rev |
line source |
nuclear@10
|
1 /*
|
nuclear@10
|
2 libdrawtext - a simple library for fast text rendering in OpenGL
|
nuclear@10
|
3 Copyright (C) 2011 John Tsiombikas <nuclear@member.fsf.org>
|
nuclear@10
|
4
|
nuclear@10
|
5 This program is free software: you can redistribute it and/or modify
|
nuclear@10
|
6 it under the terms of the GNU Lesser General Public License as published by
|
nuclear@10
|
7 the Free Software Foundation, either version 3 of the License, or
|
nuclear@10
|
8 (at your option) any later version.
|
nuclear@10
|
9
|
nuclear@10
|
10 This program is distributed in the hope that it will be useful,
|
nuclear@10
|
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
|
nuclear@10
|
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
nuclear@10
|
13 GNU Lesser General Public License for more details.
|
nuclear@10
|
14
|
nuclear@10
|
15 You should have received a copy of the GNU Lesser General Public License
|
nuclear@10
|
16 along with this program. If not, see <http://www.gnu.org/licenses/>.
|
nuclear@10
|
17 */
|
nuclear@10
|
18 #include "drawtext.h"
|
nuclear@10
|
19
|
nuclear@10
|
20 #define U8_IS_FIRST(x) (((((x) >> 7) & 1) == 0) || ((((x) >> 6) & 3) == 3))
|
nuclear@10
|
21
|
nuclear@10
|
22 static const char first_mask[] = {
|
nuclear@10
|
23 0,
|
nuclear@10
|
24 0x7f, /* single byte, 7 bits valid */
|
nuclear@10
|
25 0x1f, /* two-bytes, 5 bits valid */
|
nuclear@10
|
26 0xf, /* three-bytes, 4 bits valid */
|
nuclear@10
|
27 0x7 /* four-bytes, 3 bits valid */
|
nuclear@10
|
28 };
|
nuclear@10
|
29 static const char first_shift[] = { 0, 7, 5, 4, 3 }; /* see above */
|
nuclear@10
|
30
|
nuclear@10
|
31 #define CONT_PREFIX 0x80
|
nuclear@10
|
32 #define CONT_MASK 0x3f
|
nuclear@10
|
33 #define CONT_SHIFT 6
|
nuclear@10
|
34
|
nuclear@10
|
35 /* last charcodes for 1, 2, 3 or 4-byte utf8 chars */
|
nuclear@10
|
36 static const int utf8_lastcode[] = { 0x7f, 0x7ff, 0xfff, 0x1fffff };
|
nuclear@10
|
37
|
nuclear@10
|
38 #define prefix_mask(x) (~first_mask[x])
|
nuclear@10
|
39 #define prefix(x) ((prefix_mask(x) << 1) & 0xff)
|
nuclear@10
|
40
|
nuclear@10
|
41
|
nuclear@10
|
42 char *dtx_utf8_next_char(char *str)
|
nuclear@10
|
43 {
|
nuclear@10
|
44 return str + dtx_utf8_nbytes(str);
|
nuclear@10
|
45 }
|
nuclear@10
|
46
|
nuclear@10
|
47 int dtx_utf8_char_code(const char *str)
|
nuclear@10
|
48 {
|
nuclear@10
|
49 int i, nbytes, shift, code = 0;
|
nuclear@10
|
50 int mask;
|
nuclear@10
|
51
|
nuclear@10
|
52 if(!U8_IS_FIRST(*str)) {
|
nuclear@10
|
53 return -1;
|
nuclear@10
|
54 }
|
nuclear@10
|
55
|
nuclear@10
|
56 nbytes = dtx_utf8_nbytes(str);
|
nuclear@10
|
57 mask = first_mask[nbytes];
|
nuclear@10
|
58 shift = 0;
|
nuclear@10
|
59
|
nuclear@10
|
60 for(i=0; i<nbytes; i++) {
|
nuclear@10
|
61 if(!*str) {
|
nuclear@10
|
62 break;
|
nuclear@10
|
63 }
|
nuclear@10
|
64
|
nuclear@10
|
65 code = (code << shift) | (*str++ & mask);
|
nuclear@10
|
66 mask = 0x3f;
|
nuclear@10
|
67 shift = 6;
|
nuclear@10
|
68 }
|
nuclear@10
|
69 return code;
|
nuclear@10
|
70 }
|
nuclear@10
|
71
|
nuclear@10
|
72 int dtx_utf8_nbytes(const char *str)
|
nuclear@10
|
73 {
|
nuclear@10
|
74 int i, numset = 0;
|
nuclear@10
|
75 int c = *str;
|
nuclear@10
|
76
|
nuclear@10
|
77 if(!U8_IS_FIRST(c)) {
|
nuclear@10
|
78 for(i=0; !U8_IS_FIRST(str[i]); i++);
|
nuclear@10
|
79 return i;
|
nuclear@10
|
80 }
|
nuclear@10
|
81
|
nuclear@10
|
82 /* count the leading 1s */
|
nuclear@10
|
83 for(i=0; i<4; i++) {
|
nuclear@10
|
84 if(((c >> (7 - i)) & 1) == 0) {
|
nuclear@10
|
85 break;
|
nuclear@10
|
86 }
|
nuclear@10
|
87 numset++;
|
nuclear@10
|
88 }
|
nuclear@10
|
89
|
nuclear@10
|
90 if(!numset) {
|
nuclear@10
|
91 return 1;
|
nuclear@10
|
92 }
|
nuclear@10
|
93 return numset;
|
nuclear@10
|
94 }
|
nuclear@10
|
95
|
nuclear@10
|
96 int dtx_utf8_char_count(const char *str)
|
nuclear@10
|
97 {
|
nuclear@10
|
98 int n = 0;
|
nuclear@10
|
99
|
nuclear@10
|
100 while(*str) {
|
nuclear@10
|
101 n++;
|
nuclear@10
|
102 str = dtx_utf8_next_char((char*)str);
|
nuclear@10
|
103 }
|
nuclear@10
|
104 return n;
|
nuclear@10
|
105 }
|
nuclear@10
|
106
|
nuclear@10
|
107 size_t dtx_utf8_from_char_code(int code, char *buf)
|
nuclear@10
|
108 {
|
nuclear@10
|
109 size_t nbytes = 0;
|
nuclear@10
|
110 int i;
|
nuclear@10
|
111
|
nuclear@10
|
112 for(i=0; i<4; i++) {
|
nuclear@10
|
113 if(code <= utf8_lastcode[i]) {
|
nuclear@10
|
114 nbytes = i + 1;
|
nuclear@10
|
115 break;
|
nuclear@10
|
116 }
|
nuclear@10
|
117 }
|
nuclear@10
|
118
|
nuclear@10
|
119 if(!nbytes && buf) {
|
nuclear@10
|
120 for(i=0; i<(int)nbytes; i++) {
|
nuclear@10
|
121 int idx = nbytes - i - 1;
|
nuclear@10
|
122 int mask, shift, prefix;
|
nuclear@10
|
123
|
nuclear@10
|
124 if(idx > 0) {
|
nuclear@10
|
125 mask = CONT_MASK;
|
nuclear@10
|
126 shift = CONT_SHIFT;
|
nuclear@10
|
127 prefix = CONT_PREFIX;
|
nuclear@10
|
128 } else {
|
nuclear@10
|
129 mask = first_mask[nbytes];
|
nuclear@10
|
130 shift = first_shift[nbytes];
|
nuclear@10
|
131 prefix = prefix(nbytes);
|
nuclear@10
|
132 }
|
nuclear@10
|
133
|
nuclear@10
|
134 buf[idx] = (code & mask) | (prefix & ~mask);
|
nuclear@10
|
135 code >>= shift;
|
nuclear@10
|
136 }
|
nuclear@10
|
137 }
|
nuclear@10
|
138 return nbytes;
|
nuclear@10
|
139 }
|
nuclear@10
|
140
|
nuclear@10
|
141 size_t dtx_utf8_from_string(const wchar_t *str, char *buf)
|
nuclear@10
|
142 {
|
nuclear@10
|
143 size_t nbytes = 0;
|
nuclear@10
|
144 char *ptr = buf;
|
nuclear@10
|
145
|
nuclear@10
|
146 while(*str) {
|
nuclear@10
|
147 int cbytes = dtx_utf8_from_char_code(*str++, ptr);
|
nuclear@10
|
148 if(ptr) {
|
nuclear@10
|
149 ptr += cbytes;
|
nuclear@10
|
150 }
|
nuclear@10
|
151 nbytes += cbytes;
|
nuclear@10
|
152 }
|
nuclear@10
|
153 return nbytes;
|
nuclear@10
|
154 }
|