1 /* Access functions for CNS 11643, plane 1 handling.
2 Copyright (C) 1998-2022 Free Software Foundation, Inc.
3 This file is part of the GNU C Library.
4
5 The GNU C Library is free software; you can redistribute it and/or
6 modify it under the terms of the GNU Lesser General Public
7 License as published by the Free Software Foundation; either
8 version 2.1 of the License, or (at your option) any later version.
9
10 The GNU C Library 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 GNU
13 Lesser General Public License for more details.
14
15 You should have received a copy of the GNU Lesser General Public
16 License along with the GNU C Library; if not, see
17 <https://www.gnu.org/licenses/>. */
18
19 #include <stdint.h>
20 #include <gconv.h>
21
22 /* Table for CNS 11643, plane 1 to UCS4 conversion. */
23 extern const uint16_t __cns11643l1_to_ucs4_tab[];
24
25
uint32_t(always_inline)26 static inline uint32_t
27 __attribute ((always_inline))
28 cns11643l1_to_ucs4 (const unsigned char **s, size_t avail,
29 unsigned char offset)
30 {
31 unsigned char ch = *(*s);
32 unsigned char ch2;
33 int idx;
34
35 if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) > 0x7d)
36 return __UNKNOWN_10646_CHAR;
37
38 if (avail < 2)
39 return 0;
40
41 ch2 = (*s)[1];
42 if ((ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
43 return __UNKNOWN_10646_CHAR;
44
45 idx = (ch - 0x21 - offset) * 94 + (ch2 - 0x21 - offset);
46 if (idx > 0x21f2)
47 return __UNKNOWN_10646_CHAR;
48
49 (*s) += 2;
50
51 return __cns11643l1_to_ucs4_tab[idx] ?: ((*s) -= 2, __UNKNOWN_10646_CHAR);
52 }
53
54
55 /* Tables for the UCS4 -> CNS conversion. */
56 extern const char __cns11643l1_from_ucs4_tab1[][2];
57 extern const char __cns11643l1_from_ucs4_tab2[][2];
58 extern const char __cns11643l1_from_ucs4_tab3[][2];
59 extern const char __cns11643l1_from_ucs4_tab4[][2];
60 extern const char __cns11643l1_from_ucs4_tab5[][2];
61 extern const char __cns11643l1_from_ucs4_tab6[][2];
62 extern const char __cns11643l1_from_ucs4_tab7[][2];
63 extern const char __cns11643l1_from_ucs4_tab8[][2];
64 extern const char __cns11643l1_from_ucs4_tab9[][2];
65 extern const char __cns11643l1_from_ucs4_tab10[][2];
66 extern const char __cns11643l1_from_ucs4_tab11[][2];
67 extern const char __cns11643l1_from_ucs4_tab12[][2];
68 extern const char __cns11643l1_from_ucs4_tab13[][2];
69 extern const char __cns11643l1_from_ucs4_tab14[][2];
70
71
size_t(always_inline)72 static inline size_t
73 __attribute ((always_inline))
74 ucs4_to_cns11643l1 (uint32_t wch, unsigned char *s, size_t avail)
75 {
76 unsigned int ch = (unsigned int) wch;
77 char buf[2];
78 const char *cp = buf;
79
80 switch (ch)
81 {
82 case 0xa7 ... 0xf7:
83 cp = __cns11643l1_from_ucs4_tab1[ch - 0xa7];
84 break;
85 case 0x2c7 ... 0x2d9:
86 cp = __cns11643l1_from_ucs4_tab2[ch - 0x2c7];
87 break;
88 case 0x391 ... 0x3c9:
89 cp = __cns11643l1_from_ucs4_tab3[ch - 0x391];
90 break;
91 case 0x2013 ... 0x203e:
92 cp = __cns11643l1_from_ucs4_tab4[ch - 0x2013];
93 break;
94 case 0x2103:
95 cp = "\x22\x6a";
96 break;
97 case 0x2105:
98 cp = "\x22\x22";
99 break;
100 case 0x2109:
101 cp = "\x22\x6b";
102 break;
103 case 0x2160 ... 0x2169:
104 buf[0] = '\x24';
105 buf[1] = '\x2b' + (ch - 0x2160);
106 break;
107 case 0x2170 ... 0x2179:
108 buf[0] = '\x26';
109 buf[1] = '\x35' + (ch - 0x2170);
110 break;
111 case 0x2190 ... 0x2199:
112 cp = __cns11643l1_from_ucs4_tab5[ch - 0x2190];
113 break;
114 case 0x2215 ... 0x2267:
115 cp = __cns11643l1_from_ucs4_tab6[ch - 0x2215];
116 break;
117 case 0x22a5:
118 cp = "\x22\x47";
119 break;
120 case 0x22bf:
121 cp = "\x22\x4a";
122 break;
123 case 0x2400 ... 0x2421:
124 cp = __cns11643l1_from_ucs4_tab7[ch - 0x2400];
125 break;
126 case 0x2460 ... 0x247d:
127 cp = __cns11643l1_from_ucs4_tab8[ch - 0x2460];
128 break;
129 case 0x2500 ... 0x2642:
130 cp = __cns11643l1_from_ucs4_tab9[ch - 0x2500];
131 break;
132 case 0x3000 ... 0x3029:
133 cp = __cns11643l1_from_ucs4_tab10[ch - 0x3000];
134 break;
135 case 0x30fb:
136 cp = "\x21\x26";
137 break;
138 case 0x3105 ... 0x3129:
139 buf[0] = '\x25';
140 buf[1] = '\x47' + (ch - 0x3105);
141 break;
142 case 0x32a3:
143 cp = "\x22\x21";
144 break;
145 case 0x338e ... 0x33d5:
146 cp = __cns11643l1_from_ucs4_tab11[ch - 0x338e];
147 break;
148 case 0x4e00 ... 0x9f9c:
149 cp = __cns11643l1_from_ucs4_tab12[ch - 0x4e00];
150 break;
151 case 0xfe30 ... 0xfe6b:
152 cp = __cns11643l1_from_ucs4_tab13[ch - 0xfe30];
153 break;
154 case 0xff01 ... 0xff5d:
155 cp = __cns11643l1_from_ucs4_tab14[ch - 0xff01];
156 break;
157 case 0xffe0:
158 cp = "\x22\x66";
159 break;
160 case 0xffe1:
161 cp = "\x22\x67";
162 break;
163 case 0xffe5:
164 cp = "\x22\x64";
165 break;
166 default:
167 return __UNKNOWN_10646_CHAR;
168 }
169
170 if (cp[0] == '\0')
171 return __UNKNOWN_10646_CHAR;
172
173 if (avail < 2)
174 return 0;
175
176 s[0] = cp[0];
177 s[1] = cp[1];
178
179 return 2;
180 }
181