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
|
/* Copyright (c) 2002-2018 Dovecot authors, see the included COPYING file */
#include "lib.h"
#include "buffer.h"
#include "charset-utf8-private.h"
#ifdef HAVE_ICONV
#include <iconv.h>
#include <ctype.h>
struct charset_translation {
iconv_t cd;
normalizer_func_t *normalizer;
};
static int
iconv_charset_to_utf8_begin(const char *charset, normalizer_func_t *normalizer,
struct charset_translation **t_r)
{
struct charset_translation *t;
iconv_t cd;
if (charset_is_utf8(charset))
cd = (iconv_t)-1;
else {
if (strcmp(charset, "UTF-8//TEST") == 0)
charset = "UTF-8";
cd = iconv_open("UTF-8", charset);
if (cd == (iconv_t)-1)
return -1;
}
t = i_new(struct charset_translation, 1);
t->cd = cd;
t->normalizer = normalizer;
*t_r = t;
return 0;
}
static void iconv_charset_to_utf8_end(struct charset_translation *t)
{
if (t->cd != (iconv_t)-1)
iconv_close(t->cd);
i_free(t);
}
static void iconv_charset_to_utf8_reset(struct charset_translation *t)
{
if (t->cd != (iconv_t)-1)
(void)iconv(t->cd, NULL, NULL, NULL, NULL);
}
static bool
charset_to_utf8_try(struct charset_translation *t,
const unsigned char *src, size_t *src_size, buffer_t *dest,
enum charset_result *result)
{
ICONV_CONST char *ic_srcbuf;
char tmpbuf[8192], *ic_destbuf;
size_t srcleft, destleft, tmpbuf_used;
bool ret = TRUE;
if (t->cd == (iconv_t)-1) {
/* input is already supposed to be UTF-8 */
*result = charset_utf8_to_utf8(t->normalizer, src, src_size, dest);
return TRUE;
}
destleft = sizeof(tmpbuf);
ic_destbuf = tmpbuf;
srcleft = *src_size;
ic_srcbuf = (ICONV_CONST char *) src;
if (iconv(t->cd, &ic_srcbuf, &srcleft,
&ic_destbuf, &destleft) != SIZE_MAX) {
i_assert(srcleft == 0);
*result = CHARSET_RET_OK;
} else if (errno == E2BIG) {
/* set result just to avoid compiler warning */
*result = CHARSET_RET_INCOMPLETE_INPUT;
ret = FALSE;
} else if (errno == EINVAL) {
i_assert(srcleft <= CHARSET_MAX_PENDING_BUF_SIZE);
*result = CHARSET_RET_INCOMPLETE_INPUT;
} else {
/* should be EILSEQ */
*result = CHARSET_RET_INVALID_INPUT;
ret = FALSE;
}
*src_size -= srcleft;
/* we just converted data to UTF-8. it shouldn't be invalid, but
Solaris iconv appears to pass invalid data through sometimes
(e.g. 8 bit characters with UTF-7) */
tmpbuf_used = sizeof(tmpbuf) - destleft;
if (charset_utf8_to_utf8(t->normalizer, (void *)tmpbuf,
&tmpbuf_used, dest) != CHARSET_RET_OK)
*result = CHARSET_RET_INVALID_INPUT;
return ret;
}
static enum charset_result
iconv_charset_to_utf8(struct charset_translation *t,
const unsigned char *src, size_t *src_size,
buffer_t *dest)
{
enum charset_result result;
size_t pos, size;
size_t prev_invalid_pos = SIZE_MAX;
bool ret;
for (pos = 0;;) {
i_assert(pos <= *src_size);
size = *src_size - pos;
ret = charset_to_utf8_try(t, src + pos, &size, dest, &result);
pos += size;
if (ret)
break;
if (result == CHARSET_RET_INVALID_INPUT) {
if (prev_invalid_pos != dest->used) {
buffer_append(dest, UNICODE_REPLACEMENT_CHAR_UTF8,
strlen(UNICODE_REPLACEMENT_CHAR_UTF8));
prev_invalid_pos = dest->used;
}
if (pos < *src_size)
pos++;
}
}
if (prev_invalid_pos != SIZE_MAX)
result = CHARSET_RET_INVALID_INPUT;
i_assert(*src_size - pos <= CHARSET_MAX_PENDING_BUF_SIZE);
*src_size = pos;
return result;
}
const struct charset_utf8_vfuncs charset_iconv = {
.to_utf8_begin = iconv_charset_to_utf8_begin,
.to_utf8_end = iconv_charset_to_utf8_end,
.to_utf8_reset = iconv_charset_to_utf8_reset,
.to_utf8 = iconv_charset_to_utf8,
};
#endif
|