/src/gnutls/lib/unistring/uninorm/composition.c
Line | Count | Source |
1 | | /* Canonical composition of Unicode characters. |
2 | | Copyright (C) 2002, 2006, 2009, 2011-2025 Free Software Foundation, Inc. |
3 | | Written by Bruno Haible <bruno@clisp.org>, 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 <string.h> |
24 | | |
25 | | struct composition_rule { char codes[6]; unsigned int combined; }; |
26 | | |
27 | | #include "composition-table.h" |
28 | | #include "composition-table-bounds.h" |
29 | | |
30 | | ucs4_t |
31 | | uc_composition (ucs4_t uc1, ucs4_t uc2) |
32 | 1.78M | { |
33 | 1.78M | if (uc1 <= UNINORM_COMPOSE_MAX_ARG1 && uc2 <= UNINORM_COMPOSE_MAX_ARG2) |
34 | 1.77M | { |
35 | 1.77M | if (uc2 >= 0x1161 && uc2 < 0x1161 + 21 |
36 | 1.77M | && uc1 >= 0x1100 && uc1 < 0x1100 + 19) |
37 | 3.44k | { |
38 | | /* Hangul: Combine single letter L and single letter V to form |
39 | | two-letter syllable LV. */ |
40 | 3.44k | return 0xAC00 + ((uc1 - 0x1100) * 21 + (uc2 - 0x1161)) * 28; |
41 | 3.44k | } |
42 | 1.76M | else if (uc2 > 0x11A7 && uc2 < 0x11A7 + 28 |
43 | 1.76M | && uc1 >= 0xAC00 && uc1 < 0xD7A4 && ((uc1 - 0xAC00) % 28) == 0) |
44 | 3.10k | { |
45 | | /* Hangul: Combine two-letter syllable LV with single-letter T |
46 | | to form three-letter syllable LVT. */ |
47 | 3.10k | return uc1 + (uc2 - 0x11A7); |
48 | 3.10k | } |
49 | 1.76M | else |
50 | 1.76M | { |
51 | | #if 0 |
52 | | unsigned int uc = MUL1 * uc1 * MUL2 * uc2; |
53 | | unsigned int index1 = uc >> composition_header_0; |
54 | | if (index1 < composition_header_1) |
55 | | { |
56 | | int lookup1 = u_composition.level1[index1]; |
57 | | if (lookup1 >= 0) |
58 | | { |
59 | | unsigned int index2 = (uc >> composition_header_2) & composition_header_3; |
60 | | int lookup2 = u_composition.level2[lookup1 + index2]; |
61 | | if (lookup2 >= 0) |
62 | | { |
63 | | unsigned int index3 = (uc & composition_header_4); |
64 | | unsigned int lookup3 = u_composition.level3[lookup2 + index3]; |
65 | | if ((lookup3 >> 16) == uc2) |
66 | | return lookup3 & ((1U << 16) - 1); |
67 | | } |
68 | | } |
69 | | } |
70 | | #else |
71 | 1.76M | char codes[6]; |
72 | 1.76M | const struct composition_rule *rule; |
73 | | |
74 | 1.76M | codes[0] = (uc1 >> 16) & 0xff; |
75 | 1.76M | codes[1] = (uc1 >> 8) & 0xff; |
76 | 1.76M | codes[2] = uc1 & 0xff; |
77 | 1.76M | codes[3] = (uc2 >> 16) & 0xff; |
78 | 1.76M | codes[4] = (uc2 >> 8) & 0xff; |
79 | 1.76M | codes[5] = uc2 & 0xff; |
80 | | |
81 | 1.76M | rule = gl_uninorm_compose_lookup (codes, 6); |
82 | 1.76M | if (rule != NULL) |
83 | 94.3k | return rule->combined; |
84 | 1.76M | #endif |
85 | 1.76M | } |
86 | 1.77M | } |
87 | 1.68M | return 0; |
88 | 1.78M | } |