1 /* Copyright (C) 2002-2023 Free Software Foundation, Inc.
2 This file is part of the GNU C Library.
3
4 The GNU C Library is free software; you can redistribute it and/or
5 modify it under the terms of the GNU Lesser General Public
6 License as published by the Free Software Foundation; either
7 version 2.1 of the License, or (at your option) any later version.
8
9 The GNU C Library is distributed in the hope that it will be useful,
10 but WITHOUT ANY WARRANTY; without even the implied warranty of
11 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12 Lesser General Public License for more details.
13
14 You should have received a copy of the GNU Lesser General Public
15 License along with the GNU C Library; if not, see
16 <https://www.gnu.org/licenses/>. */
17
18 #include <assert.h>
19 #include <ctype.h>
20 #include <string.h>
21 #include "wcsmbsload.h"
22 #include <dlfcn.h>
23 #include <errno.h>
24 #include <gconv.h>
25 #include <stdlib.h>
26 #include <string.h>
27 #include <wchar.h>
28 #include <wcsmbsload.h>
29
30 #include <pointer_guard.h>
31
32 #ifndef EILSEQ
33 # define EILSEQ EINVAL
34 #endif
35
36
37 size_t
38 attribute_hidden
39 __mbsrtowcs_l (wchar_t *dst, const char **src, size_t len, mbstate_t *ps,
40 locale_t l)
41 {
42 struct __gconv_step_data data;
43 size_t result;
44 int status;
45 struct __gconv_step *towc;
46 size_t non_reversible;
47 const struct gconv_fcts *fcts;
48
49 /* Tell where we want the result. */
50 data.__invocation_counter = 0;
51 data.__internal_use = 1;
52 data.__flags = __GCONV_IS_LAST;
53 data.__statep = ps;
54
55 /* Get the conversion functions. */
56 fcts = get_gconv_fcts (l->__locales[LC_CTYPE]);
57
58 /* Get the structure with the function pointers. */
59 towc = fcts->towc;
60 __gconv_fct fct = towc->__fct;
61 if (towc->__shlib_handle != NULL)
62 PTR_DEMANGLE (fct);
63
64 /* We have to handle DST == NULL special. */
65 if (dst == NULL)
66 {
67 mbstate_t temp_state;
68 wchar_t buf[64]; /* Just an arbitrary size. */
69 const unsigned char *inbuf = (const unsigned char *) *src;
70 const unsigned char *srcend = inbuf + strlen (*src) + 1;
71
72 temp_state = *data.__statep;
73 data.__statep = &temp_state;
74
75 result = 0;
76 data.__outbufend = (unsigned char *) buf + sizeof (buf);
77 do
78 {
79 data.__outbuf = (unsigned char *) buf;
80
81 status = DL_CALL_FCT (fct, (towc, &data, &inbuf, srcend, NULL,
82 &non_reversible, 0, 1));
83
84 result += (wchar_t *) data.__outbuf - buf;
85 }
86 while (status == __GCONV_FULL_OUTPUT);
87
88 if (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
89 {
90 /* There better should be a NUL wide char at the end. */
91 assert (((wchar_t *) data.__outbuf)[-1] == L'\0');
92 /* Don't count the NUL character in. */
93 --result;
94 }
95 }
96 else
97 {
98 /* This code is based on the safe assumption that all internal
99 multi-byte encodings use the NUL byte only to mark the end
100 of the string. */
101 const unsigned char *srcp = (const unsigned char *) *src;
102 const unsigned char *srcend;
103
104 data.__outbuf = (unsigned char *) dst;
105 data.__outbufend = data.__outbuf + len * sizeof (wchar_t);
106
107 status = __GCONV_FULL_OUTPUT;
108
109 while (len > 0)
110 {
111 /* Pessimistic guess as to how much input we can use. In the
112 worst case we need one input byte for one output wchar_t. */
113 srcend = srcp + __strnlen ((const char *) srcp, len) + 1;
114
115 status = DL_CALL_FCT (fct, (towc, &data, &srcp, srcend, NULL,
116 &non_reversible, 0, 1));
117 if ((status != __GCONV_EMPTY_INPUT
118 && status != __GCONV_INCOMPLETE_INPUT)
119 /* Not all input read. */
120 || srcp != srcend
121 /* Reached the end of the input. */
122 || srcend[-1] == '\0')
123 break;
124
125 len = (wchar_t *) data.__outbufend - (wchar_t *) data.__outbuf;
126 }
127
128 /* Make the end if the input known to the caller. */
129 *src = (const char *) srcp;
130
131 result = (wchar_t *) data.__outbuf - dst;
132
133 /* We have to determine whether the last character converted
134 is the NUL character. */
135 if ((status == __GCONV_OK || status == __GCONV_EMPTY_INPUT)
136 && ((wchar_t *) dst)[result - 1] == L'\0')
137 {
138 assert (result > 0);
139 assert (__mbsinit (data.__statep));
140 *src = NULL;
141 --result;
142 }
143 }
144
145 /* There must not be any problems with the conversion but illegal input
146 characters. */
147 assert (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT
148 || status == __GCONV_ILLEGAL_INPUT
149 || status == __GCONV_INCOMPLETE_INPUT
150 || status == __GCONV_FULL_OUTPUT);
151
152 if (status != __GCONV_OK && status != __GCONV_FULL_OUTPUT
153 && status != __GCONV_EMPTY_INPUT && status != __GCONV_INCOMPLETE_INPUT)
154 {
155 result = (size_t) -1;
156 __set_errno (EILSEQ);
157 }
158
159 return result;
160 }