1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
|
#include <iconv.h>
#include <errno.h>
#include <wchar.h>
#include <string.h>
#include <stdlib.h>
#include <limits.h>
#include <stdint.h>
#define UTF_32BE 0300
#define UTF_16LE 0301
#define UTF_16BE 0302
#define UTF_32LE 0303
#define UCS2BE 0304
#define UCS2LE 0305
#define US_ASCII 0306
#define WCHAR_T 0307
#define UTF_8 0310
/* FIXME: these are not implemented yet
* EUC: A1-FE A1-FE
* GBK: 81-FE 40-7E,80-FE
* Big5: A1-FE 40-7E,A1-FE
*/
/* Definitions of charmaps. Each charmap consists of:
* 1. Empty-string-terminated list of null-terminated aliases.
* 2. Special type code or number of elided entries.
* 3. Character table (size determined by field 2). */
static const unsigned char charmaps[] =
"utf8\0\0\310"
"wchart\0\0\307"
"ucs2\0ucs2be\0\0\304"
"ucs2le\0\0\305"
"utf16\0utf16be\0\0\302"
"utf16le\0\0\301"
"ucs4\0ucs4be\0utf32\0utf32be\0\0\300"
"ucs4le\0utf32le\0\0\303"
"ascii\0usascii\0iso646\0iso646us\0\0\306"
#include "codepages.h"
;
static const unsigned short legacy_chars[] = {
#include "legacychars.h"
};
static int fuzzycmp(const unsigned char *a, const unsigned char *b)
{
for (; *a && *b; a++, b++) {
while (*a && (*a|32U)-'a'>26 && *a-'0'>10U) a++;
if ((*a|32U) != *b) return 1;
}
return *a != *b;
}
static size_t find_charmap(const void *name)
{
const unsigned char *s;
for (s=charmaps; *s; ) {
if (!fuzzycmp(name, s)) {
for (; *s; s+=strlen((void *)s)+1);
return s+1-charmaps;
}
s += strlen((void *)s)+1;
if (!*s) {
if (s[1] > 0200) s+=2;
else s+=2+(128U-s[1])/4*5;
}
}
return -1;
}
iconv_t iconv_open(const char *to, const char *from)
{
size_t f, t;
if ((t = find_charmap(to))==-1 || (f = find_charmap(from))==-1) {
errno = EINVAL;
return (iconv_t)-1;
}
return (void *)(f<<16 | t);
}
int iconv_close(iconv_t cd)
{
return 0;
}
static unsigned get_16(const unsigned char *s, int e)
{
e &= 1;
return s[e]<<8 | s[1-e];
}
static void put_16(unsigned char *s, unsigned c, int e)
{
e &= 1;
s[e] = c>>8;
s[1-e] = c;
}
static unsigned get_32(const unsigned char *s, int e)
{
e &= 3;
return s[e]+0U<<24 | s[e^1]<<16 | s[e^2]<<8 | s[e^3];
}
static void put_32(unsigned char *s, unsigned c, int e)
{
e &= 3;
s[e^0] = c>>24;
s[e^1] = c>>16;
s[e^2] = c>>8;
s[e^3] = c;
}
/* Adapt as needed */
#define mbrtowc_utf8 mbrtowc
#define wctomb_utf8 wctomb
#include <stdio.h>
size_t iconv(iconv_t cd0, char **in, size_t *inb, char **out, size_t *outb)
{
size_t x=0;
unsigned long cd = (unsigned long)cd0;
unsigned to = cd & 0xffff;
unsigned from = cd >> 16;
const unsigned char *map = charmaps+from+1;
const unsigned char *tomap = charmaps+to+1;
mbstate_t st = {0};
wchar_t wc;
unsigned c, d;
size_t k, l;
int err;
unsigned char type = map[-1];
unsigned char totype = tomap[-1];
if (!in || !*in || !*inb) return 0;
for (; *inb; *in+=l, *inb-=l) {
c = *(unsigned char *)*in;
l = 1;
if (c >= 128) switch (type) {
case UTF_8:
l = mbrtowc_utf8(&wc, *in, *inb, &st);
if (!l) l++;
else if (l == (size_t)-1) goto ilseq;
else if (l == (size_t)-2) goto starved;
c = wc;
break;
case US_ASCII:
goto ilseq;
case WCHAR_T:
l = sizeof(wchar_t);
if (*inb < l) goto starved;
c = *(wchar_t *)*in;
if (0) {
case UTF_32BE:
case UTF_32LE:
l = 4;
if (*inb < 4) goto starved;
c = get_32((void *)*in, type);
}
if (c-0xd800u < 0x800u || c >= 0x110000u) goto ilseq;
break;
case UCS2BE:
case UCS2LE:
case UTF_16BE:
case UTF_16LE:
l = 2;
if (*inb < 2) goto starved;
c = get_16((void *)*in, type);
if ((unsigned)(c-0xdc00) < 0x400) goto ilseq;
if ((unsigned)(c-0xd800) < 0x400) {
if (type-UCS2BE < 2U) goto ilseq;
l = 4;
if (*inb < 4) goto starved;
d = get_16((void *)(*in + 2), from);
if ((unsigned)(c-0xdc00) >= 0x400) goto ilseq;
c = ((c-0xd800)<<10) | (d-0xdc00);
}
break;
default:
if (c < 128+type) break;
c -= 128+type;
c = legacy_chars[ map[c*5/4]>>2*c%8 |
map[c*5/4+1]<<8-2*c%8 & 1023 ];
if (!c) c = *(unsigned char *)*in;
if (c==1) goto ilseq;
}
switch (totype) {
case WCHAR_T:
if (*outb < sizeof(wchar_t)) goto toobig;
*(wchar_t *)*out = c;
*out += sizeof(wchar_t);
*outb -= sizeof(wchar_t);
break;
case UTF_8:
if (*outb < 4) {
char tmp[4];
k = wctomb_utf8(tmp, c);
if (*outb < k) goto toobig;
memcpy(*out, tmp, k);
} else k = wctomb_utf8(*out, c);
*out += k;
*outb -= k;
break;
case US_ASCII:
if (c > 0x7f) subst: x++, c='*';
default:
if (*outb < 1) goto toobig;
if (c < 128+totype) {
revout:
*(*out)++ = c;
*outb -= 1;
break;
}
d = c;
for (c=0; c<128-totype; c++) {
if (d == legacy_chars[ map[c*5/4]>>2*c%8 |
map[c*5/4+1]<<8-2*c%8 & 1023 ]) {
c += 128;
goto revout;
}
}
goto subst;
case UCS2BE:
case UCS2LE:
case UTF_16BE:
case UTF_16LE:
if (c < 0x10000) {
if (*outb < 2) goto toobig;
put_16((void *)*out, c, totype);
*out += 2;
*outb -= 2;
break;
}
if (type-UCS2BE < 2U) goto ilseq;
if (*outb < 4) goto toobig;
put_16((void *)*out, (c>>10)|0xd800, totype);
put_16((void *)(*out + 2), (c&0x3ff)|0xdc00, totype);
*out += 4;
*outb -= 4;
break;
case UTF_32BE:
case UTF_32LE:
if (*outb < 4) goto toobig;
put_32((void *)*out, c, totype);
*out += 4;
*outb -= 4;
break;
}
}
return x;
ilseq:
err = EILSEQ;
x = -1;
goto end;
toobig:
err = E2BIG;
x = -1;
goto end;
starved:
err = EINVAL;
x = -1;
end:
errno = err;
return x;
}
|