| 1 |
/* Normalization forms (composition and decomposition) of Unicode strings. |
| 2 |
Copyright (C) 2001-2002, 2009-2024 Free Software Foundation, Inc. |
| 3 |
Written by Bruno Haible <[email protected]>, 2009. |
| 4 |
|
| 5 |
This file is free software: you can redistribute it and/or modify |
| 6 |
it under the terms of the GNU Lesser General Public License as |
| 7 |
published by the Free Software Foundation; either version 2.1 of the |
| 8 |
License, or (at your option) any later version. |
| 9 |
|
| 10 |
This file is distributed in the hope that it will be useful, |
| 11 |
but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 |
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
| 13 |
GNU Lesser General Public License for more details. |
| 14 |
|
| 15 |
You should have received a copy of the GNU Lesser General Public License |
| 16 |
along with this program. If not, see <https://www.gnu.org/licenses/>. */ |
| 17 |
|
| 18 |
#ifndef _UNINORM_H |
| 19 |
#define _UNINORM_H |
| 20 |
|
| 21 |
/* Get size_t. */ |
| 22 |
#include <stddef.h> |
| 23 |
|
| 24 |
#include "unitypes.h" |
| 25 |
|
| 26 |
#if @HAVE_UNISTRING_WOE32DLL_H@ |
| 27 |
# include <unistring/woe32dll.h> |
| 28 |
#else |
| 29 |
# define LIBUNISTRING_DLL_VARIABLE |
| 30 |
#endif |
| 31 |
|
| 32 |
|
| 33 |
#ifdef __cplusplus |
| 34 |
extern "C" { |
| 35 |
#endif |
| 36 |
|
| 37 |
|
| 38 |
/* Conventions: |
| 39 |
|
| 40 |
All functions prefixed with u8_ operate on UTF-8 encoded strings. |
| 41 |
Their unit is an uint8_t (1 byte). |
| 42 |
|
| 43 |
All functions prefixed with u16_ operate on UTF-16 encoded strings. |
| 44 |
Their unit is an uint16_t (a 2-byte word). |
| 45 |
|
| 46 |
All functions prefixed with u32_ operate on UCS-4 encoded strings. |
| 47 |
Their unit is an uint32_t (a 4-byte word). |
| 48 |
|
| 49 |
All argument pairs (s, n) denote a Unicode string s[0..n-1] with exactly |
| 50 |
n units. |
| 51 |
|
| 52 |
Functions returning a string result take a (resultbuf, lengthp) argument |
| 53 |
pair. If resultbuf is not NULL and the result fits into *lengthp units, |
| 54 |
it is put in resultbuf, and resultbuf is returned. Otherwise, a freshly |
| 55 |
allocated string is returned. In both cases, *lengthp is set to the |
| 56 |
length (number of units) of the returned string. In case of error, |
| 57 |
NULL is returned and errno is set. */ |
| 58 |
|
| 59 |
|
| 60 |
enum |
| 61 |
{ |
| 62 |
UC_DECOMP_CANONICAL,/* Canonical decomposition. */ |
| 63 |
UC_DECOMP_FONT, /* <font> A font variant (e.g. a blackletter form). */ |
| 64 |
UC_DECOMP_NOBREAK, /* <noBreak> A no-break version of a space or hyphen. */ |
| 65 |
UC_DECOMP_INITIAL, /* <initial> An initial presentation form (Arabic). */ |
| 66 |
UC_DECOMP_MEDIAL, /* <medial> A medial presentation form (Arabic). */ |
| 67 |
UC_DECOMP_FINAL, /* <final> A final presentation form (Arabic). */ |
| 68 |
UC_DECOMP_ISOLATED,/* <isolated> An isolated presentation form (Arabic). */ |
| 69 |
UC_DECOMP_CIRCLE, /* <circle> An encircled form. */ |
| 70 |
UC_DECOMP_SUPER, /* <super> A superscript form. */ |
| 71 |
UC_DECOMP_SUB, /* <sub> A subscript form. */ |
| 72 |
UC_DECOMP_VERTICAL,/* <vertical> A vertical layout presentation form. */ |
| 73 |
UC_DECOMP_WIDE, /* <wide> A wide (or zenkaku) compatibility character. */ |
| 74 |
UC_DECOMP_NARROW, /* <narrow> A narrow (or hankaku) compatibility character. */ |
| 75 |
UC_DECOMP_SMALL, /* <small> A small variant form (CNS compatibility). */ |
| 76 |
UC_DECOMP_SQUARE, /* <square> A CJK squared font variant. */ |
| 77 |
UC_DECOMP_FRACTION,/* <fraction> A vulgar fraction form. */ |
| 78 |
UC_DECOMP_COMPAT /* <compat> Otherwise unspecified compatibility character. */ |
| 79 |
}; |
| 80 |
|
| 81 |
/* Maximum size of decomposition of a single Unicode character. */ |
| 82 |
#define UC_DECOMPOSITION_MAX_LENGTH 32 |
| 83 |
|
| 84 |
/* Return the character decomposition mapping of a Unicode character. |
| 85 |
DECOMPOSITION must point to an array of at least UC_DECOMPOSITION_MAX_LENGTH |
| 86 |
ucs_t elements. |
| 87 |
When a decomposition exists, DECOMPOSITION[0..N-1] and *DECOMP_TAG are |
| 88 |
filled and N is returned. Otherwise -1 is returned. */ |
| 89 |
extern int |
| 90 |
uc_decomposition (ucs4_t uc, int *decomp_tag, ucs4_t *decomposition); |
| 91 |
|
| 92 |
/* Return the canonical character decomposition mapping of a Unicode character. |
| 93 |
DECOMPOSITION must point to an array of at least UC_DECOMPOSITION_MAX_LENGTH |
| 94 |
ucs_t elements. |
| 95 |
When a decomposition exists, DECOMPOSITION[0..N-1] is filled and N is |
| 96 |
returned. Otherwise -1 is returned. */ |
| 97 |
extern int |
| 98 |
uc_canonical_decomposition (ucs4_t uc, ucs4_t *decomposition); |
| 99 |
|
| 100 |
|
| 101 |
/* Attempt to combine the Unicode characters uc1, uc2. |
| 102 |
uc1 is known to have canonical combining class 0. |
| 103 |
Return the combination of uc1 and uc2, if it exists. |
| 104 |
Return 0 otherwise. |
| 105 |
Not all decompositions can be recombined using this function. See the |
| 106 |
Unicode file CompositionExclusions.txt for details. */ |
| 107 |
extern ucs4_t |
| 108 |
uc_composition (ucs4_t uc1, ucs4_t uc2) |
| 109 |
_UC_ATTRIBUTE_CONST; |
| 110 |
|
| 111 |
|
| 112 |
/* An object of type uninorm_t denotes a Unicode normalization form. */ |
| 113 |
struct unicode_normalization_form; |
| 114 |
typedef const struct unicode_normalization_form *uninorm_t; |
| 115 |
|
| 116 |
/* UNINORM_NFD: Normalization form D: canonical decomposition. */ |
| 117 |
extern @GNULIB_UNINORM_NFD_DLL_VARIABLE@ const struct unicode_normalization_form uninorm_nfd; |
| 118 |
#define UNINORM_NFD (&uninorm_nfd) |
| 119 |
|
| 120 |
/* UNINORM_NFC: Normalization form C: canonical decomposition, then |
| 121 |
canonical composition. */ |
| 122 |
extern @GNULIB_UNINORM_NFC_DLL_VARIABLE@ const struct unicode_normalization_form uninorm_nfc; |
| 123 |
#define UNINORM_NFC (&uninorm_nfc) |
| 124 |
|
| 125 |
/* UNINORM_NFKD: Normalization form KD: compatibility decomposition. */ |
| 126 |
extern @GNULIB_UNINORM_NFKD_DLL_VARIABLE@ const struct unicode_normalization_form uninorm_nfkd; |
| 127 |
#define UNINORM_NFKD (&uninorm_nfkd) |
| 128 |
|
| 129 |
/* UNINORM_NFKC: Normalization form KC: compatibility decomposition, then |
| 130 |
canonical composition. */ |
| 131 |
extern @GNULIB_UNINORM_NFKC_DLL_VARIABLE@ const struct unicode_normalization_form uninorm_nfkc; |
| 132 |
#define UNINORM_NFKC (&uninorm_nfkc) |
| 133 |
|
| 134 |
/* Test whether a normalization form does compatibility decomposition. */ |
| 135 |
#define uninorm_is_compat_decomposing(nf) \ |
| 136 |
((* (const unsigned int *) (nf) >> 0) & 1) |
| 137 |
|
| 138 |
/* Test whether a normalization form includes canonical composition. */ |
| 139 |
#define uninorm_is_composing(nf) \ |
| 140 |
((* (const unsigned int *) (nf) >> 1) & 1) |
| 141 |
|
| 142 |
/* Return the decomposing variant of a normalization form. |
| 143 |
This maps NFC,NFD -> NFD and NFKC,NFKD -> NFKD. */ |
| 144 |
extern uninorm_t |
| 145 |
uninorm_decomposing_form (uninorm_t nf) |
| 146 |
_UC_ATTRIBUTE_PURE; |
| 147 |
|
| 148 |
|
| 149 |
/* Return the specified normalization form of a string. */ |
| 150 |
extern uint8_t * |
| 151 |
u8_normalize (uninorm_t nf, const uint8_t *s, size_t n, |
| 152 |
uint8_t *_UC_RESTRICT resultbuf, size_t *lengthp); |
| 153 |
extern uint16_t * |
| 154 |
u16_normalize (uninorm_t nf, const uint16_t *s, size_t n, |
| 155 |
uint16_t *_UC_RESTRICT resultbuf, size_t *lengthp); |
| 156 |
extern uint32_t * |
| 157 |
u32_normalize (uninorm_t nf, const uint32_t *s, size_t n, |
| 158 |
uint32_t *_UC_RESTRICT resultbuf, size_t *lengthp); |
| 159 |
|
| 160 |
|
| 161 |
/* Compare S1 and S2, ignoring differences in normalization. |
| 162 |
NF must be either UNINORM_NFD or UNINORM_NFKD. |
| 163 |
If successful, set *RESULTP to -1 if S1 < S2, 0 if S1 = S2, 1 if S1 > S2, and |
| 164 |
return 0. Upon failure, return -1 with errno set. */ |
| 165 |
extern int |
| 166 |
u8_normcmp (const uint8_t *s1, size_t n1, const uint8_t *s2, size_t n2, |
| 167 |
uninorm_t nf, int *resultp); |
| 168 |
extern int |
| 169 |
u16_normcmp (const uint16_t *s1, size_t n1, const uint16_t *s2, size_t n2, |
| 170 |
uninorm_t nf, int *resultp); |
| 171 |
extern int |
| 172 |
u32_normcmp (const uint32_t *s1, size_t n1, const uint32_t *s2, size_t n2, |
| 173 |
uninorm_t nf, int *resultp); |
| 174 |
|
| 175 |
|
| 176 |
/* Converts the string S of length N to a NUL-terminated byte sequence, in such |
| 177 |
a way that comparing uN_normxfrm (S1) and uN_normxfrm (S2) with uN_cmp2() is |
| 178 |
equivalent to comparing S1 and S2 with uN_normcoll(). |
| 179 |
NF must be either UNINORM_NFC or UNINORM_NFKC. */ |
| 180 |
extern char * |
| 181 |
u8_normxfrm (const uint8_t *s, size_t n, uninorm_t nf, |
| 182 |
char *resultbuf, size_t *lengthp); |
| 183 |
extern char * |
| 184 |
u16_normxfrm (const uint16_t *s, size_t n, uninorm_t nf, |
| 185 |
char *resultbuf, size_t *lengthp); |
| 186 |
extern char * |
| 187 |
u32_normxfrm (const uint32_t *s, size_t n, uninorm_t nf, |
| 188 |
char *resultbuf, size_t *lengthp); |
| 189 |
|
| 190 |
|
| 191 |
/* Compare S1 and S2, ignoring differences in normalization, using the |
| 192 |
collation rules of the current locale. |
| 193 |
NF must be either UNINORM_NFC or UNINORM_NFKC. |
| 194 |
If successful, set *RESULTP to -1 if S1 < S2, 0 if S1 = S2, 1 if S1 > S2, and |
| 195 |
return 0. Upon failure, return -1 with errno set. */ |
| 196 |
extern int |
| 197 |
u8_normcoll (const uint8_t *s1, size_t n1, const uint8_t *s2, size_t n2, |
| 198 |
uninorm_t nf, int *resultp); |
| 199 |
extern int |
| 200 |
u16_normcoll (const uint16_t *s1, size_t n1, const uint16_t *s2, size_t n2, |
| 201 |
uninorm_t nf, int *resultp); |
| 202 |
extern int |
| 203 |
u32_normcoll (const uint32_t *s1, size_t n1, const uint32_t *s2, size_t n2, |
| 204 |
uninorm_t nf, int *resultp); |
| 205 |
|
| 206 |
|
| 207 |
/* Normalization of a stream of Unicode characters. |
| 208 |
|
| 209 |
A "stream of Unicode characters" is essentially a function that accepts an |
| 210 |
ucs4_t argument repeatedly, optionally combined with a function that |
| 211 |
"flushes" the stream. */ |
| 212 |
|
| 213 |
/* Data type of a stream of Unicode characters that normalizes its input |
| 214 |
according to a given normalization form and passes the normalized character |
| 215 |
sequence to the encapsulated stream of Unicode characters. */ |
| 216 |
struct uninorm_filter; |
| 217 |
|
| 218 |
/* Bring data buffered in the filter to its destination, the encapsulated |
| 219 |
stream, then close and free the filter. |
| 220 |
Return 0 if successful, or -1 with errno set upon failure. */ |
| 221 |
extern int |
| 222 |
uninorm_filter_free (struct uninorm_filter *filter); |
| 223 |
|
| 224 |
/* Create and return a normalization filter for Unicode characters. |
| 225 |
The pair (stream_func, stream_data) is the encapsulated stream. |
| 226 |
stream_func (stream_data, uc) receives the Unicode character uc |
| 227 |
and returns 0 if successful, or -1 with errno set upon failure. |
| 228 |
Return the new filter, or NULL with errno set upon failure. */ |
| 229 |
extern struct uninorm_filter * |
| 230 |
uninorm_filter_create (uninorm_t nf, |
| 231 |
int (*stream_func) (void *stream_data, ucs4_t uc), |
| 232 |
void *stream_data) |
| 233 |
_GL_ATTRIBUTE_DEALLOC (uninorm_filter_free, 1); |
| 234 |
|
| 235 |
/* Stuff a Unicode character into a normalizing filter. |
| 236 |
Return 0 if successful, or -1 with errno set upon failure. */ |
| 237 |
extern int |
| 238 |
uninorm_filter_write (struct uninorm_filter *filter, ucs4_t uc); |
| 239 |
|
| 240 |
/* Bring data buffered in the filter to its destination, the encapsulated |
| 241 |
stream. |
| 242 |
Return 0 if successful, or -1 with errno set upon failure. |
| 243 |
Note! If after calling this function, additional characters are written |
| 244 |
into the filter, the resulting character sequence in the encapsulated stream |
| 245 |
will not necessarily be normalized. */ |
| 246 |
extern int |
| 247 |
uninorm_filter_flush (struct uninorm_filter *filter); |
| 248 |
|
| 249 |
|
| 250 |
#ifdef __cplusplus |
| 251 |
} |
| 252 |
#endif |
| 253 |
|
| 254 |
|
| 255 |
#endif /* _UNINORM_H */ |