(checkResult);
240  * \endcode
241  *
242  * C++ users can take advantage of a few syntactical conveniences.  The following snippet is functionally
243  * equivalent to the one above:
244  *
245  * \code{.cpp}
246  * UErrorCode status = U_ZERO_ERROR;
247  * UnicodeString str((UChar*) u"p\u0430ypal");  // with U+0430 CYRILLIC SMALL LETTER A
248  *
249  * // Get the default set of allowable characters:
250  * UnicodeSet allowed;
251  * allowed.addAll(*uspoof_getRecommendedUnicodeSet(&status));
252  * allowed.addAll(*uspoof_getInclusionUnicodeSet(&status));
253  *
254  * LocalUSpoofCheckerPointer sc(uspoof_open(&status));
255  * uspoof_setAllowedChars(sc.getAlias(), allowed.toUSet(), &status);
256  * uspoof_setRestrictionLevel(sc.getAlias(), USPOOF_MODERATELY_RESTRICTIVE);
257  *
258  * LocalUSpoofCheckResultPointer checkResult(uspoof_openCheckResult(&status));
259  * int32_t bitmask = uspoof_check2UnicodeString(sc.getAlias(), str, checkResult.getAlias(), &status);
260  *
261  * int32_t failures1 = bitmask;
262  * int32_t failures2 = uspoof_getCheckResultChecks(checkResult.getAlias(), &status);
263  * assert(failures1 == failures2);
264  * // checks that failed: 0x00000010 (status: U_ZERO_ERROR)
265  * printf("checks that failed: %#010x (status: %s)\n", failures1, u_errorName(status));
266  *
267  * // Explicit cleanup not necessary.
268  * \endcode
269  *
270  * The return value is a bitmask of the checks that failed. In this case, there was one check that failed:
271  * {@link USPOOF_RESTRICTION_LEVEL}, corresponding to the fifth bit (16). The possible checks are:
272  *
273  * <ul>
274  * <li><code>RESTRICTION_LEVEL</code>: flags strings that violate the
275  * <a href="http://unicode.org/reports/tr39/#Restriction_Level_Detection">Restriction Level</a> test as specified in UTS
276  * 39; in most cases, this means flagging strings that contain characters from multiple different scripts.</li>
277  * <li><code>INVISIBLE</code>: flags strings that contain invisible characters, such as zero-width spaces, or character
278  * sequences that are likely not to display, such as multiple occurrences of the same non-spacing mark.</li>
279  * <li><code>CHAR_LIMIT</code>: flags strings that contain characters outside of a specified set of acceptable
280  * characters. See {@link uspoof_setAllowedChars} and {@link uspoof_setAllowedLocales}.</li>
281  * <li><code>MIXED_NUMBERS</code>: flags strings that contain digits from multiple different numbering systems.</li>
282  * </ul>
283  *
284  * <p>
285  * These checks can be enabled independently of each other. For example, if you were interested in checking for only the
286  * INVISIBLE and MIXED_NUMBERS conditions, you could do:
287  *
288  * \code{.c}
289  * UErrorCode status = U_ZERO_ERROR;
290  * UChar* str = (UChar*) u"8\u09EA";  // 8 mixed with U+09EA BENGALI DIGIT FOUR
291  *
292  * USpoofChecker* sc = uspoof_open(&status);
293  * uspoof_setChecks(sc, USPOOF_INVISIBLE | USPOOF_MIXED_NUMBERS, &status);
294  *
295  * int32_t bitmask = uspoof_check2(sc, str, -1, NULL, &status);
296  * UBool result = bitmask != 0;
297  * // fails checks: 1 (status: U_ZERO_ERROR)
298  * printf("fails checks: %d (status: %s)\n", result, u_errorName(status));
299  * uspoof_close(sc);
300  * \endcode
301  *
302  * Here is an example in C++ showing how to compute the restriction level of a string:
303  *
304  * \code{.cpp}
305  * UErrorCode status = U_ZERO_ERROR;
306  * UnicodeString str((UChar*) u"p\u0430ypal");  // with U+0430 CYRILLIC SMALL LETTER A
307  *
308  * // Get the default set of allowable characters:
309  * UnicodeSet allowed;
310  * allowed.addAll(*uspoof_getRecommendedUnicodeSet(&status));
311  * allowed.addAll(*uspoof_getInclusionUnicodeSet(&status));
312  *
313  * LocalUSpoofCheckerPointer sc(uspoof_open(&status));
314  * uspoof_setAllowedChars(sc.getAlias(), allowed.toUSet(), &status);
315  * uspoof_setRestrictionLevel(sc.getAlias(), USPOOF_MODERATELY_RESTRICTIVE);
316  * uspoof_setChecks(sc.getAlias(), USPOOF_RESTRICTION_LEVEL | USPOOF_AUX_INFO, &status);
317  *
318  * LocalUSpoofCheckResultPointer checkResult(uspoof_openCheckResult(&status));
319  * int32_t bitmask = uspoof_check2UnicodeString(sc.getAlias(), str, checkResult.getAlias(), &status);
320  *
321  * URestrictionLevel restrictionLevel = uspoof_getCheckResultRestrictionLevel(checkResult.getAlias(), &status);
322  * // Since USPOOF_AUX_INFO was enabled, the restriction level is also available in the upper bits of the bitmask:
323  * assert((restrictionLevel & bitmask) == restrictionLevel);
324  * // Restriction level: 0x50000000 (status: U_ZERO_ERROR)
325  * printf("Restriction level: %#010x (status: %s)\n", restrictionLevel, u_errorName(status));
326  * \endcode
327  *
328  * The code '0x50000000' corresponds to the restriction level USPOOF_MINIMALLY_RESTRICTIVE.  Since
329  * USPOOF_MINIMALLY_RESTRICTIVE is weaker than USPOOF_MODERATELY_RESTRICTIVE, the string fails the check.
330  *
331  * <b>Note:</b> The Restriction Level is the most powerful of the checks. The full logic is documented in
332  * <a href="http://unicode.org/reports/tr39/#Restriction_Level_Detection">UTS 39</a>, but the basic idea is that strings
333  * are restricted to contain characters from only a single script, <em>except</em> that most scripts are allowed to have
334  * Latin characters interspersed. Although the default restriction level is <code>HIGHLY_RESTRICTIVE</code>, it is
335  * recommended that users set their restriction level to <code>MODERATELY_RESTRICTIVE</code>, which allows Latin mixed
336  * with all other scripts except Cyrillic, Greek, and Cherokee, with which it is often confusable. For more details on
337  * the levels, see UTS 39 or {@link URestrictionLevel}. The Restriction Level test is aware of the set of
338  * allowed characters set in {@link uspoof_setAllowedChars}. Note that characters which have script code
339  * COMMON or INHERITED, such as numbers and punctuation, are ignored when computing whether a string has multiple
340  * scripts.
341  *
342  * <h2>Additional Information</h2>
343  *
344  * A <code>USpoofChecker</code> instance may be used repeatedly to perform checks on any number of identifiers.
345  *
346  * <b>Thread Safety:</b> The test functions for checking a single identifier, or for testing whether
347  * two identifiers are possible confusable, are thread safe. They may called concurrently, from multiple threads,
348  * using the same USpoofChecker instance.
349  *
350  * More generally, the standard ICU thread safety rules apply: functions that take a const USpoofChecker parameter are
351  * thread safe. Those that take a non-const USpoofChecker are not thread safe..
352  *
353  * @stable ICU 4.6
354  */
355 
356 struct USpoofChecker;
357 /**
358  * @stable ICU 4.2
359  */
360 typedef struct USpoofChecker USpoofChecker; /**< typedef for C of USpoofChecker */
361 
362 struct USpoofCheckResult;
363 /**
364  * @see uspoof_openCheckResult
365  * @stable ICU 58
366  */
367 typedef struct USpoofCheckResult USpoofCheckResult;
368 
369 /**
370  * Enum for the kinds of checks that USpoofChecker can perform.
371  * These enum values are used both to select the set of checks that
372  * will be performed, and to report results from the check function.
373  *
374  * @stable ICU 4.2
375  */
376 typedef enum USpoofChecks {
377     /**
378      * When performing the two-string {@link uspoof_areConfusable} test, this flag in the return value indicates
379      * that the two strings are visually confusable and that they are from the same script, according to UTS 39 section
380      * 4.
381      *
382      * @see uspoof_areConfusable
383      * @stable ICU 4.2
384      */
385     USPOOF_SINGLE_SCRIPT_CONFUSABLE =   1,
386 
387     /**
388      * When performing the two-string {@link uspoof_areConfusable} test, this flag in the return value indicates
389      * that the two strings are visually confusable and that they are <b>not</b> from the same script, according to UTS
390      * 39 section 4.
391      *
392      * @see uspoof_areConfusable
393      * @stable ICU 4.2
394      */
395     USPOOF_MIXED_SCRIPT_CONFUSABLE  =   2,
396 
397     /**
398      * When performing the two-string {@link uspoof_areConfusable} test, this flag in the return value indicates
399      * that the two strings are visually confusable and that they are not from the same script but both of them are
400      * single-script strings, according to UTS 39 section 4.
401      *
402      * @see uspoof_areConfusable
403      * @stable ICU 4.2
404      */
405     USPOOF_WHOLE_SCRIPT_CONFUSABLE  =   4,
406 
407     /**
408      * Enable this flag in {@link uspoof_setChecks} to turn on all types of confusables.  You may set
409      * the checks to some subset of SINGLE_SCRIPT_CONFUSABLE, MIXED_SCRIPT_CONFUSABLE, or WHOLE_SCRIPT_CONFUSABLE to
410      * make {@link uspoof_areConfusable} return only those types of confusables.
411      *
412      * @see uspoof_areConfusable
413      * @see uspoof_getSkeleton
414      * @stable ICU 58
415      */
416     USPOOF_CONFUSABLE               =   USPOOF_SINGLE_SCRIPT_CONFUSABLE | USPOOF_MIXED_SCRIPT_CONFUSABLE | USPOOF_WHOLE_SCRIPT_CONFUSABLE,
417 
418 #ifndef U_HIDE_DEPRECATED_API
419     /**
420       * This flag is deprecated and no longer affects the behavior of SpoofChecker.
421       *
422       * @deprecated ICU 58  Any case confusable mappings were removed from UTS 39; the corresponding ICU API was deprecated.
423       */
424     USPOOF_ANY_CASE                 =   8,
425 #endif  /* U_HIDE_DEPRECATED_API */
426 
427     /**
428       * Check that an identifier is no looser than the specified RestrictionLevel.
429       * The default if {@link uspoof_setRestrictionLevel} is not called is HIGHLY_RESTRICTIVE.
430       *
431       * If USPOOF_AUX_INFO is enabled the actual restriction level of the
432       * identifier being tested will also be returned by uspoof_check().
433       *
434       * @see URestrictionLevel
435       * @see uspoof_setRestrictionLevel
436       * @see USPOOF_AUX_INFO
437       *
438       * @stable ICU 51
439       */
440     USPOOF_RESTRICTION_LEVEL        = 16,
441 
442 #ifndef U_HIDE_DEPRECATED_API
443     /** Check that an identifier contains only characters from a
444       * single script (plus chars from the common and inherited scripts.)
445       * Applies to checks of a single identifier check only.
446       * @deprecated ICU 51  Use RESTRICTION_LEVEL instead.
447       */
448     USPOOF_SINGLE_SCRIPT            =  USPOOF_RESTRICTION_LEVEL,
449 #endif  /* U_HIDE_DEPRECATED_API */
450 
451     /** Check an identifier for the presence of invisible characters,
452       * such as zero-width spaces, or character sequences that are
453       * likely not to display, such as multiple occurrences of the same
454       * non-spacing mark.  This check does not test the input string as a whole
455       * for conformance to any particular syntax for identifiers.
456       */
457     USPOOF_INVISIBLE                =  32,
458 
459     /** Check that an identifier contains only characters from a specified set
460       * of acceptable characters.  See {@link uspoof_setAllowedChars} and
461       * {@link uspoof_setAllowedLocales}.  Note that a string that fails this check
462       * will also fail the {@link USPOOF_RESTRICTION_LEVEL} check.
463       */
464     USPOOF_CHAR_LIMIT               =  64,
465 
466     /**
467      * Check that an identifier does not mix numbers from different numbering systems.
468      * For more information, see UTS 39 section 5.3.
469      *
470      * @stable ICU 51
471      */
472     USPOOF_MIXED_NUMBERS            = 128,
473 
474 #ifndef U_HIDE_DRAFT_API
475     /**
476      * Check that an identifier does not have a combining character following a character in which that
477      * combining character would be hidden; for example 'i' followed by a U+0307 combining dot.
478      *
479      * More specifically, the following characters are forbidden from preceding a U+0307:
480      * <ul>
481      * <li>Those with the Soft_Dotted Unicode property (which includes 'i' and 'j')</li>
482      * <li>Latin lowercase letter 'l'</li>
483      * <li>Dotless 'i' and 'j' ('ı' and 'ȷ', U+0131 and U+0237)</li>
484      * <li>Any character whose confusable prototype ends with such a character
485      * (Soft_Dotted, 'l', 'ı', or 'ȷ')</li>
486      * </ul>
487      * In addition, combining characters are allowed between the above characters and U+0307 except those
488      * with combining class 0 or combining class "Above" (230, same class as U+0307).
489      *
490      * This list and the number of combing characters considered by this check may grow over time.
491      *
492      * @draft ICU 62
493      */
494     USPOOF_HIDDEN_OVERLAY            = 256,
495 #endif  /* U_HIDE_DRAFT_API */
496 
497    /**
498      * Enable all spoof checks.
499      *
500      * @stable ICU 4.6
501      */
502     USPOOF_ALL_CHECKS               = 0xFFFF,
503 
504     /**
505       * Enable the return of auxillary (non-error) information in the
506       * upper bits of the check results value.
507       *
508       * If this "check" is not enabled, the results of {@link uspoof_check} will be
509       * zero when an identifier passes all of the enabled checks.
510       *
511       * If this "check" is enabled, (uspoof_check() & {@link USPOOF_ALL_CHECKS}) will
512       * be zero when an identifier passes all checks.
513       *
514       * @stable ICU 51
515       */
516     USPOOF_AUX_INFO                  = 0x40000000
517 
518     } USpoofChecks;
519 
520 
521     /**
522      * Constants from UAX #39 for use in {@link uspoof_setRestrictionLevel}, and
523      * for returned identifier restriction levels in check results.
524      *
525      * @stable ICU 51
526      *
527      * @see uspoof_setRestrictionLevel
528      * @see uspoof_check
529      */
530     typedef enum URestrictionLevel {
531         /**
532          * All characters in the string are in the identifier profile and all characters in the string are in the
533          * ASCII range.
534          *
535          * @stable ICU 51
536          */
537         USPOOF_ASCII = 0x10000000,
538         /**
539          * The string classifies as ASCII-Only, or all characters in the string are in the identifier profile and
540          * the string is single-script, according to the definition in UTS 39 section 5.1.
541          *
542          * @stable ICU 53
543          */
544         USPOOF_SINGLE_SCRIPT_RESTRICTIVE = 0x20000000,
545         /**
546          * The string classifies as Single Script, or all characters in the string are in the identifier profile and
547          * the string is covered by any of the following sets of scripts, according to the definition in UTS 39
548          * section 5.1:
549          * <ul>
550          *   <li>Latin + Han + Bopomofo (or equivalently: Latn + Hanb)</li>
551          *   <li>Latin + Han + Hiragana + Katakana (or equivalently: Latn + Jpan)</li>
552          *   <li>Latin + Han + Hangul (or equivalently: Latn +Kore)</li>
553          * </ul>
554          * This is the default restriction in ICU.
555          *
556          * @stable ICU 51
557          */
558         USPOOF_HIGHLY_RESTRICTIVE = 0x30000000,
559         /**
560          * The string classifies as Highly Restrictive, or all characters in the string are in the identifier profile
561          * and the string is covered by Latin and any one other Recommended or Aspirational script, except Cyrillic,
562          * Greek, and Cherokee.
563          *
564          * @stable ICU 51
565          */
566         USPOOF_MODERATELY_RESTRICTIVE = 0x40000000,
567         /**
568          * All characters in the string are in the identifier profile.  Allow arbitrary mixtures of scripts.
569          *
570          * @stable ICU 51
571          */
572         USPOOF_MINIMALLY_RESTRICTIVE = 0x50000000,
573         /**
574          * Any valid identifiers, including characters outside of the Identifier Profile.
575          *
576          * @stable ICU 51
577          */
578         USPOOF_UNRESTRICTIVE = 0x60000000,
579         /**
580          * Mask for selecting the Restriction Level bits from the return value of {@link uspoof_check}.
581          *
582          * @stable ICU 53
583          */
584         USPOOF_RESTRICTION_LEVEL_MASK = 0x7F000000,
585 #ifndef U_HIDE_INTERNAL_API
586         /**
587          * An undefined restriction level.
588          * @internal
589          */
590         USPOOF_UNDEFINED_RESTRICTIVE = -1
591 #endif  /* U_HIDE_INTERNAL_API */
592     } URestrictionLevel;
593 
594 /**
595  *  Create a Unicode Spoof Checker, configured to perform all
596  *  checks except for USPOOF_LOCALE_LIMIT and USPOOF_CHAR_LIMIT.
597  *  Note that additional checks may be added in the future,
598  *  resulting in the changes to the default checking behavior.
599  *
600  *  @param status  The error code, set if this function encounters a problem.
601  *  @return        the newly created Spoof Checker
602  *  @stable ICU 4.2
603  */
604 U_STABLE USpoofChecker * U_EXPORT2
605 uspoof_open(UErrorCode *status);
606 
607 
608 /**
609  * Open a Spoof checker from its serialized form, stored in 32-bit-aligned memory.
610  * Inverse of uspoof_serialize().
611  * The memory containing the serialized data must remain valid and unchanged
612  * as long as the spoof checker, or any cloned copies of the spoof checker,
613  * are in use.  Ownership of the memory remains with the caller.
614  * The spoof checker (and any clones) must be closed prior to deleting the
615  * serialized data.
616  *
617  * @param data a pointer to 32-bit-aligned memory containing the serialized form of spoof data
618  * @param length the number of bytes available at data;
619  *               can be more than necessary
620  * @param pActualLength receives the actual number of bytes at data taken up by the data;
621  *                      can be NULL
622  * @param pErrorCode ICU error code
623  * @return the spoof checker.
624  *
625  * @see uspoof_open
626  * @see uspoof_serialize
627  * @stable ICU 4.2
628  */
629 U_STABLE USpoofChecker * U_EXPORT2
630 uspoof_openFromSerialized(const void *data, int32_t length, int32_t *pActualLength,
631                           UErrorCode *pErrorCode);
632 
633 /**
634   * Open a Spoof Checker from the source form of the spoof data.
635   * The input corresponds to the Unicode data file confusables.txt
636   * as described in Unicode UAX #39.  The syntax of the source data
637   * is as described in UAX #39 for this file, and the content of
638   * this file is acceptable input.
639   *
640   * The character encoding of the (char *) input text is UTF-8.
641   *
642   * @param confusables a pointer to the confusable characters definitions,
643   *                    as found in file confusables.txt from unicode.org.
644   * @param confusablesLen The length of the confusables text, or -1 if the
645   *                    input string is zero terminated.
646   * @param confusablesWholeScript
647   *                    Deprecated in ICU 58.  No longer used.
648   * @param confusablesWholeScriptLen
649   *                    Deprecated in ICU 58.  No longer used.
650   * @param errType     In the event of an error in the input, indicates
651   *                    which of the input files contains the error.
652   *                    The value is one of USPOOF_SINGLE_SCRIPT_CONFUSABLE or
653   *                    USPOOF_WHOLE_SCRIPT_CONFUSABLE, or
654   *                    zero if no errors are found.
655   * @param pe          In the event of an error in the input, receives the position
656   *                    in the input text (line, offset) of the error.
657   * @param status      an in/out ICU UErrorCode.  Among the possible errors is
658   *                    U_PARSE_ERROR, which is used to report syntax errors
659   *                    in the input.
660   * @return            A spoof checker that uses the rules from the input files.
661   * @stable ICU 4.2
662   */
663 U_STABLE USpoofChecker * U_EXPORT2
664 uspoof_openFromSource(const char *confusables,  int32_t confusablesLen,
665                       const char *confusablesWholeScript, int32_t confusablesWholeScriptLen,
666                       int32_t *errType, UParseError *pe, UErrorCode *status);
667 
668 
669 /**
670   * Close a Spoof Checker, freeing any memory that was being held by
671   *   its implementation.
672   * @stable ICU 4.2
673   */
674 U_STABLE void U_EXPORT2
675 uspoof_close(USpoofChecker *sc);
676 
677 #if U_SHOW_CPLUSPLUS_API
678 
679 U_NAMESPACE_BEGIN
680 
681 /**
682  * \class LocalUSpoofCheckerPointer
683  * "Smart pointer" class, closes a USpoofChecker via uspoof_close().
684  * For most methods see the LocalPointerBase base class.
685  *
686  * @see LocalPointerBase
687  * @see LocalPointer
688  * @stable ICU 4.4
689  */
690 U_DEFINE_LOCAL_OPEN_POINTER(LocalUSpoofCheckerPointer, USpoofChecker, uspoof_close);
691 
692 U_NAMESPACE_END
693 
694 #endif
695 
696 /**
697  * Clone a Spoof Checker.  The clone will be set to perform the same checks
698  *   as the original source.
699  *
700  * @param sc       The source USpoofChecker
701  * @param status   The error code, set if this function encounters a problem.
702  * @return
703  * @stable ICU 4.2
704  */
705 U_STABLE USpoofChecker * U_EXPORT2
706 uspoof_clone(const USpoofChecker *sc, UErrorCode *status);
707 
708 
709 /**
710  * Specify the bitmask of checks that will be performed by {@link uspoof_check}. Calling this method
711  * overwrites any checks that may have already been enabled. By default, all checks are enabled.
712  *
713  * To enable specific checks and disable all others, the "whitelisted" checks should be ORed together. For
714  * example, to fail strings containing characters outside of the set specified by {@link uspoof_setAllowedChars} and
715  * also strings that contain digits from mixed numbering systems:
716  *
717  * <pre>
718  * {@code
719  * uspoof_setChecks(USPOOF_CHAR_LIMIT | USPOOF_MIXED_NUMBERS);
720  * }
721  * </pre>
722  *
723  * To disable specific checks and enable all others, the "blacklisted" checks should be ANDed away from
724  * ALL_CHECKS. For example, if you are not planning to use the {@link uspoof_areConfusable} functionality,
725  * it is good practice to disable the CONFUSABLE check:
726  *
727  * <pre>
728  * {@code
729  * uspoof_setChecks(USPOOF_ALL_CHECKS & ~USPOOF_CONFUSABLE);
730  * }
731  * </pre>
732  *
733  * Note that methods such as {@link uspoof_setAllowedChars}, {@link uspoof_setAllowedLocales}, and
734  * {@link uspoof_setRestrictionLevel} will enable certain checks when called. Those methods will OR the check they
735  * enable onto the existing bitmask specified by this method. For more details, see the documentation of those
736  * methods.
737  *
738  * @param sc       The USpoofChecker
739  * @param checks         The set of checks that this spoof checker will perform.
740  *                 The value is a bit set, obtained by OR-ing together
741  *                 values from enum USpoofChecks.
742  * @param status   The error code, set if this function encounters a problem.
743  * @stable ICU 4.2
744  *
745  */
746 U_STABLE void U_EXPORT2
747 uspoof_setChecks(USpoofChecker *sc, int32_t checks, UErrorCode *status);
748 
749 /**
750  * Get the set of checks that this Spoof Checker has been configured to perform.
751  *
752  * @param sc       The USpoofChecker
753  * @param status   The error code, set if this function encounters a problem.
754  * @return         The set of checks that this spoof checker will perform.
755  *                 The value is a bit set, obtained by OR-ing together
756  *                 values from enum USpoofChecks.
757  * @stable ICU 4.2
758  *
759  */
760 U_STABLE int32_t U_EXPORT2
761 uspoof_getChecks(const USpoofChecker *sc, UErrorCode *status);
762 
763 /**
764  * Set the loosest restriction level allowed for strings. The default if this is not called is
765  * {@link USPOOF_HIGHLY_RESTRICTIVE}. Calling this method enables the {@link USPOOF_RESTRICTION_LEVEL} and
766  * {@link USPOOF_MIXED_NUMBERS} checks, corresponding to Sections 5.1 and 5.2 of UTS 39. To customize which checks are
767  * to be performed by {@link uspoof_check}, see {@link uspoof_setChecks}.
768  *
769  * @param sc       The USpoofChecker
770  * @param restrictionLevel The loosest restriction level allowed.
771  * @see URestrictionLevel
772  * @stable ICU 51
773  */
774 U_STABLE void U_EXPORT2
775 uspoof_setRestrictionLevel(USpoofChecker *sc, URestrictionLevel restrictionLevel);
776 
777 
778 /**
779   * Get the Restriction Level that will be tested if the checks include {@link USPOOF_RESTRICTION_LEVEL}.
780   *
781   * @return The restriction level
782   * @see URestrictionLevel
783   * @stable ICU 51
784   */
785 U_STABLE URestrictionLevel U_EXPORT2
786 uspoof_getRestrictionLevel(const USpoofChecker *sc);
787 
788 /**
789  * Limit characters that are acceptable in identifiers being checked to those
790  * normally used with the languages associated with the specified locales.
791  * Any previously specified list of locales is replaced by the new settings.
792  *
793  * A set of languages is determined from the locale(s), and
794  * from those a set of acceptable Unicode scripts is determined.
795  * Characters from this set of scripts, along with characters from
796  * the "common" and "inherited" Unicode Script categories
797  * will be permitted.
798  *
799  * Supplying an empty string removes all restrictions;
800  * characters from any script will be allowed.
801  *
802  * The {@link USPOOF_CHAR_LIMIT} test is automatically enabled for this
803  * USpoofChecker when calling this function with a non-empty list
804  * of locales.
805  *
806  * The Unicode Set of characters that will be allowed is accessible
807  * via the uspoof_getAllowedChars() function.  uspoof_setAllowedLocales()
808  * will <i>replace</i> any previously applied set of allowed characters.
809  *
810  * Adjustments, such as additions or deletions of certain classes of characters,
811  * can be made to the result of uspoof_setAllowedLocales() by
812  * fetching the resulting set with uspoof_getAllowedChars(),
813  * manipulating it with the Unicode Set API, then resetting the
814  * spoof detectors limits with uspoof_setAllowedChars().
815  *
816  * @param sc           The USpoofChecker
817  * @param localesList  A list list of locales, from which the language
818  *                     and associated script are extracted.  The locales
819  *                     are comma-separated if there is more than one.
820  *                     White space may not appear within an individual locale,
821  *                     but is ignored otherwise.
822  *                     The locales are syntactically like those from the
823  *                     HTTP Accept-Language header.
824  *                     If the localesList is empty, no restrictions will be placed on
825  *                     the allowed characters.
826  *
827  * @param status       The error code, set if this function encounters a problem.
828  * @stable ICU 4.2
829  */
830 U_STABLE void U_EXPORT2
831 uspoof_setAllowedLocales(USpoofChecker *sc, const char *localesList, UErrorCode *status);
832 
833 /**
834  * Get a list of locales for the scripts that are acceptable in strings
835  *  to be checked.  If no limitations on scripts have been specified,
836  *  an empty string will be returned.
837  *
838  *  uspoof_setAllowedChars() will reset the list of allowed to be empty.
839  *
840  *  The format of the returned list is the same as that supplied to
841  *  uspoof_setAllowedLocales(), but returned list may not be identical
842  *  to the originally specified string; the string may be reformatted,
843  *  and information other than languages from
844  *  the originally specified locales may be omitted.
845  *
846  * @param sc           The USpoofChecker
847  * @param status       The error code, set if this function encounters a problem.
848  * @return             A string containing a list of  locales corresponding
849  *                     to the acceptable scripts, formatted like an
850  *                     HTTP Accept Language value.
851  *
852  * @stable ICU 4.2
853  */
854 U_STABLE const char * U_EXPORT2
855 uspoof_getAllowedLocales(USpoofChecker *sc, UErrorCode *status);
856 
857 
858 /**
859  * Limit the acceptable characters to those specified by a Unicode Set.
860  *   Any previously specified character limit is
861  *   is replaced by the new settings.  This includes limits on
862  *   characters that were set with the uspoof_setAllowedLocales() function.
863  *
864  * The USPOOF_CHAR_LIMIT test is automatically enabled for this
865  * USpoofChecker by this function.
866  *
867  * @param sc       The USpoofChecker
868  * @param chars    A Unicode Set containing the list of
869  *                 characters that are permitted.  Ownership of the set
870  *                 remains with the caller.  The incoming set is cloned by
871  *                 this function, so there are no restrictions on modifying
872  *                 or deleting the USet after calling this function.
873  * @param status   The error code, set if this function encounters a problem.
874  * @stable ICU 4.2
875  */
876 U_STABLE void U_EXPORT2
877 uspoof_setAllowedChars(USpoofChecker *sc, const USet *chars, UErrorCode *status);
878 
879 
880 /**
881  * Get a USet for the characters permitted in an identifier.
882  * This corresponds to the limits imposed by the Set Allowed Characters
883  * functions. Limitations imposed by other checks will not be
884  * reflected in the set returned by this function.
885  *
886  * The returned set will be frozen, meaning that it cannot be modified
887  * by the caller.
888  *
889  * Ownership of the returned set remains with the Spoof Detector.  The
890  * returned set will become invalid if the spoof detector is closed,
891  * or if a new set of allowed characters is specified.
892  *
893  *
894  * @param sc       The USpoofChecker
895  * @param status   The error code, set if this function encounters a problem.
896  * @return         A USet containing the characters that are permitted by
897  *                 the USPOOF_CHAR_LIMIT test.
898  * @stable ICU 4.2
899  */
900 U_STABLE const USet * U_EXPORT2
901 uspoof_getAllowedChars(const USpoofChecker *sc, UErrorCode *status);
902 
903 
904 #if U_SHOW_CPLUSPLUS_API
905 /**
906  * Limit the acceptable characters to those specified by a Unicode Set.
907  *   Any previously specified character limit is
908  *   is replaced by the new settings.    This includes limits on
909  *   characters that were set with the uspoof_setAllowedLocales() function.
910  *
911  * The USPOOF_CHAR_LIMIT test is automatically enabled for this
912  * USoofChecker by this function.
913  *
914  * @param sc       The USpoofChecker
915  * @param chars    A Unicode Set containing the list of
916  *                 characters that are permitted.  Ownership of the set
917  *                 remains with the caller.  The incoming set is cloned by
918  *                 this function, so there are no restrictions on modifying
919  *                 or deleting the UnicodeSet after calling this function.
920  * @param status   The error code, set if this function encounters a problem.
921  * @stable ICU 4.2
922  */
923 U_STABLE void U_EXPORT2
924 uspoof_setAllowedUnicodeSet(USpoofChecker *sc, const icu::UnicodeSet *chars, UErrorCode *status);
925 
926 
927 /**
928  * Get a UnicodeSet for the characters permitted in an identifier.
929  * This corresponds to the limits imposed by the Set Allowed Characters /
930  * UnicodeSet functions. Limitations imposed by other checks will not be
931  * reflected in the set returned by this function.
932  *
933  * The returned set will be frozen, meaning that it cannot be modified
934  * by the caller.
935  *
936  * Ownership of the returned set remains with the Spoof Detector.  The
937  * returned set will become invalid if the spoof detector is closed,
938  * or if a new set of allowed characters is specified.
939  *
940  *
941  * @param sc       The USpoofChecker
942  * @param status   The error code, set if this function encounters a problem.
943  * @return         A UnicodeSet containing the characters that are permitted by
944  *                 the USPOOF_CHAR_LIMIT test.
945  * @stable ICU 4.2
946  */
947 U_STABLE const icu::UnicodeSet * U_EXPORT2
948 uspoof_getAllowedUnicodeSet(const USpoofChecker *sc, UErrorCode *status);
949 #endif
950 
951 
952 /**
953  * Check the specified string for possible security issues.
954  * The text to be checked will typically be an identifier of some sort.
955  * The set of checks to be performed is specified with uspoof_setChecks().
956  *
957  * \note
958  *   Consider using the newer API, {@link uspoof_check2}, instead.
959  *   The newer API exposes additional information from the check procedure
960  *   and is otherwise identical to this method.
961  *
962  * @param sc      The USpoofChecker
963  * @param id      The identifier to be checked for possible security issues,
964  *                in UTF-16 format.
965  * @param length  the length of the string to be checked, expressed in
966  *                16 bit UTF-16 code units, or -1 if the string is
967  *                zero terminated.
968  * @param position  Deprecated in ICU 51.  Always returns zero.
969  *                Originally, an out parameter for the index of the first
970  *                string position that failed a check.
971  *                This parameter may be NULL.
972  * @param status  The error code, set if an error occurred while attempting to
973  *                perform the check.
974  *                Spoofing or security issues detected with the input string are
975  *                not reported here, but through the function's return value.
976  * @return        An integer value with bits set for any potential security
977  *                or spoofing issues detected.  The bits are defined by
978  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
979  *                will be zero if the input string passes all of the
980  *                enabled checks.
981  * @see uspoof_check2
982  * @stable ICU 4.2
983  */
984 U_STABLE int32_t U_EXPORT2
985 uspoof_check(const USpoofChecker *sc,
986                          const UChar *id, int32_t length,
987                          int32_t *position,
988                          UErrorCode *status);
989 
990 
991 /**
992  * Check the specified string for possible security issues.
993  * The text to be checked will typically be an identifier of some sort.
994  * The set of checks to be performed is specified with uspoof_setChecks().
995  *
996  * \note
997  *   Consider using the newer API, {@link uspoof_check2UTF8}, instead.
998  *   The newer API exposes additional information from the check procedure
999  *   and is otherwise identical to this method.
1000  *
1001  * @param sc      The USpoofChecker
1002  * @param id      A identifier to be checked for possible security issues, in UTF8 format.
1003  * @param length  the length of the string to be checked, or -1 if the string is
1004  *                zero terminated.
1005  * @param position  Deprecated in ICU 51.  Always returns zero.
1006  *                Originally, an out parameter for the index of the first
1007  *                string position that failed a check.
1008  *                This parameter may be NULL.
1009  * @param status  The error code, set if an error occurred while attempting to
1010  *                perform the check.
1011  *                Spoofing or security issues detected with the input string are
1012  *                not reported here, but through the function's return value.
1013  *                If the input contains invalid UTF-8 sequences,
1014  *                a status of U_INVALID_CHAR_FOUND will be returned.
1015  * @return        An integer value with bits set for any potential security
1016  *                or spoofing issues detected.  The bits are defined by
1017  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1018  *                will be zero if the input string passes all of the
1019  *                enabled checks.
1020  * @see uspoof_check2UTF8
1021  * @stable ICU 4.2
1022  */
1023 U_STABLE int32_t U_EXPORT2
1024 uspoof_checkUTF8(const USpoofChecker *sc,
1025                  const char *id, int32_t length,
1026                  int32_t *position,
1027                  UErrorCode *status);
1028 
1029 
1030 #if U_SHOW_CPLUSPLUS_API
1031 /**
1032  * Check the specified string for possible security issues.
1033  * The text to be checked will typically be an identifier of some sort.
1034  * The set of checks to be performed is specified with uspoof_setChecks().
1035  *
1036  * \note
1037  *   Consider using the newer API, {@link uspoof_check2UnicodeString}, instead.
1038  *   The newer API exposes additional information from the check procedure
1039  *   and is otherwise identical to this method.
1040  *
1041  * @param sc      The USpoofChecker
1042  * @param id      A identifier to be checked for possible security issues.
1043  * @param position  Deprecated in ICU 51.  Always returns zero.
1044  *                Originally, an out parameter for the index of the first
1045  *                string position that failed a check.
1046  *                This parameter may be NULL.
1047  * @param status  The error code, set if an error occurred while attempting to
1048  *                perform the check.
1049  *                Spoofing or security issues detected with the input string are
1050  *                not reported here, but through the function's return value.
1051  * @return        An integer value with bits set for any potential security
1052  *                or spoofing issues detected.  The bits are defined by
1053  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1054  *                will be zero if the input string passes all of the
1055  *                enabled checks.
1056  * @see uspoof_check2UnicodeString
1057  * @stable ICU 4.2
1058  */
1059 U_STABLE int32_t U_EXPORT2
1060 uspoof_checkUnicodeString(const USpoofChecker *sc,
1061                           const icu::UnicodeString &id,
1062                           int32_t *position,
1063                           UErrorCode *status);
1064 #endif
1065 
1066 
1067 /**
1068  * Check the specified string for possible security issues.
1069  * The text to be checked will typically be an identifier of some sort.
1070  * The set of checks to be performed is specified with uspoof_setChecks().
1071  *
1072  * @param sc      The USpoofChecker
1073  * @param id      The identifier to be checked for possible security issues,
1074  *                in UTF-16 format.
1075  * @param length  the length of the string to be checked, or -1 if the string is
1076  *                zero terminated.
1077  * @param checkResult  An instance of USpoofCheckResult to be filled with
1078  *                details about the identifier.  Can be NULL.
1079  * @param status  The error code, set if an error occurred while attempting to
1080  *                perform the check.
1081  *                Spoofing or security issues detected with the input string are
1082  *                not reported here, but through the function's return value.
1083  * @return        An integer value with bits set for any potential security
1084  *                or spoofing issues detected.  The bits are defined by
1085  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1086  *                will be zero if the input string passes all of the
1087  *                enabled checks.  Any information in this bitmask will be
1088  *                consistent with the information saved in the optional
1089  *                checkResult parameter.
1090  * @see uspoof_openCheckResult
1091  * @see uspoof_check2UTF8
1092  * @see uspoof_check2UnicodeString
1093  * @stable ICU 58
1094  */
1095 U_STABLE int32_t U_EXPORT2
1096 uspoof_check2(const USpoofChecker *sc,
1097     const UChar* id, int32_t length,
1098     USpoofCheckResult* checkResult,
1099     UErrorCode *status);
1100 
1101 /**
1102  * Check the specified string for possible security issues.
1103  * The text to be checked will typically be an identifier of some sort.
1104  * The set of checks to be performed is specified with uspoof_setChecks().
1105  *
1106  * This version of {@link uspoof_check} accepts a USpoofCheckResult, which
1107  * returns additional information about the identifier.  For more
1108  * information, see {@link uspoof_openCheckResult}.
1109  *
1110  * @param sc      The USpoofChecker
1111  * @param id      A identifier to be checked for possible security issues, in UTF8 format.
1112  * @param length  the length of the string to be checked, or -1 if the string is
1113  *                zero terminated.
1114  * @param checkResult  An instance of USpoofCheckResult to be filled with
1115  *                details about the identifier.  Can be NULL.
1116  * @param status  The error code, set if an error occurred while attempting to
1117  *                perform the check.
1118  *                Spoofing or security issues detected with the input string are
1119  *                not reported here, but through the function's return value.
1120  * @return        An integer value with bits set for any potential security
1121  *                or spoofing issues detected.  The bits are defined by
1122  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1123  *                will be zero if the input string passes all of the
1124  *                enabled checks.  Any information in this bitmask will be
1125  *                consistent with the information saved in the optional
1126  *                checkResult parameter.
1127  * @see uspoof_openCheckResult
1128  * @see uspoof_check2
1129  * @see uspoof_check2UnicodeString
1130  * @stable ICU 58
1131  */
1132 U_STABLE int32_t U_EXPORT2
1133 uspoof_check2UTF8(const USpoofChecker *sc,
1134     const char *id, int32_t length,
1135     USpoofCheckResult* checkResult,
1136     UErrorCode *status);
1137 
1138 #if U_SHOW_CPLUSPLUS_API
1139 /**
1140  * Check the specified string for possible security issues.
1141  * The text to be checked will typically be an identifier of some sort.
1142  * The set of checks to be performed is specified with uspoof_setChecks().
1143  *
1144  * @param sc      The USpoofChecker
1145  * @param id      A identifier to be checked for possible security issues.
1146  * @param checkResult  An instance of USpoofCheckResult to be filled with
1147  *                details about the identifier.  Can be NULL.
1148  * @param status  The error code, set if an error occurred while attempting to
1149  *                perform the check.
1150  *                Spoofing or security issues detected with the input string are
1151  *                not reported here, but through the function's return value.
1152  * @return        An integer value with bits set for any potential security
1153  *                or spoofing issues detected.  The bits are defined by
1154  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1155  *                will be zero if the input string passes all of the
1156  *                enabled checks.  Any information in this bitmask will be
1157  *                consistent with the information saved in the optional
1158  *                checkResult parameter.
1159  * @see uspoof_openCheckResult
1160  * @see uspoof_check2
1161  * @see uspoof_check2UTF8
1162  * @stable ICU 58
1163  */
1164 U_STABLE int32_t U_EXPORT2
1165 uspoof_check2UnicodeString(const USpoofChecker *sc,
1166     const icu::UnicodeString &id,
1167     USpoofCheckResult* checkResult,
1168     UErrorCode *status);
1169 #endif
1170 
1171 /**
1172  * Create a USpoofCheckResult, used by the {@link uspoof_check2} class of functions to return
1173  * information about the identifier.  Information includes:
1174  * <ul>
1175  *   <li>A bitmask of the checks that failed</li>
1176  *   <li>The identifier's restriction level (UTS 39 section 5.2)</li>
1177  *   <li>The set of numerics in the string (UTS 39 section 5.3)</li>
1178  * </ul>
1179  * The data held in a USpoofCheckResult is cleared whenever it is passed into a new call
1180  * of {@link uspoof_check2}.
1181  *
1182  * @param status  The error code, set if this function encounters a problem.
1183  * @return        the newly created USpoofCheckResult
1184  * @see uspoof_check2
1185  * @see uspoof_check2UTF8
1186  * @see uspoof_check2UnicodeString
1187  * @stable ICU 58
1188  */
1189 U_STABLE USpoofCheckResult* U_EXPORT2
1190 uspoof_openCheckResult(UErrorCode *status);
1191 
1192 /**
1193  * Close a USpoofCheckResult, freeing any memory that was being held by
1194  *   its implementation.
1195  *
1196  * @param checkResult  The instance of USpoofCheckResult to close
1197  * @stable ICU 58
1198  */
1199 U_STABLE void U_EXPORT2
1200 uspoof_closeCheckResult(USpoofCheckResult *checkResult);
1201 
1202 #if U_SHOW_CPLUSPLUS_API
1203 
1204 U_NAMESPACE_BEGIN
1205 
1206 /**
1207  * \class LocalUSpoofCheckResultPointer
1208  * "Smart pointer" class, closes a USpoofCheckResult via `uspoof_closeCheckResult()`.
1209  * For most methods see the LocalPointerBase base class.
1210  *
1211  * @see LocalPointerBase
1212  * @see LocalPointer
1213  * @stable ICU 58
1214  */
1215 
1216 /**
1217  * \cond
1218  * Note: Doxygen is giving a bogus warning on this U_DEFINE_LOCAL_OPEN_POINTER.
1219  *       For now, suppress with a Doxygen cond
1220  */
1221 U_DEFINE_LOCAL_OPEN_POINTER(LocalUSpoofCheckResultPointer, USpoofCheckResult, uspoof_closeCheckResult);
1222 /** \endcond */
1223 
1224 U_NAMESPACE_END
1225 
1226 #endif
1227 
1228 /**
1229  * Indicates which of the spoof check(s) have failed. The value is a bitwise OR of the constants for the tests
1230  * in question: USPOOF_RESTRICTION_LEVEL, USPOOF_CHAR_LIMIT, and so on.
1231  *
1232  * @param checkResult  The instance of USpoofCheckResult created by {@link uspoof_openCheckResult}
1233  * @param status       The error code, set if an error occurred.
1234  * @return        An integer value with bits set for any potential security
1235  *                or spoofing issues detected.  The bits are defined by
1236  *                enum USpoofChecks.  (returned_value & USPOOF_ALL_CHECKS)
1237  *                will be zero if the input string passes all of the
1238  *                enabled checks.
1239  * @see uspoof_setChecks
1240  * @stable ICU 58
1241  */
1242 U_STABLE int32_t U_EXPORT2
1243 uspoof_getCheckResultChecks(const USpoofCheckResult *checkResult, UErrorCode *status);
1244 
1245 /**
1246  * Gets the restriction level that the text meets, if the USPOOF_RESTRICTION_LEVEL check
1247  * was enabled; otherwise, undefined.
1248  *
1249  * @param checkResult  The instance of USpoofCheckResult created by {@link uspoof_openCheckResult}
1250  * @param status       The error code, set if an error occurred.
1251  * @return             The restriction level contained in the USpoofCheckResult
1252  * @see uspoof_setRestrictionLevel
1253  * @stable ICU 58
1254  */
1255 U_STABLE URestrictionLevel U_EXPORT2
1256 uspoof_getCheckResultRestrictionLevel(const USpoofCheckResult *checkResult, UErrorCode *status);
1257 
1258 /**
1259  * Gets the set of numerics found in the string, if the USPOOF_MIXED_NUMBERS check was enabled;
1260  * otherwise, undefined.  The set will contain the zero digit from each decimal number system found
1261  * in the input string.  Ownership of the returned USet remains with the USpoofCheckResult.
1262  * The USet will be free'd when {@link uspoof_closeCheckResult} is called.
1263  *
1264  * @param checkResult  The instance of USpoofCheckResult created by {@link uspoof_openCheckResult}
1265  * @return             The set of numerics contained in the USpoofCheckResult
1266  * @param status       The error code, set if an error occurred.
1267  * @stable ICU 58
1268  */
1269 U_STABLE const USet* U_EXPORT2
1270 uspoof_getCheckResultNumerics(const USpoofCheckResult *checkResult, UErrorCode *status);
1271 
1272 
1273 /**
1274  * Check the whether two specified strings are visually confusable.
1275  *
1276  * If the strings are confusable, the return value will be nonzero, as long as
1277  * {@link USPOOF_CONFUSABLE} was enabled in uspoof_setChecks().
1278  *
1279  * The bits in the return value correspond to flags for each of the classes of
1280  * confusables applicable to the two input strings.  According to UTS 39
1281  * section 4, the possible flags are:
1282  *
1283  * <ul>
1284  *   <li>{@link USPOOF_SINGLE_SCRIPT_CONFUSABLE}</li>
1285  *   <li>{@link USPOOF_MIXED_SCRIPT_CONFUSABLE}</li>
1286  *   <li>{@link USPOOF_WHOLE_SCRIPT_CONFUSABLE}</li>
1287  * </ul>
1288  *
1289  * If one or more of the above flags were not listed in uspoof_setChecks(), this
1290  * function will never report that class of confusable.  The check
1291  * {@link USPOOF_CONFUSABLE} enables all three flags.
1292  *
1293  *
1294  * @param sc      The USpoofChecker
1295  * @param id1     The first of the two identifiers to be compared for
1296  *                confusability.  The strings are in UTF-16 format.
1297  * @param length1 the length of the first identifer, expressed in
1298  *                16 bit UTF-16 code units, or -1 if the string is
1299  *                nul terminated.
1300  * @param id2     The second of the two identifiers to be compared for
1301  *                confusability.  The identifiers are in UTF-16 format.
1302  * @param length2 The length of the second identifiers, expressed in
1303  *                16 bit UTF-16 code units, or -1 if the string is
1304  *                nul terminated.
1305  * @param status  The error code, set if an error occurred while attempting to
1306  *                perform the check.
1307  *                Confusability of the identifiers is not reported here,
1308  *                but through this function's return value.
1309  * @return        An integer value with bit(s) set corresponding to
1310  *                the type of confusability found, as defined by
1311  *                enum USpoofChecks.  Zero is returned if the identifiers
1312  *                are not confusable.
1313  *
1314  * @stable ICU 4.2
1315  */
1316 U_STABLE int32_t U_EXPORT2
1317 uspoof_areConfusable(const USpoofChecker *sc,
1318                      const UChar *id1, int32_t length1,
1319                      const UChar *id2, int32_t length2,
1320                      UErrorCode *status);
1321 
1322 
1323 
1324 /**
1325  * A version of {@link uspoof_areConfusable} accepting strings in UTF-8 format.
1326  *
1327  * @param sc      The USpoofChecker
1328  * @param id1     The first of the two identifiers to be compared for
1329  *                confusability.  The strings are in UTF-8 format.
1330  * @param length1 the length of the first identifiers, in bytes, or -1
1331  *                if the string is nul terminated.
1332  * @param id2     The second of the two identifiers to be compared for
1333  *                confusability.  The strings are in UTF-8 format.
1334  * @param length2 The length of the second string in bytes, or -1
1335  *                if the string is nul terminated.
1336  * @param status  The error code, set if an error occurred while attempting to
1337  *                perform the check.
1338  *                Confusability of the strings is not reported here,
1339  *                but through this function's return value.
1340  * @return        An integer value with bit(s) set corresponding to
1341  *                the type of confusability found, as defined by
1342  *                enum USpoofChecks.  Zero is returned if the strings
1343  *                are not confusable.
1344  *
1345  * @stable ICU 4.2
1346  *
1347  * @see uspoof_areConfusable
1348  */
1349 U_STABLE int32_t U_EXPORT2
1350 uspoof_areConfusableUTF8(const USpoofChecker *sc,
1351                          const char *id1, int32_t length1,
1352                          const char *id2, int32_t length2,
1353                          UErrorCode *status);
1354 
1355 
1356 
1357 
1358 #if U_SHOW_CPLUSPLUS_API
1359 /**
1360  * A version of {@link uspoof_areConfusable} accepting UnicodeStrings.
1361  *
1362  * @param sc      The USpoofChecker
1363  * @param s1     The first of the two identifiers to be compared for
1364  *                confusability.  The strings are in UTF-8 format.
1365  * @param s2     The second of the two identifiers to be compared for
1366  *                confusability.  The strings are in UTF-8 format.
1367  * @param status  The error code, set if an error occurred while attempting to
1368  *                perform the check.
1369  *                Confusability of the identifiers is not reported here,
1370  *                but through this function's return value.
1371  * @return        An integer value with bit(s) set corresponding to
1372  *                the type of confusability found, as defined by
1373  *                enum USpoofChecks.  Zero is returned if the identifiers
1374  *                are not confusable.
1375  *
1376  * @stable ICU 4.2
1377  *
1378  * @see uspoof_areConfusable
1379  */
1380 U_STABLE int32_t U_EXPORT2
1381 uspoof_areConfusableUnicodeString(const USpoofChecker *sc,
1382                                   const icu::UnicodeString &s1,
1383                                   const icu::UnicodeString &s2,
1384                                   UErrorCode *status);
1385 #endif
1386 
1387 
1388 /**
1389  *  Get the "skeleton" for an identifier.
1390  *  Skeletons are a transformation of the input identifier;
1391  * Two identifiers are confusable if their skeletons are identical.
1392  *  See Unicode UAX #39 for additional information.
1393  *
1394  *  Using skeletons directly makes it possible to quickly check
1395  *  whether an identifier is confusable with any of some large
1396  *  set of existing identifiers, by creating an efficiently
1397  *  searchable collection of the skeletons.
1398  *
1399  * @param sc      The USpoofChecker
1400  * @param type    Deprecated in ICU 58.  You may pass any number.
1401  *                Originally, controlled which of the Unicode confusable data
1402  *                tables to use.
1403  * @param id      The input identifier whose skeleton will be computed.
1404  * @param length  The length of the input identifier, expressed in 16 bit
1405  *                UTF-16 code units, or -1 if the string is zero terminated.
1406  * @param dest    The output buffer, to receive the skeleton string.
1407  * @param destCapacity  The length of the output buffer, in 16 bit units.
1408  *                The destCapacity may be zero, in which case the function will
1409  *                return the actual length of the skeleton.
1410  * @param status  The error code, set if an error occurred while attempting to
1411  *                perform the check.
1412  * @return        The length of the skeleton string.  The returned length
1413  *                is always that of the complete skeleton, even when the
1414  *                supplied buffer is too small (or of zero length)
1415  *
1416  * @stable ICU 4.2
1417  * @see uspoof_areConfusable
1418  */
1419 U_STABLE int32_t U_EXPORT2
1420 uspoof_getSkeleton(const USpoofChecker *sc,
1421                    uint32_t type,
1422                    const UChar *id,  int32_t length,
1423                    UChar *dest, int32_t destCapacity,
1424                    UErrorCode *status);
1425 
1426 /**
1427  *  Get the "skeleton" for an identifier.
1428  *  Skeletons are a transformation of the input identifier;
1429  *  Two identifiers are confusable if their skeletons are identical.
1430  *  See Unicode UAX #39 for additional information.
1431  *
1432  *  Using skeletons directly makes it possible to quickly check
1433  *  whether an identifier is confusable with any of some large
1434  *  set of existing identifiers, by creating an efficiently
1435  *  searchable collection of the skeletons.
1436  *
1437  * @param sc      The USpoofChecker
1438  * @param type    Deprecated in ICU 58.  You may pass any number.
1439  *                Originally, controlled which of the Unicode confusable data
1440  *                tables to use.
1441  * @param id      The UTF-8 format identifier whose skeleton will be computed.
1442  * @param length  The length of the input string, in bytes,
1443  *                or -1 if the string is zero terminated.
1444  * @param dest    The output buffer, to receive the skeleton string.
1445  * @param destCapacity  The length of the output buffer, in bytes.
1446  *                The destCapacity may be zero, in which case the function will
1447  *                return the actual length of the skeleton.
1448  * @param status  The error code, set if an error occurred while attempting to
1449  *                perform the check.  Possible Errors include U_INVALID_CHAR_FOUND
1450  *                   for invalid UTF-8 sequences, and
1451  *                   U_BUFFER_OVERFLOW_ERROR if the destination buffer is too small
1452  *                   to hold the complete skeleton.
1453  * @return        The length of the skeleton string, in bytes.  The returned length
1454  *                is always that of the complete skeleton, even when the
1455  *                supplied buffer is too small (or of zero length)
1456  *
1457  * @stable ICU 4.2
1458  */
1459 U_STABLE int32_t U_EXPORT2
1460 uspoof_getSkeletonUTF8(const USpoofChecker *sc,
1461                        uint32_t type,
1462                        const char *id,  int32_t length,
1463                        char *dest, int32_t destCapacity,
1464                        UErrorCode *status);
1465 
1466 #if U_SHOW_CPLUSPLUS_API
1467 /**
1468  *  Get the "skeleton" for an identifier.
1469  *  Skeletons are a transformation of the input identifier;
1470  *  Two identifiers are confusable if their skeletons are identical.
1471  *  See Unicode UAX #39 for additional information.
1472  *
1473  *  Using skeletons directly makes it possible to quickly check
1474  *  whether an identifier is confusable with any of some large
1475  *  set of existing identifiers, by creating an efficiently
1476  *  searchable collection of the skeletons.
1477  *
1478  * @param sc      The USpoofChecker.
1479  * @param type    Deprecated in ICU 58.  You may pass any number.
1480  *                Originally, controlled which of the Unicode confusable data
1481  *                tables to use.
1482  * @param id      The input identifier whose skeleton will be computed.
1483  * @param dest    The output identifier, to receive the skeleton string.
1484  * @param status  The error code, set if an error occurred while attempting to
1485  *                perform the check.
1486  * @return        A reference to the destination (skeleton) string.
1487  *
1488  * @stable ICU 4.2
1489  */
1490 U_I18N_API icu::UnicodeString & U_EXPORT2
1491 uspoof_getSkeletonUnicodeString(const USpoofChecker *sc,
1492                                 uint32_t type,
1493                                 const icu::UnicodeString &id,
1494                                 icu::UnicodeString &dest,
1495                                 UErrorCode *status);
1496 #endif   /* U_SHOW_CPLUSPLUS_API */
1497 
1498 /**
1499   * Get the set of Candidate Characters for Inclusion in Identifiers, as defined
1500   * in http://unicode.org/Public/security/latest/xidmodifications.txt
1501   * and documented in http://www.unicode.org/reports/tr39/, Unicode Security Mechanisms.
1502   *
1503   * The returned set is frozen. Ownership of the set remains with the ICU library; it must not
1504   * be deleted by the caller.
1505   *
1506   * @param status The error code, set if a problem occurs while creating the set.
1507   *
1508   * @stable ICU 51
1509   */
1510 U_STABLE const USet * U_EXPORT2
1511 uspoof_getInclusionSet(UErrorCode *status);
1512 
1513 /**
1514   * Get the set of characters from Recommended Scripts for Inclusion in Identifiers, as defined
1515   * in http://unicode.org/Public/security/latest/xidmodifications.txt
1516   * and documented in http://www.unicode.org/reports/tr39/, Unicode Security Mechanisms.
1517   *
1518   * The returned set is frozen. Ownership of the set remains with the ICU library; it must not
1519   * be deleted by the caller.
1520   *
1521   * @param status The error code, set if a problem occurs while creating the set.
1522   *
1523   * @stable ICU 51
1524   */
1525 U_STABLE const USet * U_EXPORT2
1526 uspoof_getRecommendedSet(UErrorCode *status);
1527 
1528 #if U_SHOW_CPLUSPLUS_API
1529 
1530 /**
1531   * Get the set of Candidate Characters for Inclusion in Identifiers, as defined
1532   * in http://unicode.org/Public/security/latest/xidmodifications.txt
1533   * and documented in http://www.unicode.org/reports/tr39/, Unicode Security Mechanisms.
1534   *
1535   * The returned set is frozen. Ownership of the set remains with the ICU library; it must not
1536   * be deleted by the caller.
1537   *
1538   * @param status The error code, set if a problem occurs while creating the set.
1539   *
1540   * @stable ICU 51
1541   */
1542 U_STABLE const icu::UnicodeSet * U_EXPORT2
1543 uspoof_getInclusionUnicodeSet(UErrorCode *status);
1544 
1545 /**
1546   * Get the set of characters from Recommended Scripts for Inclusion in Identifiers, as defined
1547   * in http://unicode.org/Public/security/latest/xidmodifications.txt
1548   * and documented in http://www.unicode.org/reports/tr39/, Unicode Security Mechanisms.
1549   *
1550   * The returned set is frozen. Ownership of the set remains with the ICU library; it must not
1551   * be deleted by the caller.
1552   *
1553   * @param status The error code, set if a problem occurs while creating the set.
1554   *
1555   * @stable ICU 51
1556   */
1557 U_STABLE const icu::UnicodeSet * U_EXPORT2
1558 uspoof_getRecommendedUnicodeSet(UErrorCode *status);
1559 
1560 #endif /* U_SHOW_CPLUSPLUS_API */
1561 
1562 /**
1563  * Serialize the data for a spoof detector into a chunk of memory.
1564  * The flattened spoof detection tables can later be used to efficiently
1565  * instantiate a new Spoof Detector.
1566  *
1567  * The serialized spoof checker includes only the data compiled from the
1568  * Unicode data tables by uspoof_openFromSource(); it does not include
1569  * include any other state or configuration that may have been set.
1570  *
1571  * @param sc   the Spoof Detector whose data is to be serialized.
1572  * @param data a pointer to 32-bit-aligned memory to be filled with the data,
1573  *             can be NULL if capacity==0
1574  * @param capacity the number of bytes available at data,
1575  *                 or 0 for preflighting
1576  * @param status an in/out ICU UErrorCode; possible errors include:
1577  * - U_BUFFER_OVERFLOW_ERROR if the data storage block is too small for serialization
1578  * - U_ILLEGAL_ARGUMENT_ERROR  the data or capacity parameters are bad
1579  * @return the number of bytes written or needed for the spoof data
1580  *
1581  * @see utrie2_openFromSerialized()
1582  * @stable ICU 4.2
1583  */
1584 U_STABLE int32_t U_EXPORT2
1585 uspoof_serialize(USpoofChecker *sc,
1586                  void *data, int32_t capacity,
1587                  UErrorCode *status);
1588 
1589 
1590 #endif
1591 
1592 #endif   /* USPOOF_H */
1593