| 1 |
/* Canonical decomposition of Unicode characters. |
| 2 |
Copyright (C) 2009-2026 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 |
#include <config.h> |
| 19 |
|
| 20 |
/* Specification. */ |
| 21 |
#include "uninorm.h" |
| 22 |
|
| 23 |
#include <stdlib.h> |
| 24 |
|
| 25 |
#include "uninorm/decomposition-table.h" |
| 26 |
|
| 27 |
int |
| 28 |
uc_canonical_decomposition (ucs4_t uc, ucs4_t *decomposition) |
| 29 |
{ |
| 30 |
if (uc >= 0xAC00 && uc < 0xD7A4) |
| 31 |
{ |
| 32 |
/* Hangul syllable. See Unicode standard, chapter 3, section |
| 33 |
"Hangul Syllable Decomposition", See also the clarification at |
| 34 |
<https://www.unicode.org/versions/Unicode5.1.0/>, section |
| 35 |
"Clarification of Hangul Jamo Handling". */ |
| 36 |
uc -= 0xAC00; |
| 37 |
unsigned int t = uc % 28; |
| 38 |
|
| 39 |
if (t == 0) |
| 40 |
{ |
| 41 |
uc = uc / 28; |
| 42 |
unsigned int v = uc % 21; |
| 43 |
unsigned int l = uc / 21; |
| 44 |
|
| 45 |
decomposition[0] = 0x1100 + l; |
| 46 |
decomposition[1] = 0x1161 + v; |
| 47 |
return 2; |
| 48 |
} |
| 49 |
else |
| 50 |
{ |
| 51 |
#if 1 /* Return the pairwise decomposition, not the full decomposition. */ |
| 52 |
decomposition[0] = 0xAC00 + uc - t; /* = 0xAC00 + (l * 21 + v) * 28; */ |
| 53 |
decomposition[1] = 0x11A7 + t; |
| 54 |
return 2; |
| 55 |
#else |
| 56 |
uc = uc / 28; |
| 57 |
unsigned int v = uc % 21; |
| 58 |
unsigned int l = uc / 21; |
| 59 |
|
| 60 |
decomposition[0] = 0x1100 + l; |
| 61 |
decomposition[1] = 0x1161 + v; |
| 62 |
decomposition[2] = 0x11A7 + t; |
| 63 |
return 3; |
| 64 |
#endif |
| 65 |
} |
| 66 |
} |
| 67 |
else if (uc < 0x110000) |
| 68 |
{ |
| 69 |
unsigned short entry = decomp_index (uc); |
| 70 |
/* An entry of (unsigned short)(-1) denotes an absent entry. |
| 71 |
Otherwise, bit 15 of the entry tells whether the decomposition |
| 72 |
is a canonical one. */ |
| 73 |
if (entry < 0x8000) |
| 74 |
{ |
| 75 |
const unsigned char *p = &gl_uninorm_decomp_chars_table[3 * entry]; |
| 76 |
unsigned int element = (p[0] << 16) | (p[1] << 8) | p[2]; |
| 77 |
/* The first element has 5 bits for the decomposition type. */ |
| 78 |
if (((element >> 18) & 0x1f) != UC_DECOMP_CANONICAL) |
| 79 |
abort (); |
| 80 |
unsigned int length = 1; |
| 81 |
for (;;) |
| 82 |
{ |
| 83 |
/* Every element has an 18 bits wide Unicode code point. */ |
| 84 |
*decomposition = element & 0x3ffff; |
| 85 |
/* Bit 23 tells whether there are more elements, */ |
| 86 |
if ((element & (1 << 23)) == 0) |
| 87 |
break; |
| 88 |
p += 3; |
| 89 |
element = (p[0] << 16) | (p[1] << 8) | p[2]; |
| 90 |
decomposition++; |
| 91 |
length++; |
| 92 |
} |
| 93 |
return length; |
| 94 |
} |
| 95 |
} |
| 96 |
return -1; |
| 97 |
} |