jcs.org
mutt stable branch with some hacks
1/*
2 * This is an implementation of wcwidth() and wcswidth() (defined in
3 * IEEE Std 1002.1-2001) for Unicode.
4 *
5 * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html
6 * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html
7 *
8 * Markus Kuhn -- 2007-05-26 (Unicode 5.0)
9 *
10 * Permission to use, copy, modify, and distribute this software
11 * for any purpose and without fee is hereby granted. The author
12 * disclaims all warranties with regard to this software.
13 *
14 * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c
15 */
16
17/* Changes made for mutt:
18 * - Adapted for Mutt by Edmund Grimley Evans.
19 * - Changed 'first'/'last' members of combined[] to wchar_t from
20 * unsigned short to fix compiler warnings, 2007-11-13, Rocco Rutte
21 */
22
23#if HAVE_CONFIG_H
24# include "config.h"
25#endif
26
27#ifndef HAVE_WC_FUNCS
28
29#include "mutt.h"
30#include "mbyte.h"
31
32#include <ctype.h>
33
34/* The following two functions define the column width of an ISO 10646
35 * character as follows:
36 *
37 * - The null character (U+0000) has a column width of 0.
38 *
39 * - Other C0/C1 control characters and DEL will lead to a return
40 * value of -1.
41 *
42 * - Non-spacing and enclosing combining characters (general
43 * category code Mn or Me in the Unicode database) have a
44 * column width of 0.
45 *
46 * - SOFT HYPHEN (U+00AD) has a column width of 1.
47 *
48 * - Other format characters (general category code Cf in the Unicode
49 * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0.
50 *
51 * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF)
52 * have a column width of 0.
53 *
54 * - Spacing characters in the East Asian Wide (W) or East Asian
55 * Full-width (F) category as defined in Unicode Technical
56 * Report #11 have a column width of 2.
57 *
58 * - All remaining characters (including all printable
59 * ISO 8859-1 and WGL4 characters, Unicode control characters,
60 * etc.) have a column width of 1.
61 *
62 * This implementation assumes that wchar_t characters are encoded
63 * in ISO 10646.
64 */
65
66int wcwidth_ucs(wchar_t ucs)
67{
68 /* sorted list of non-overlapping intervals of non-spacing characters */
69 /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */
70 static const struct interval {
71 wchar_t first;
72 wchar_t last;
73 } combining[] = {
74 { 0x0300, 0x036f }, { 0x0483, 0x0486 }, { 0x0488, 0x0489 },
75 { 0x0591, 0x05bd }, { 0x05bf, 0x05bf }, { 0x05c1, 0x05c2 },
76 { 0x05c4, 0x05c5 }, { 0x05c7, 0x05c7 }, { 0x0600, 0x0603 },
77 { 0x0610, 0x0615 }, { 0x064b, 0x065e }, { 0x0670, 0x0670 },
78 { 0x06d6, 0x06e4 }, { 0x06e7, 0x06e8 }, { 0x06ea, 0x06ed },
79 { 0x070f, 0x070f }, { 0x0711, 0x0711 }, { 0x0730, 0x074a },
80 { 0x07a6, 0x07b0 }, { 0x07eb, 0x07f3 }, { 0x0901, 0x0902 },
81 { 0x093c, 0x093c }, { 0x0941, 0x0948 }, { 0x094d, 0x094d },
82 { 0x0951, 0x0954 }, { 0x0962, 0x0963 }, { 0x0981, 0x0981 },
83 { 0x09bc, 0x09bc }, { 0x09c1, 0x09c4 }, { 0x09cd, 0x09cd },
84 { 0x09e2, 0x09e3 }, { 0x0a01, 0x0a02 }, { 0x0a3c, 0x0a3c },
85 { 0x0a41, 0x0a42 }, { 0x0a47, 0x0a48 }, { 0x0a4b, 0x0a4d },
86 { 0x0a70, 0x0a71 }, { 0x0a81, 0x0a82 }, { 0x0abc, 0x0abc },
87 { 0x0ac1, 0x0ac5 }, { 0x0ac7, 0x0ac8 }, { 0x0acd, 0x0acd },
88 { 0x0ae2, 0x0ae3 }, { 0x0b01, 0x0b01 }, { 0x0b3c, 0x0b3c },
89 { 0x0b3f, 0x0b3f }, { 0x0b41, 0x0b43 }, { 0x0b4d, 0x0b4d },
90 { 0x0b56, 0x0b56 }, { 0x0b82, 0x0b82 }, { 0x0bc0, 0x0bc0 },
91 { 0x0bcd, 0x0bcd }, { 0x0c3e, 0x0c40 }, { 0x0c46, 0x0c48 },
92 { 0x0c4a, 0x0c4d }, { 0x0c55, 0x0c56 }, { 0x0cbc, 0x0cbc },
93 { 0x0cbf, 0x0cbf }, { 0x0cc6, 0x0cc6 }, { 0x0ccc, 0x0ccd },
94 { 0x0ce2, 0x0ce3 }, { 0x0d41, 0x0d43 }, { 0x0d4d, 0x0d4d },
95 { 0x0dca, 0x0dca }, { 0x0dd2, 0x0dd4 }, { 0x0dd6, 0x0dd6 },
96 { 0x0e31, 0x0e31 }, { 0x0e34, 0x0e3a }, { 0x0e47, 0x0e4e },
97 { 0x0eb1, 0x0eb1 }, { 0x0eb4, 0x0eb9 }, { 0x0ebb, 0x0ebc },
98 { 0x0ec8, 0x0ecd }, { 0x0f18, 0x0f19 }, { 0x0f35, 0x0f35 },
99 { 0x0f37, 0x0f37 }, { 0x0f39, 0x0f39 }, { 0x0f71, 0x0f7e },
100 { 0x0f80, 0x0f84 }, { 0x0f86, 0x0f87 }, { 0x0f90, 0x0f97 },
101 { 0x0f99, 0x0fbc }, { 0x0fc6, 0x0fc6 }, { 0x102d, 0x1030 },
102 { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 },
103 { 0x1058, 0x1059 }, { 0x1160, 0x11ff }, { 0x135f, 0x135f },
104 { 0x1712, 0x1714 }, { 0x1732, 0x1734 }, { 0x1752, 0x1753 },
105 { 0x1772, 0x1773 }, { 0x17b4, 0x17b5 }, { 0x17b7, 0x17bd },
106 { 0x17c6, 0x17c6 }, { 0x17c9, 0x17d3 }, { 0x17dd, 0x17dd },
107 { 0x180b, 0x180d }, { 0x18a9, 0x18a9 }, { 0x1920, 0x1922 },
108 { 0x1927, 0x1928 }, { 0x1932, 0x1932 }, { 0x1939, 0x193b },
109 { 0x1a17, 0x1a18 }, { 0x1b00, 0x1b03 }, { 0x1b34, 0x1b34 },
110 { 0x1b36, 0x1b3a }, { 0x1b3c, 0x1b3c }, { 0x1b42, 0x1b42 },
111 { 0x1b6b, 0x1b73 }, { 0x1dc0, 0x1dca }, { 0x1dfe, 0x1dff },
112 { 0x200b, 0x200f }, { 0x202a, 0x202e }, { 0x2060, 0x2063 },
113 { 0x206a, 0x206f }, { 0x20d0, 0x20ef }, { 0x302a, 0x302f },
114 { 0x3099, 0x309a }, { 0xa806, 0xa806 }, { 0xa80b, 0xa80b },
115 { 0xa825, 0xa826 }, { 0xfb1e, 0xfb1e }, { 0xfe00, 0xfe0f },
116 { 0xfe20, 0xfe23 }, { 0xfeff, 0xfeff }, { 0xfff9, 0xfffb },
117 { 0x10a01, 0x10a03 }, { 0x10a05, 0x10a06 }, { 0x10a0c, 0x10a0f },
118 { 0x10a38, 0x10a3a }, { 0x10a3f, 0x10a3f }, { 0x1d167, 0x1d169 },
119 { 0x1d173, 0x1d182 }, { 0x1d185, 0x1d18b }, { 0x1d1aa, 0x1d1ad },
120 { 0x1d242, 0x1d244 }, { 0xe0001, 0xe0001 }, { 0xe0020, 0xe007f },
121 { 0xe0100, 0xe01ef }
122 };
123 int min = 0;
124 int max = sizeof(combining) / sizeof(struct interval) - 1;
125 int mid;
126
127 /* test for 8-bit control characters */
128 if (ucs == 0)
129 return 0;
130 if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0))
131 return -1;
132
133 /* first quick check for Latin-1 etc. characters */
134 if (ucs < combining[0].first)
135 return 1;
136
137 /* binary search in table of non-spacing characters */
138 while (max >= min)
139 {
140 mid = (min + max) / 2;
141 if (combining[mid].last < ucs)
142 min = mid + 1;
143 else if (combining[mid].first > ucs)
144 max = mid - 1;
145 else if (combining[mid].first <= ucs && combining[mid].last >= ucs)
146 return 0;
147 }
148
149 /* if we arrive here, ucs is not a combining or C0/C1 control character */
150
151 /* fast test for majority of non-wide scripts */
152 if (ucs < 0x1100)
153 return 1;
154
155 return 1 +
156 (ucs >= 0x1100 &&
157 (ucs <= 0x115f || /* Hangul Jamo init. consonants */
158 ucs == 0x2329 || ucs == 0x232a ||
159 (ucs >= 0x2e80 && ucs <= 0xa4cf &&
160 ucs != 0x303f) || /* CJK ... Yi */
161 (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */
162 (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */
163 (ucs >= 0xfe10 && ucs <= 0xfe19) || /* Vertical forms */
164 (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */
165 (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */
166 (ucs >= 0xffe0 && ucs <= 0xffe6) ||
167 (ucs >= 0x20000 && ucs <= 0x2fffd) ||
168 (ucs >= 0x30000 && ucs <= 0x3fffd)));
169}
170
171#endif /* !HAVE_WC_FUNCS */
172
173#if 0 /* original */
174int wcswidth(const wchar_t *pwcs, size_t n)
175{
176 int w, width = 0;
177
178 for (;*pwcs && n-- > 0; pwcs++)
179 if ((w = wcwidth(*pwcs)) < 0)
180 return -1;
181 else
182 width += w;
183
184 return width;
185}
186#endif