1 /* Access functions for GB2312 conversion.
2 Copyright (C) 1998-2023 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 #ifndef _GB2312_H
20 #define _GB2312_H 1
21
22 #include <gconv.h>
23 #include <stdint.h>
24 #include <assert.h>
25
26 /* Conversion table. */
27 extern const uint16_t __gb2312_to_ucs[];
28
29
30 static inline uint32_t
31 __attribute ((always_inline))
32 gb2312_to_ucs4 (const unsigned char **s, size_t avail, unsigned char offset)
33 {
34 unsigned char ch = *(*s);
35 unsigned char ch2;
36 int idx;
37
38 if (ch < offset || (ch - offset) <= 0x20 || (ch - offset) > 0x77)
39 return __UNKNOWN_10646_CHAR;
40
41 if (avail < 2)
42 return 0;
43
44 ch2 = (*s)[1];
45 if ((ch2 - offset) <= 0x20 || (ch2 - offset) >= 0x7f)
46 return __UNKNOWN_10646_CHAR;
47
48 idx = (ch - 0x21 - offset) * 94 + (ch2 - 0x21 - offset);
49 if (idx > 0x1ff1)
50 return __UNKNOWN_10646_CHAR;
51
52 (*s) += 2;
53
54 return __gb2312_to_ucs[idx] ?: ((*s) -= 2, __UNKNOWN_10646_CHAR);
55 }
56
57
58 extern const char __gb2312_from_ucs4_tab1[][2];
59 extern const char __gb2312_from_ucs4_tab2[][2];
60 extern const char __gb2312_from_ucs4_tab3[][2];
61 extern const char __gb2312_from_ucs4_tab4[][2];
62 extern const char __gb2312_from_ucs4_tab5[][2];
63 extern const char __gb2312_from_ucs4_tab6[][2];
64 extern const char __gb2312_from_ucs4_tab7[][2];
65 extern const char __gb2312_from_ucs4_tab8[][2];
66 extern const char __gb2312_from_ucs4_tab9[][2];
67
68 static inline size_t
69 __attribute ((always_inline))
70 ucs4_to_gb2312 (uint32_t wch, unsigned char *s, size_t avail)
71 {
72 unsigned int ch = (unsigned int) wch;
73 char buf[2];
74 const char *cp = buf;
75
76 switch (ch)
77 {
78 case 0xa4 ... 0x101:
79 cp = __gb2312_from_ucs4_tab1[ch - 0xa4];
80 break;
81 case 0x113:
82 cp = "\x28\x25";
83 break;
84 case 0x11b:
85 cp = "\x28\x27";
86 break;
87 case 0x12b:
88 cp = "\x28\x29";
89 break;
90 case 0x14d:
91 cp = "\x28\x2d";
92 break;
93 case 0x16b:
94 cp = "\x28\x31";
95 break;
96 case 0x1ce:
97 cp = "\x28\x23";
98 break;
99 case 0x1d0:
100 cp = "\x28\x2b";
101 break;
102 case 0x1d2:
103 cp = "\x28\x2f";
104 break;
105 case 0x1d4:
106 cp = "\x28\x33";
107 break;
108 case 0x1d6:
109 cp = "\x28\x35";
110 break;
111 case 0x1d8:
112 cp = "\x28\x36";
113 break;
114 case 0x1da:
115 cp = "\x28\x37";
116 break;
117 case 0x1dc:
118 cp = "\x28\x38";
119 break;
120 case 0x2c7:
121 cp = "\x21\x26";
122 break;
123 case 0x2c9:
124 cp = "\x21\x25";
125 break;
126 case 0x391 ... 0x3c9:
127 cp = __gb2312_from_ucs4_tab2[ch - 0x391];
128 break;
129 case 0x401 ... 0x451:
130 cp = __gb2312_from_ucs4_tab3[ch - 0x401];
131 break;
132 case 0x2015 ... 0x203b:
133 cp = __gb2312_from_ucs4_tab4[ch - 0x2015];
134 break;
135 case 0x2103 ... 0x22a5:
136 cp = __gb2312_from_ucs4_tab5[ch - 0x2103];
137 break;
138 case 0x2312:
139 cp = "\x21\x50";
140 break;
141 case 0x2460 ... 0x249b:
142 cp = __gb2312_from_ucs4_tab6[ch - 0x2460];
143 break;
144 case 0x2500 ... 0x254b:
145 buf[0] = '\x29';
146 buf[1] = '\x24' + (ch % 256);
147 break;
148 case 0x25a0:
149 cp = "\x21\x76";
150 break;
151 case 0x25a1:
152 cp = "\x21\x75";
153 break;
154 case 0x25b2:
155 cp = "\x21\x78";
156 break;
157 case 0x25b3:
158 cp = "\x21\x77";
159 break;
160 case 0x25c6:
161 cp = "\x21\x74";
162 break;
163 case 0x25c7:
164 cp = "\x21\x73";
165 break;
166 case 0x25cb:
167 cp = "\x21\x70";
168 break;
169 case 0x25ce:
170 cp = "\x21\x72";
171 break;
172 case 0x25cf:
173 cp = "\x21\x71";
174 break;
175 case 0x2605:
176 cp = "\x21\x6f";
177 break;
178 case 0x2606:
179 cp = "\x21\x6e";
180 break;
181 case 0x2640:
182 cp = "\x21\x62";
183 break;
184 case 0x2642:
185 cp = "\x21\x61";
186 break;
187 case 0x3000 ... 0x3129:
188 cp = __gb2312_from_ucs4_tab7[ch - 0x3000];
189 break;
190 case 0x3220 ... 0x3229:
191 buf[0] = '\x22';
192 buf[1] = '\x65' + (ch - 0x3220);
193 break;
194 case 0x4e00 ... 0x9fa0:
195 cp = __gb2312_from_ucs4_tab8[ch - 0x4e00];
196 break;
197 case 0xff01 ... 0xff5e:
198 cp = __gb2312_from_ucs4_tab9[ch - 0xff01];
199 break;
200 case 0xffe0:
201 cp = "\x21\x69";
202 break;
203 case 0xffe1:
204 cp = "\x21\x6a";
205 break;
206 case 0xffe3:
207 cp = "\x23\x7e";
208 break;
209 case 0xffe5:
210 cp = "\x23\x24";
211 break;
212 default:
213 return __UNKNOWN_10646_CHAR;
214 }
215
216 if (cp[0] == '\0')
217 return __UNKNOWN_10646_CHAR;
218
219 assert (cp[1] != '\0');
220
221 if (avail < 2)
222 return 0;
223
224 s[0] = cp[0];
225 s[1] = cp[1];
226
227 return 2;
228 }
229
230 #endif /* gb2312.h */