1 // © 2016 and later: Unicode, Inc. and others.
2 // License & terms of use: http://www.unicode.org/copyright.html
3 /*
4 *******************************************************************************
5 * Copyright (c) 1996-2016, International Business Machines Corporation
6 *               and others. All Rights Reserved.
7 *******************************************************************************
8 * File unorm.h
9 *
10 * Created by: Vladimir Weinstein 12052000
11 *
12 * Modification history :
13 *
14 * Date        Name        Description
15 * 02/01/01    synwee      Added normalization quickcheck enum and method.
16 */
17 #ifndef UNORM_H
18 #define UNORM_H
19 
20 #include "unicode/utypes.h"
21 
22 #if !UCONFIG_NO_NORMALIZATION
23 
24 #include "unicode/uiter.h"
25 #include "unicode/unorm2.h"
26 
27 /**
28  * \file
29  * \brief C API: Unicode Normalization
30  *
31  * Old Unicode normalization API.
32  *
33  * This API has been replaced by the unorm2.h API and is only available
34  * for backward compatibility. The functions here simply delegate to the
35  * unorm2.h functions, for example unorm2_getInstance() and unorm2_normalize().
36  * There is one exception: The new API does not provide a replacement for unorm_compare().
37  * Its declaration has been moved to unorm2.h.
38  *
39  * <code>unorm_normalize</code> transforms Unicode text into an equivalent composed or
40  * decomposed form, allowing for easier sorting and searching of text.
41  * <code>unorm_normalize</code> supports the standard normalization forms described in
42  * <a href="http://www.unicode.org/unicode/reports/tr15/" target="unicode">
43  * Unicode Standard Annex #15: Unicode Normalization Forms</a>.
44  *
45  * Characters with accents or other adornments can be encoded in
46  * several different ways in Unicode.  For example, take the character A-acute.
47  * In Unicode, this can be encoded as a single character (the
48  * "composed" form):
49  *
50  * \code
51  *      00C1    LATIN CAPITAL LETTER A WITH ACUTE
52  * \endcode
53  *
54  * or as two separate characters (the "decomposed" form):
55  *
56  * \code
57  *      0041    LATIN CAPITAL LETTER A
58  *      0301    COMBINING ACUTE ACCENT
59  * \endcode
60  *
61  * To a user of your program, however, both of these sequences should be
62  * treated as the same "user-level" character "A with acute accent".  When you are searching or
63  * comparing text, you must ensure that these two sequences are treated
64  * equivalently.  In addition, you must handle characters with more than one
65  * accent.  Sometimes the order of a character's combining accents is
66  * significant, while in other cases accent sequences in different orders are
67  * really equivalent.
68  *
69  * Similarly, the string "ffi" can be encoded as three separate letters:
70  *
71  * \code
72  *      0066    LATIN SMALL LETTER F
73  *      0066    LATIN SMALL LETTER F
74  *      0069    LATIN SMALL LETTER I
75  * \endcode
76  *
77  * or as the single character
78  *
79  * \code
80  *      FB03    LATIN SMALL LIGATURE FFI
81  * \endcode
82  *
83  * The ffi ligature is not a distinct semantic character, and strictly speaking
84  * it shouldn't be in Unicode at all, but it was included for compatibility
85  * with existing character sets that already provided it.  The Unicode standard
86  * identifies such characters by giving them "compatibility" decompositions
87  * into the corresponding semantic characters.  When sorting and searching, you
88  * will often want to use these mappings.
89  *
90  * <code>unorm_normalize</code> helps solve these problems by transforming text into the
91  * canonical composed and decomposed forms as shown in the first example above.
92  * In addition, you can have it perform compatibility decompositions so that
93  * you can treat compatibility characters the same as their equivalents.
94  * Finally, <code>unorm_normalize</code> rearranges accents into the proper canonical
95  * order, so that you do not have to worry about accent rearrangement on your
96  * own.
97  *
98  * Form FCD, "Fast C or D", is also designed for collation.
99  * It allows to work on strings that are not necessarily normalized
100  * with an algorithm (like in collation) that works under "canonical closure", i.e., it treats precomposed
101  * characters and their decomposed equivalents the same.
102  *
103  * It is not a normalization form because it does not provide for uniqueness of representation. Multiple strings
104  * may be canonically equivalent (their NFDs are identical) and may all conform to FCD without being identical
105  * themselves.
106  *
107  * The form is defined such that the "raw decomposition", the recursive canonical decomposition of each character,
108  * results in a string that is canonically ordered. This means that precomposed characters are allowed for as long
109  * as their decompositions do not need canonical reordering.
110  *
111  * Its advantage for a process like collation is that all NFD and most NFC texts - and many unnormalized texts -
112  * already conform to FCD and do not need to be normalized (NFD) for such a process. The FCD quick check will
113  * return UNORM_YES for most strings in practice.
114  *
115  * unorm_normalize(UNORM_FCD) may be implemented with UNORM_NFD.
116  *
117  * For more details on FCD see the collation design document:
118  * http://source.icu-project.org/repos/icu/icuhtml/trunk/design/collation/ICU_collation_design.htm
119  *
120  * ICU collation performs either NFD or FCD normalization automatically if normalization
121  * is turned on for the collator object.
122  * Beyond collation and string search, normalized strings may be useful for string equivalence comparisons,
123  * transliteration/transcription, unique representations, etc.
124  *
125  * The W3C generally recommends to exchange texts in NFC.
126  * Note also that most legacy character encodings use only precomposed forms and often do not
127  * encode any combining marks by themselves. For conversion to such character encodings the
128  * Unicode text needs to be normalized to NFC.
129  * For more usage examples, see the Unicode Standard Annex.
130  */
131 
132 // Do not conditionalize the following enum with #ifndef U_HIDE_DEPRECATED_API,
133 // it is needed for layout of Normalizer object.
134 /**
135  * Constants for normalization modes.
136  * @deprecated ICU 56 Use unorm2.h instead.
137  */
138 typedef enum {
139   /** No decomposition/composition. @deprecated ICU 56 Use unorm2.h instead. */
140   UNORM_NONE = 1,
141   /** Canonical decomposition. @deprecated ICU 56 Use unorm2.h instead. */
142   UNORM_NFD = 2,
143   /** Compatibility decomposition. @deprecated ICU 56 Use unorm2.h instead. */
144   UNORM_NFKD = 3,
145   /** Canonical decomposition followed by canonical composition. @deprecated ICU 56 Use unorm2.h instead. */
146   UNORM_NFC = 4,
147   /** Default normalization. @deprecated ICU 56 Use unorm2.h instead. */
148   UNORM_DEFAULT = UNORM_NFC,
149   /** Compatibility decomposition followed by canonical composition. @deprecated ICU 56 Use unorm2.h instead. */
150   UNORM_NFKC =5,
151   /** "Fast C or D" form. @deprecated ICU 56 Use unorm2.h instead. */
152   UNORM_FCD = 6,
153 
154   /** One more than the highest normalization mode constant. @deprecated ICU 56 Use unorm2.h instead. */
155   UNORM_MODE_COUNT
156 } UNormalizationMode;
157 
158 #ifndef U_HIDE_DEPRECATED_API
159 
160 /**
161  * Constants for options flags for normalization.
162  * Use 0 for default options,
163  * including normalization according to the Unicode version
164  * that is currently supported by ICU (see u_getUnicodeVersion).
165  * @deprecated ICU 56 Use unorm2.h instead.
166  */
167 enum {
168     /**
169      * Options bit set value to select Unicode 3.2 normalization
170      * (except NormalizationCorrections).
171      * At most one Unicode version can be selected at a time.
172      * @deprecated ICU 56 Use unorm2.h instead.
173      */
174     UNORM_UNICODE_3_2=0x20
175 };
176 
177 /**
178  * Lowest-order bit number of unorm_compare() options bits corresponding to
179  * normalization options bits.
180  *
181  * The options parameter for unorm_compare() uses most bits for
182  * itself and for various comparison and folding flags.
183  * The most significant bits, however, are shifted down and passed on
184  * to the normalization implementation.
185  * (That is, from unorm_compare(..., options, ...),
186  * options>>UNORM_COMPARE_NORM_OPTIONS_SHIFT will be passed on to the
187  * internal normalization functions.)
188  *
189  * @see unorm_compare
190  * @deprecated ICU 56 Use unorm2.h instead.
191  */
192 #define UNORM_COMPARE_NORM_OPTIONS_SHIFT 20
193 
194 /**
195  * Normalize a string.
196  * The string will be normalized according the specified normalization mode
197  * and options.
198  * The source and result buffers must not be the same, nor overlap.
199  *
200  * @param source The string to normalize.
201  * @param sourceLength The length of source, or -1 if NUL-terminated.
202  * @param mode The normalization mode; one of UNORM_NONE,
203  *             UNORM_NFD, UNORM_NFC, UNORM_NFKC, UNORM_NFKD, UNORM_DEFAULT.
204  * @param options The normalization options, ORed together (0 for no options).
205  * @param result A pointer to a buffer to receive the result string.
206  *               The result string is NUL-terminated if possible.
207  * @param resultLength The maximum size of result.
208  * @param status A pointer to a UErrorCode to receive any errors.
209  * @return The total buffer size needed; if greater than resultLength,
210  *         the output was truncated, and the error code is set to U_BUFFER_OVERFLOW_ERROR.
211  * @deprecated ICU 56 Use unorm2.h instead.
212  */
213 U_DEPRECATED int32_t U_EXPORT2
214 unorm_normalize(const UChar *source, int32_t sourceLength,
215                 UNormalizationMode mode, int32_t options,
216                 UChar *result, int32_t resultLength,
217                 UErrorCode *status);
218 
219 /**
220  * Performing quick check on a string, to quickly determine if the string is
221  * in a particular normalization format.
222  * Three types of result can be returned UNORM_YES, UNORM_NO or
223  * UNORM_MAYBE. Result UNORM_YES indicates that the argument
224  * string is in the desired normalized format, UNORM_NO determines that
225  * argument string is not in the desired normalized format. A
226  * UNORM_MAYBE result indicates that a more thorough check is required,
227  * the user may have to put the string in its normalized form and compare the
228  * results.
229  *
230  * @param source       string for determining if it is in a normalized format
231  * @param sourcelength length of source to test, or -1 if NUL-terminated
232  * @param mode         which normalization form to test for
233  * @param status       a pointer to a UErrorCode to receive any errors
234  * @return UNORM_YES, UNORM_NO or UNORM_MAYBE
235  *
236  * @see unorm_isNormalized
237  * @deprecated ICU 56 Use unorm2.h instead.
238  */
239 U_DEPRECATED UNormalizationCheckResult U_EXPORT2
240 unorm_quickCheck(const UChar *source, int32_t sourcelength,
241                  UNormalizationMode mode,
242                  UErrorCode *status);
243 
244 /**
245  * Performing quick check on a string; same as unorm_quickCheck but
246  * takes an extra options parameter like most normalization functions.
247  *
248  * @param src        String that is to be tested if it is in a normalization format.
249  * @param srcLength  Length of source to test, or -1 if NUL-terminated.
250  * @param mode       Which normalization form to test for.
251  * @param options    The normalization options, ORed together (0 for no options).
252  * @param pErrorCode ICU error code in/out parameter.
253  *                   Must fulfill U_SUCCESS before the function call.
254  * @return UNORM_YES, UNORM_NO or UNORM_MAYBE
255  *
256  * @see unorm_quickCheck
257  * @see unorm_isNormalized
258  * @deprecated ICU 56 Use unorm2.h instead.
259  */
260 U_DEPRECATED UNormalizationCheckResult U_EXPORT2
261 unorm_quickCheckWithOptions(const UChar *src, int32_t srcLength,
262                             UNormalizationMode mode, int32_t options,
263                             UErrorCode *pErrorCode);
264 
265 /**
266  * Test if a string is in a given normalization form.
267  * This is semantically equivalent to source.equals(normalize(source, mode)) .
268  *
269  * Unlike unorm_quickCheck(), this function returns a definitive result,
270  * never a "maybe".
271  * For NFD, NFKD, and FCD, both functions work exactly the same.
272  * For NFC and NFKC where quickCheck may return "maybe", this function will
273  * perform further tests to arrive at a TRUE/FALSE result.
274  *
275  * @param src        String that is to be tested if it is in a normalization format.
276  * @param srcLength  Length of source to test, or -1 if NUL-terminated.
277  * @param mode       Which normalization form to test for.
278  * @param pErrorCode ICU error code in/out parameter.
279  *                   Must fulfill U_SUCCESS before the function call.
280  * @return Boolean value indicating whether the source string is in the
281  *         "mode" normalization form.
282  *
283  * @see unorm_quickCheck
284  * @deprecated ICU 56 Use unorm2.h instead.
285  */
286 U_DEPRECATED UBool U_EXPORT2
287 unorm_isNormalized(const UChar *src, int32_t srcLength,
288                    UNormalizationMode mode,
289                    UErrorCode *pErrorCode);
290 
291 /**
292  * Test if a string is in a given normalization form; same as unorm_isNormalized but
293  * takes an extra options parameter like most normalization functions.
294  *
295  * @param src        String that is to be tested if it is in a normalization format.
296  * @param srcLength  Length of source to test, or -1 if NUL-terminated.
297  * @param mode       Which normalization form to test for.
298  * @param options    The normalization options, ORed together (0 for no options).
299  * @param pErrorCode ICU error code in/out parameter.
300  *                   Must fulfill U_SUCCESS before the function call.
301  * @return Boolean value indicating whether the source string is in the
302  *         "mode/options" normalization form.
303  *
304  * @see unorm_quickCheck
305  * @see unorm_isNormalized
306  * @deprecated ICU 56 Use unorm2.h instead.
307  */
308 U_DEPRECATED UBool U_EXPORT2
309 unorm_isNormalizedWithOptions(const UChar *src, int32_t srcLength,
310                               UNormalizationMode mode, int32_t options,
311                               UErrorCode *pErrorCode);
312 
313 /**
314  * Iterative normalization forward.
315  * This function (together with unorm_previous) is somewhat
316  * similar to the C++ Normalizer class (see its non-static functions).
317  *
318  * Iterative normalization is useful when only a small portion of a longer
319  * string/text needs to be processed.
320  *
321  * For example, the likelihood may be high that processing the first 10% of some
322  * text will be sufficient to find certain data.
323  * Another example: When one wants to concatenate two normalized strings and get a
324  * normalized result, it is much more efficient to normalize just a small part of
325  * the result around the concatenation place instead of re-normalizing everything.
326  *
327  * The input text is an instance of the C character iteration API UCharIterator.
328  * It may wrap around a simple string, a CharacterIterator, a Replaceable, or any
329  * other kind of text object.
330  *
331  * If a buffer overflow occurs, then the caller needs to reset the iterator to the
332  * old index and call the function again with a larger buffer - if the caller cares
333  * for the actual output.
334  * Regardless of the output buffer, the iterator will always be moved to the next
335  * normalization boundary.
336  *
337  * This function (like unorm_previous) serves two purposes:
338  *
339  * 1) To find the next boundary so that the normalization of the part of the text
340  * from the current position to that boundary does not affect and is not affected
341  * by the part of the text beyond that boundary.
342  *
343  * 2) To normalize the text up to the boundary.
344  *
345  * The second step is optional, per the doNormalize parameter.
346  * It is omitted for operations like string concatenation, where the two adjacent
347  * string ends need to be normalized together.
348  * In such a case, the output buffer will just contain a copy of the text up to the
349  * boundary.
350  *
351  * pNeededToNormalize is an output-only parameter. Its output value is only defined
352  * if normalization was requested (doNormalize) and successful (especially, no
353  * buffer overflow).
354  * It is useful for operations like a normalizing transliterator, where one would
355  * not want to replace a piece of text if it is not modified.
356  *
357  * If doNormalize==TRUE and pNeededToNormalize!=NULL then *pNeeded... is set TRUE
358  * if the normalization was necessary.
359  *
360  * If doNormalize==FALSE then *pNeededToNormalize will be set to FALSE.
361  *
362  * If the buffer overflows, then *pNeededToNormalize will be undefined;
363  * essentially, whenever U_FAILURE is true (like in buffer overflows), this result
364  * will be undefined.
365  *
366  * @param src The input text in the form of a C character iterator.
367  * @param dest The output buffer; can be NULL if destCapacity==0 for pure preflighting.
368  * @param destCapacity The number of UChars that fit into dest.
369  * @param mode The normalization mode.
370  * @param options The normalization options, ORed together (0 for no options).
371  * @param doNormalize Indicates if the source text up to the next boundary
372  *                    is to be normalized (TRUE) or just copied (FALSE).
373  * @param pNeededToNormalize Output flag indicating if the normalization resulted in
374  *                           different text from the input.
375  *                           Not defined if an error occurs including buffer overflow.
376  *                           Always FALSE if !doNormalize.
377  * @param pErrorCode ICU error code in/out parameter.
378  *                   Must fulfill U_SUCCESS before the function call.
379  * @return Length of output (number of UChars) when successful or buffer overflow.
380  *
381  * @see unorm_previous
382  * @see unorm_normalize
383  *
384  * @deprecated ICU 56 Use unorm2.h instead.
385  */
386 U_DEPRECATED int32_t U_EXPORT2
387 unorm_next(UCharIterator *src,
388            UChar *dest, int32_t destCapacity,
389            UNormalizationMode mode, int32_t options,
390            UBool doNormalize, UBool *pNeededToNormalize,
391            UErrorCode *pErrorCode);
392 
393 /**
394  * Iterative normalization backward.
395  * This function (together with unorm_next) is somewhat
396  * similar to the C++ Normalizer class (see its non-static functions).
397  * For all details see unorm_next.
398  *
399  * @param src The input text in the form of a C character iterator.
400  * @param dest The output buffer; can be NULL if destCapacity==0 for pure preflighting.
401  * @param destCapacity The number of UChars that fit into dest.
402  * @param mode The normalization mode.
403  * @param options The normalization options, ORed together (0 for no options).
404  * @param doNormalize Indicates if the source text up to the next boundary
405  *                    is to be normalized (TRUE) or just copied (FALSE).
406  * @param pNeededToNormalize Output flag indicating if the normalization resulted in
407  *                           different text from the input.
408  *                           Not defined if an error occurs including buffer overflow.
409  *                           Always FALSE if !doNormalize.
410  * @param pErrorCode ICU error code in/out parameter.
411  *                   Must fulfill U_SUCCESS before the function call.
412  * @return Length of output (number of UChars) when successful or buffer overflow.
413  *
414  * @see unorm_next
415  * @see unorm_normalize
416  *
417  * @deprecated ICU 56 Use unorm2.h instead.
418  */
419 U_DEPRECATED int32_t U_EXPORT2
420 unorm_previous(UCharIterator *src,
421                UChar *dest, int32_t destCapacity,
422                UNormalizationMode mode, int32_t options,
423                UBool doNormalize, UBool *pNeededToNormalize,
424                UErrorCode *pErrorCode);
425 
426 /**
427  * Concatenate normalized strings, making sure that the result is normalized as well.
428  *
429  * If both the left and the right strings are in
430  * the normalization form according to "mode/options",
431  * then the result will be
432  *
433  * \code
434  *     dest=normalize(left+right, mode, options)
435  * \endcode
436  *
437  * With the input strings already being normalized,
438  * this function will use unorm_next() and unorm_previous()
439  * to find the adjacent end pieces of the input strings.
440  * Only the concatenation of these end pieces will be normalized and
441  * then concatenated with the remaining parts of the input strings.
442  *
443  * It is allowed to have dest==left to avoid copying the entire left string.
444  *
445  * @param left Left source string, may be same as dest.
446  * @param leftLength Length of left source string, or -1 if NUL-terminated.
447  * @param right Right source string. Must not be the same as dest, nor overlap.
448  * @param rightLength Length of right source string, or -1 if NUL-terminated.
449  * @param dest The output buffer; can be NULL if destCapacity==0 for pure preflighting.
450  * @param destCapacity The number of UChars that fit into dest.
451  * @param mode The normalization mode.
452  * @param options The normalization options, ORed together (0 for no options).
453  * @param pErrorCode ICU error code in/out parameter.
454  *                   Must fulfill U_SUCCESS before the function call.
455  * @return Length of output (number of UChars) when successful or buffer overflow.
456  *
457  * @see unorm_normalize
458  * @see unorm_next
459  * @see unorm_previous
460  *
461  * @deprecated ICU 56 Use unorm2.h instead.
462  */
463 U_DEPRECATED int32_t U_EXPORT2
464 unorm_concatenate(const UChar *left, int32_t leftLength,
465                   const UChar *right, int32_t rightLength,
466                   UChar *dest, int32_t destCapacity,
467                   UNormalizationMode mode, int32_t options,
468                   UErrorCode *pErrorCode);
469 
470 #endif  /* U_HIDE_DEPRECATED_API */
471 #endif /* #if !UCONFIG_NO_NORMALIZATION */
472 #endif
473