1 | /* Copyright (C) 2002-2021 Free Software Foundation, Inc. |
2 | This file is part of the GNU C Library. |
3 | Contributed by Ulrich Drepper <drepper@gnu.org>, 2002. |
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 <assert.h> |
20 | #include <ctype.h> |
21 | #include <string.h> |
22 | #include "wcsmbsload.h" |
23 | #include <dlfcn.h> |
24 | #include <errno.h> |
25 | #include <gconv.h> |
26 | #include <stdlib.h> |
27 | #include <string.h> |
28 | #include <wchar.h> |
29 | #include <wcsmbsload.h> |
30 | |
31 | #include <sysdep.h> |
32 | |
33 | #ifndef EILSEQ |
34 | # define EILSEQ EINVAL |
35 | #endif |
36 | |
37 | |
38 | size_t |
39 | attribute_hidden |
40 | __mbsrtowcs_l (wchar_t *dst, const char **src, size_t len, mbstate_t *ps, |
41 | locale_t l) |
42 | { |
43 | struct __gconv_step_data data; |
44 | size_t result; |
45 | int status; |
46 | struct __gconv_step *towc; |
47 | size_t non_reversible; |
48 | const struct gconv_fcts *fcts; |
49 | |
50 | /* Tell where we want the result. */ |
51 | data.__invocation_counter = 0; |
52 | data.__internal_use = 1; |
53 | data.__flags = __GCONV_IS_LAST; |
54 | data.__statep = ps; |
55 | |
56 | /* Get the conversion functions. */ |
57 | fcts = get_gconv_fcts (l->__locales[LC_CTYPE]); |
58 | |
59 | /* Get the structure with the function pointers. */ |
60 | towc = fcts->towc; |
61 | __gconv_fct fct = towc->__fct; |
62 | #ifdef PTR_DEMANGLE |
63 | if (towc->__shlib_handle != NULL) |
64 | PTR_DEMANGLE (fct); |
65 | #endif |
66 | |
67 | /* We have to handle DST == NULL special. */ |
68 | if (dst == NULL) |
69 | { |
70 | mbstate_t temp_state; |
71 | wchar_t buf[64]; /* Just an arbitrary size. */ |
72 | const unsigned char *inbuf = (const unsigned char *) *src; |
73 | const unsigned char *srcend = inbuf + strlen (*src) + 1; |
74 | |
75 | temp_state = *data.__statep; |
76 | data.__statep = &temp_state; |
77 | |
78 | result = 0; |
79 | data.__outbufend = (unsigned char *) buf + sizeof (buf); |
80 | do |
81 | { |
82 | data.__outbuf = (unsigned char *) buf; |
83 | |
84 | status = DL_CALL_FCT (fct, (towc, &data, &inbuf, srcend, NULL, |
85 | &non_reversible, 0, 1)); |
86 | |
87 | result += (wchar_t *) data.__outbuf - buf; |
88 | } |
89 | while (status == __GCONV_FULL_OUTPUT); |
90 | |
91 | if (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT) |
92 | { |
93 | /* There better should be a NUL wide char at the end. */ |
94 | assert (((wchar_t *) data.__outbuf)[-1] == L'\0'); |
95 | /* Don't count the NUL character in. */ |
96 | --result; |
97 | } |
98 | } |
99 | else |
100 | { |
101 | /* This code is based on the safe assumption that all internal |
102 | multi-byte encodings use the NUL byte only to mark the end |
103 | of the string. */ |
104 | const unsigned char *srcp = (const unsigned char *) *src; |
105 | const unsigned char *srcend; |
106 | |
107 | data.__outbuf = (unsigned char *) dst; |
108 | data.__outbufend = data.__outbuf + len * sizeof (wchar_t); |
109 | |
110 | status = __GCONV_FULL_OUTPUT; |
111 | |
112 | while (len > 0) |
113 | { |
114 | /* Pessimistic guess as to how much input we can use. In the |
115 | worst case we need one input byte for one output wchar_t. */ |
116 | srcend = srcp + __strnlen ((const char *) srcp, len) + 1; |
117 | |
118 | status = DL_CALL_FCT (fct, (towc, &data, &srcp, srcend, NULL, |
119 | &non_reversible, 0, 1)); |
120 | if ((status != __GCONV_EMPTY_INPUT |
121 | && status != __GCONV_INCOMPLETE_INPUT) |
122 | /* Not all input read. */ |
123 | || srcp != srcend |
124 | /* Reached the end of the input. */ |
125 | || srcend[-1] == '\0') |
126 | break; |
127 | |
128 | len = (wchar_t *) data.__outbufend - (wchar_t *) data.__outbuf; |
129 | } |
130 | |
131 | /* Make the end if the input known to the caller. */ |
132 | *src = (const char *) srcp; |
133 | |
134 | result = (wchar_t *) data.__outbuf - dst; |
135 | |
136 | /* We have to determine whether the last character converted |
137 | is the NUL character. */ |
138 | if ((status == __GCONV_OK || status == __GCONV_EMPTY_INPUT) |
139 | && ((wchar_t *) dst)[result - 1] == L'\0') |
140 | { |
141 | assert (result > 0); |
142 | assert (__mbsinit (data.__statep)); |
143 | *src = NULL; |
144 | --result; |
145 | } |
146 | } |
147 | |
148 | /* There must not be any problems with the conversion but illegal input |
149 | characters. */ |
150 | assert (status == __GCONV_OK || status == __GCONV_EMPTY_INPUT |
151 | || status == __GCONV_ILLEGAL_INPUT |
152 | || status == __GCONV_INCOMPLETE_INPUT |
153 | || status == __GCONV_FULL_OUTPUT); |
154 | |
155 | if (status != __GCONV_OK && status != __GCONV_FULL_OUTPUT |
156 | && status != __GCONV_EMPTY_INPUT && status != __GCONV_INCOMPLETE_INPUT) |
157 | { |
158 | result = (size_t) -1; |
159 | __set_errno (EILSEQ); |
160 | } |
161 | |
162 | return result; |
163 | } |
164 | |