File pdcurses/wcwidth.c added (mode: 100644) (index 00000000..9d117bf2) |
|
1 |
|
/* |
|
2 |
|
* Copyright (C) Markus Kuhn -- 2007-05-26 (Unicode 5.0) |
|
3 |
|
* Copyright (C) xaizek -- 2014 (Adapt for vifm) |
|
4 |
|
* Copyright (C) xaizek -- 2016 (Adapt for pdcurses) |
|
5 |
|
* |
|
6 |
|
* This is an implementation of wcwidth() and wcswidth() (defined in |
|
7 |
|
* IEEE Std 1002.1-2001) for Unicode. |
|
8 |
|
* |
|
9 |
|
* http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html |
|
10 |
|
* http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html |
|
11 |
|
* |
|
12 |
|
* In fixed-width output devices, Latin characters all occupy a single |
|
13 |
|
* "cell" position of equal width, whereas ideographic CJK characters |
|
14 |
|
* occupy two such cells. Interoperability between terminal-line |
|
15 |
|
* applications and (teletype-style) character terminals using the |
|
16 |
|
* UTF-8 encoding requires agreement on which character should advance |
|
17 |
|
* the cursor by how many cell positions. No established formal |
|
18 |
|
* standards exist at present on which Unicode character shall occupy |
|
19 |
|
* how many cell positions on character terminals. These routines are |
|
20 |
|
* a first attempt of defining such behavior based on simple rules |
|
21 |
|
* applied to data provided by the Unicode Consortium. |
|
22 |
|
* |
|
23 |
|
* For some graphical characters, the Unicode standard explicitly |
|
24 |
|
* defines a character-cell width via the definition of the East Asian |
|
25 |
|
* FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes. |
|
26 |
|
* In all these cases, there is no ambiguity about which width a |
|
27 |
|
* terminal shall use. For characters in the East Asian Ambiguous (A) |
|
28 |
|
* class, the width choice depends purely on a preference of backward |
|
29 |
|
* compatibility with either historic CJK or Western practice. |
|
30 |
|
* Choosing single-width for these characters is easy to justify as |
|
31 |
|
* the appropriate long-term solution, as the CJK practice of |
|
32 |
|
* displaying these characters as double-width comes from historic |
|
33 |
|
* implementation simplicity (8-bit encoded characters were displayed |
|
34 |
|
* single-width and 16-bit ones double-width, even for Greek, |
|
35 |
|
* Cyrillic, etc.) and not any typographic considerations. |
|
36 |
|
* |
|
37 |
|
* Much less clear is the choice of width for the Not East Asian |
|
38 |
|
* (Neutral) class. Existing practice does not dictate a width for any |
|
39 |
|
* of these characters. It would nevertheless make sense |
|
40 |
|
* typographically to allocate two character cells to characters such |
|
41 |
|
* as for instance EM SPACE or VOLUME INTEGRAL, which cannot be |
|
42 |
|
* represented adequately with a single-width glyph. The following |
|
43 |
|
* routines at present merely assign a single-cell width to all |
|
44 |
|
* neutral characters, in the interest of simplicity. This is not |
|
45 |
|
* entirely satisfactory and should be reconsidered before |
|
46 |
|
* establishing a formal standard in this area. At the moment, the |
|
47 |
|
* decision which Not East Asian (Neutral) characters should be |
|
48 |
|
* represented by double-width glyphs cannot yet be answered by |
|
49 |
|
* applying a simple rule from the Unicode database content. Setting |
|
50 |
|
* up a proper standard for the behavior of UTF-8 character terminals |
|
51 |
|
* will require a careful analysis not only of each Unicode character, |
|
52 |
|
* but also of each presentation form, something the author of these |
|
53 |
|
* routines has avoided to do so far. |
|
54 |
|
* |
|
55 |
|
* http://www.unicode.org/unicode/reports/tr11/ |
|
56 |
|
* |
|
57 |
|
* Permission to use, copy, modify, and distribute this software |
|
58 |
|
* for any purpose and without fee is hereby granted. The author |
|
59 |
|
* disclaims all warranties with regard to this software. |
|
60 |
|
* |
|
61 |
|
* Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c |
|
62 |
|
*/ |
|
63 |
|
|
|
64 |
|
#ifdef PDC_WIDE |
|
65 |
|
|
|
66 |
|
#include <curspriv.h> |
|
67 |
|
|
|
68 |
|
#include <wchar.h> /* wchar_t */ |
|
69 |
|
|
|
70 |
|
typedef struct |
|
71 |
|
{ |
|
72 |
|
int first; |
|
73 |
|
int last; |
|
74 |
|
} |
|
75 |
|
interval; |
|
76 |
|
|
|
77 |
|
static int bisearch(wchar_t ucs, const interval *table, int max); |
|
78 |
|
|
|
79 |
|
/* The following two functions define the column width of an ISO 10646 |
|
80 |
|
* character as follows: |
|
81 |
|
* - The null character (U+0000) has a column width of 0. |
|
82 |
|
* - Other C0/C1 control characters and DEL will lead to a return |
|
83 |
|
* value of -1. |
|
84 |
|
* - Non-spacing and enclosing combining characters (general |
|
85 |
|
* category code Mn or Me in the Unicode database) have a |
|
86 |
|
* column width of 0. |
|
87 |
|
* - SOFT HYPHEN (U+00AD) has a column width of 1. |
|
88 |
|
* - Other format characters (general category code Cf in the Unicode |
|
89 |
|
* database) and ZERO WIDTH SPACE (U+200B) have a column width of 0. |
|
90 |
|
* - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF) |
|
91 |
|
* have a column width of 0. |
|
92 |
|
* - Spacing characters in the East Asian Wide (W) or East Asian |
|
93 |
|
* Full-width (F) category as defined in Unicode Technical |
|
94 |
|
* Report #11 have a column width of 2. |
|
95 |
|
* - All remaining characters (including all printable |
|
96 |
|
* ISO 8859-1 and WGL4 characters, Unicode control characters, |
|
97 |
|
* etc.) have a column width of 1. |
|
98 |
|
* This implementation assumes that wchar_t characters are encoded |
|
99 |
|
* in ISO 10646. |
|
100 |
|
*/ |
|
101 |
|
int |
|
102 |
|
compat_wcwidth(wchar_t ucs) |
|
103 |
|
{ |
|
104 |
|
/* Sorted list of non-overlapping intervals of non-spacing characters |
|
105 |
|
* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c". */ |
|
106 |
|
static const interval combining[] = |
|
107 |
|
{ |
|
108 |
|
{ 0x0300, 0x036F }, { 0x0483, 0x0486 }, { 0x0488, 0x0489 }, |
|
109 |
|
{ 0x0591, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 }, |
|
110 |
|
{ 0x05C4, 0x05C5 }, { 0x05C7, 0x05C7 }, { 0x0600, 0x0603 }, |
|
111 |
|
{ 0x0610, 0x0615 }, { 0x064B, 0x065E }, { 0x0670, 0x0670 }, |
|
112 |
|
{ 0x06D6, 0x06E4 }, { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, |
|
113 |
|
{ 0x070F, 0x070F }, { 0x0711, 0x0711 }, { 0x0730, 0x074A }, |
|
114 |
|
{ 0x07A6, 0x07B0 }, { 0x07EB, 0x07F3 }, { 0x0901, 0x0902 }, |
|
115 |
|
{ 0x093C, 0x093C }, { 0x0941, 0x0948 }, { 0x094D, 0x094D }, |
|
116 |
|
{ 0x0951, 0x0954 }, { 0x0962, 0x0963 }, { 0x0981, 0x0981 }, |
|
117 |
|
{ 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 }, { 0x09CD, 0x09CD }, |
|
118 |
|
{ 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 }, { 0x0A3C, 0x0A3C }, |
|
119 |
|
{ 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 }, { 0x0A4B, 0x0A4D }, |
|
120 |
|
{ 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 }, { 0x0ABC, 0x0ABC }, |
|
121 |
|
{ 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 }, { 0x0ACD, 0x0ACD }, |
|
122 |
|
{ 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 }, { 0x0B3C, 0x0B3C }, |
|
123 |
|
{ 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 }, { 0x0B4D, 0x0B4D }, |
|
124 |
|
{ 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 }, { 0x0BC0, 0x0BC0 }, |
|
125 |
|
{ 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 }, { 0x0C46, 0x0C48 }, |
|
126 |
|
{ 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 }, { 0x0CBC, 0x0CBC }, |
|
127 |
|
{ 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 }, { 0x0CCC, 0x0CCD }, |
|
128 |
|
{ 0x0CE2, 0x0CE3 }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D }, |
|
129 |
|
{ 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 }, |
|
130 |
|
{ 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E }, |
|
131 |
|
{ 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC }, |
|
132 |
|
{ 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 }, |
|
133 |
|
{ 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E }, |
|
134 |
|
{ 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 }, |
|
135 |
|
{ 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 }, |
|
136 |
|
{ 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 }, |
|
137 |
|
{ 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x135F, 0x135F }, |
|
138 |
|
{ 0x1712, 0x1714 }, { 0x1732, 0x1734 }, { 0x1752, 0x1753 }, |
|
139 |
|
{ 0x1772, 0x1773 }, { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD }, |
|
140 |
|
{ 0x17C6, 0x17C6 }, { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD }, |
|
141 |
|
{ 0x180B, 0x180D }, { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 }, |
|
142 |
|
{ 0x1927, 0x1928 }, { 0x1932, 0x1932 }, { 0x1939, 0x193B }, |
|
143 |
|
{ 0x1A17, 0x1A18 }, { 0x1B00, 0x1B03 }, { 0x1B34, 0x1B34 }, |
|
144 |
|
{ 0x1B36, 0x1B3A }, { 0x1B3C, 0x1B3C }, { 0x1B42, 0x1B42 }, |
|
145 |
|
{ 0x1B6B, 0x1B73 }, { 0x1DC0, 0x1DCA }, { 0x1DFE, 0x1DFF }, |
|
146 |
|
{ 0x200B, 0x200F }, { 0x202A, 0x202E }, { 0x2060, 0x2063 }, |
|
147 |
|
{ 0x206A, 0x206F }, { 0x20D0, 0x20EF }, { 0x302A, 0x302F }, |
|
148 |
|
{ 0x3099, 0x309A }, { 0xA806, 0xA806 }, { 0xA80B, 0xA80B }, |
|
149 |
|
{ 0xA825, 0xA826 }, { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F }, |
|
150 |
|
{ 0xFE20, 0xFE23 }, { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB }, |
|
151 |
|
{ 0x10A01, 0x10A03 }, { 0x10A05, 0x10A06 }, { 0x10A0C, 0x10A0F }, |
|
152 |
|
{ 0x10A38, 0x10A3A }, { 0x10A3F, 0x10A3F }, { 0x1D167, 0x1D169 }, |
|
153 |
|
{ 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B }, { 0x1D1AA, 0x1D1AD }, |
|
154 |
|
{ 0x1D242, 0x1D244 }, { 0xE0001, 0xE0001 }, { 0xE0020, 0xE007F }, |
|
155 |
|
{ 0xE0100, 0xE01EF } |
|
156 |
|
}; |
|
157 |
|
|
|
158 |
|
/* Test for 8-bit control characters. */ |
|
159 |
|
if(ucs == 0) |
|
160 |
|
{ |
|
161 |
|
return 0; |
|
162 |
|
} |
|
163 |
|
if(ucs < 32 || (ucs >= 0x7f && ucs < 0xa0)) |
|
164 |
|
{ |
|
165 |
|
return -1; |
|
166 |
|
} |
|
167 |
|
|
|
168 |
|
/* Binary search in table of non-spacing characters. */ |
|
169 |
|
if(bisearch(ucs, combining, sizeof(combining)/sizeof(combining[0]) - 1)) |
|
170 |
|
{ |
|
171 |
|
return 0; |
|
172 |
|
} |
|
173 |
|
|
|
174 |
|
/* If we arrive here, ucs is not a combining or C0/C1 control character. */ |
|
175 |
|
|
|
176 |
|
return 1 + |
|
177 |
|
(ucs >= 0x1100 && |
|
178 |
|
(ucs <= 0x115f || /* Hangul Jamo init. consonants */ |
|
179 |
|
ucs == 0x2329 || ucs == 0x232a || |
|
180 |
|
(ucs >= 0x2e80 && ucs <= 0xa4cf && |
|
181 |
|
ucs != 0x303f) || /* CJK ... Yi */ |
|
182 |
|
(ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */ |
|
183 |
|
(ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */ |
|
184 |
|
(ucs >= 0xfe10 && ucs <= 0xfe19) || /* Vertical forms */ |
|
185 |
|
(ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */ |
|
186 |
|
(ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */ |
|
187 |
|
(ucs >= 0xffe0 && ucs <= 0xffe6) || |
|
188 |
|
(ucs >= 0x20000 && ucs <= 0x2fffd) || |
|
189 |
|
(ucs >= 0x30000 && ucs <= 0x3fffd))); |
|
190 |
|
} |
|
191 |
|
|
|
192 |
|
/* Auxiliary function for binary search in interval table. */ |
|
193 |
|
static int |
|
194 |
|
bisearch(wchar_t ucs, const interval *table, int max) |
|
195 |
|
{ |
|
196 |
|
int min; |
|
197 |
|
|
|
198 |
|
if(ucs < table[0].first || ucs > table[max].last) |
|
199 |
|
{ |
|
200 |
|
return 0; |
|
201 |
|
} |
|
202 |
|
|
|
203 |
|
min = 0; |
|
204 |
|
while(max >= min) |
|
205 |
|
{ |
|
206 |
|
const int mid = (min + max)/2; |
|
207 |
|
if(ucs > table[mid].last) |
|
208 |
|
{ |
|
209 |
|
min = mid + 1; |
|
210 |
|
} |
|
211 |
|
else if(ucs < table[mid].first) |
|
212 |
|
{ |
|
213 |
|
max = mid - 1; |
|
214 |
|
} |
|
215 |
|
else |
|
216 |
|
{ |
|
217 |
|
return 1; |
|
218 |
|
} |
|
219 |
|
} |
|
220 |
|
|
|
221 |
|
return 0; |
|
222 |
|
} |
|
223 |
|
|
|
224 |
|
#endif |