1 | /* Determine the number of screen columns needed for a string. |
2 | Copyright (C) 2000-2019 Free Software Foundation, Inc. |
3 | |
4 | This program is free software: you can redistribute it and/or modify |
5 | it under the terms of the GNU General Public License as published by |
6 | the Free Software Foundation; either version 3 of the License, or |
7 | (at your option) any later version. |
8 | |
9 | This program 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 |
12 | GNU General Public License for more details. |
13 | |
14 | You should have received a copy of the GNU General Public License |
15 | along with this program. If not, see <https://www.gnu.org/licenses/>. */ |
16 | |
17 | /* Written by Bruno Haible <haible@clisp.cons.org>. */ |
18 | |
19 | #include <config.h> |
20 | |
21 | /* Specification. */ |
22 | #include "mbswidth.h" |
23 | |
24 | /* Get MB_CUR_MAX. */ |
25 | #include <stdlib.h> |
26 | |
27 | #include <string.h> |
28 | |
29 | /* Get isprint(). */ |
30 | #include <ctype.h> |
31 | |
32 | /* Get mbstate_t, mbrtowc(), mbsinit(), wcwidth(). */ |
33 | #include <wchar.h> |
34 | |
35 | /* Get iswcntrl(). */ |
36 | #include <wctype.h> |
37 | |
38 | /* Get INT_MAX. */ |
39 | #include <limits.h> |
40 | |
41 | /* Returns the number of columns needed to represent the multibyte |
42 | character string pointed to by STRING. If a non-printable character |
43 | occurs, and MBSW_REJECT_UNPRINTABLE is specified, -1 is returned. |
44 | With flags = MBSW_REJECT_INVALID | MBSW_REJECT_UNPRINTABLE, this is |
45 | the multibyte analogue of the wcswidth function. */ |
46 | int |
47 | mbswidth (const char *string, int flags) |
48 | { |
49 | return mbsnwidth (string, strlen (string), flags); |
50 | } |
51 | |
52 | /* Returns the number of columns needed to represent the multibyte |
53 | character string pointed to by STRING of length NBYTES. If a |
54 | non-printable character occurs, and MBSW_REJECT_UNPRINTABLE is |
55 | specified, -1 is returned. */ |
56 | int |
57 | mbsnwidth (const char *string, size_t nbytes, int flags) |
58 | { |
59 | const char *p = string; |
60 | const char *plimit = p + nbytes; |
61 | int width; |
62 | |
63 | width = 0; |
64 | if (MB_CUR_MAX > 1) |
65 | { |
66 | while (p < plimit) |
67 | switch (*p) |
68 | { |
69 | case ' ': case '!': case '"': case '#': case '%': |
70 | case '&': case '\'': case '(': case ')': case '*': |
71 | case '+': case ',': case '-': case '.': case '/': |
72 | case '0': case '1': case '2': case '3': case '4': |
73 | case '5': case '6': case '7': case '8': case '9': |
74 | case ':': case ';': case '<': case '=': case '>': |
75 | case '?': |
76 | case 'A': case 'B': case 'C': case 'D': case 'E': |
77 | case 'F': case 'G': case 'H': case 'I': case 'J': |
78 | case 'K': case 'L': case 'M': case 'N': case 'O': |
79 | case 'P': case 'Q': case 'R': case 'S': case 'T': |
80 | case 'U': case 'V': case 'W': case 'X': case 'Y': |
81 | case 'Z': |
82 | case '[': case '\\': case ']': case '^': case '_': |
83 | case 'a': case 'b': case 'c': case 'd': case 'e': |
84 | case 'f': case 'g': case 'h': case 'i': case 'j': |
85 | case 'k': case 'l': case 'm': case 'n': case 'o': |
86 | case 'p': case 'q': case 'r': case 's': case 't': |
87 | case 'u': case 'v': case 'w': case 'x': case 'y': |
88 | case 'z': case '{': case '|': case '}': case '~': |
89 | /* These characters are printable ASCII characters. */ |
90 | p++; |
91 | width++; |
92 | break; |
93 | default: |
94 | /* If we have a multibyte sequence, scan it up to its end. */ |
95 | { |
96 | mbstate_t mbstate; |
97 | memset (&mbstate, 0, sizeof mbstate); |
98 | do |
99 | { |
100 | wchar_t wc; |
101 | size_t bytes; |
102 | int w; |
103 | |
104 | bytes = mbrtowc (&wc, p, plimit - p, &mbstate); |
105 | |
106 | if (bytes == (size_t) -1) |
107 | /* An invalid multibyte sequence was encountered. */ |
108 | { |
109 | if (!(flags & MBSW_REJECT_INVALID)) |
110 | { |
111 | p++; |
112 | width++; |
113 | break; |
114 | } |
115 | else |
116 | return -1; |
117 | } |
118 | |
119 | if (bytes == (size_t) -2) |
120 | /* An incomplete multibyte character at the end. */ |
121 | { |
122 | if (!(flags & MBSW_REJECT_INVALID)) |
123 | { |
124 | p = plimit; |
125 | width++; |
126 | break; |
127 | } |
128 | else |
129 | return -1; |
130 | } |
131 | |
132 | if (bytes == 0) |
133 | /* A null wide character was encountered. */ |
134 | bytes = 1; |
135 | |
136 | w = wcwidth (wc); |
137 | if (w >= 0) |
138 | /* A printable multibyte character. */ |
139 | { |
140 | if (w > INT_MAX - width) |
141 | goto overflow; |
142 | width += w; |
143 | } |
144 | else |
145 | /* An unprintable multibyte character. */ |
146 | if (!(flags & MBSW_REJECT_UNPRINTABLE)) |
147 | { |
148 | if (!iswcntrl (wc)) |
149 | { |
150 | if (width == INT_MAX) |
151 | goto overflow; |
152 | width++; |
153 | } |
154 | } |
155 | else |
156 | return -1; |
157 | |
158 | p += bytes; |
159 | } |
160 | while (! mbsinit (&mbstate)); |
161 | } |
162 | break; |
163 | } |
164 | return width; |
165 | } |
166 | |
167 | while (p < plimit) |
168 | { |
169 | unsigned char c = (unsigned char) *p++; |
170 | |
171 | if (isprint (c)) |
172 | { |
173 | if (width == INT_MAX) |
174 | goto overflow; |
175 | width++; |
176 | } |
177 | else if (!(flags & MBSW_REJECT_UNPRINTABLE)) |
178 | { |
179 | if (!iscntrl (c)) |
180 | { |
181 | if (width == INT_MAX) |
182 | goto overflow; |
183 | width++; |
184 | } |
185 | } |
186 | else |
187 | return -1; |
188 | } |
189 | return width; |
190 | |
191 | overflow: |
192 | return INT_MAX; |
193 | } |
194 | |