Commit | Line | Data |
---|---|---|
9255ee31 EZ |
1 | /* |
2 | * This is an implementation of wcwidth() and wcswidth() as defined in | |
3 | * "The Single UNIX Specification, Version 2, The Open Group, 1997" | |
4 | * <http://www.UNIX-systems.org/online.html> | |
5 | * | |
6 | * Markus Kuhn -- 2001-09-08 -- public domain | |
7 | */ | |
8 | ||
9 | #include <wchar.h> | |
10 | ||
11 | struct interval { | |
12 | unsigned short first; | |
13 | unsigned short last; | |
14 | }; | |
15 | ||
16 | /* auxiliary function for binary search in interval table */ | |
17 | static int bisearch(wchar_t ucs, const struct interval *table, int max) { | |
18 | int min = 0; | |
19 | int mid; | |
20 | ||
21 | if (ucs < table[0].first || ucs > table[max].last) | |
22 | return 0; | |
23 | while (max >= min) { | |
24 | mid = (min + max) / 2; | |
25 | if (ucs > table[mid].last) | |
26 | min = mid + 1; | |
27 | else if (ucs < table[mid].first) | |
28 | max = mid - 1; | |
29 | else | |
30 | return 1; | |
31 | } | |
32 | ||
33 | return 0; | |
34 | } | |
35 | ||
36 | ||
37 | /* The following functions define the column width of an ISO 10646 | |
38 | * character as follows: | |
39 | * | |
40 | * - The null character (U+0000) has a column width of 0. | |
41 | * | |
42 | * - Other C0/C1 control characters and DEL will lead to a return | |
43 | * value of -1. | |
44 | * | |
45 | * - Non-spacing and enclosing combining characters (general | |
46 | * category code Mn or Me in the Unicode database) have a | |
47 | * column width of 0. | |
48 | * | |
49 | * - Other format characters (general category code Cf in the Unicode | |
50 | * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0. | |
51 | * | |
52 | * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF) | |
53 | * have a column width of 0. | |
54 | * | |
55 | * - Spacing characters in the East Asian Wide (W) or East Asian | |
56 | * FullWidth (F) category as defined in Unicode Technical | |
57 | * Report #11 have a column width of 2. | |
58 | * | |
59 | * - All remaining characters (including all printable | |
60 | * ISO 8859-1 and WGL4 characters, Unicode control characters, | |
61 | * etc.) have a column width of 1. | |
62 | * | |
63 | * This implementation assumes that wchar_t characters are encoded | |
64 | * in ISO 10646. | |
65 | */ | |
66 | ||
67 | int wcwidth(wchar_t ucs) | |
68 | { | |
69 | /* sorted list of non-overlapping intervals of non-spacing characters */ | |
70 | static const struct interval combining[] = { | |
71 | { 0x0300, 0x034E }, { 0x0360, 0x0362 }, { 0x0483, 0x0486 }, | |
72 | { 0x0488, 0x0489 }, { 0x0591, 0x05A1 }, { 0x05A3, 0x05B9 }, | |
73 | { 0x05BB, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 }, | |
74 | { 0x05C4, 0x05C4 }, { 0x064B, 0x0655 }, { 0x0670, 0x0670 }, | |
75 | { 0x06D6, 0x06E4 }, { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, | |
76 | { 0x070F, 0x070F }, { 0x0711, 0x0711 }, { 0x0730, 0x074A }, | |
77 | { 0x07A6, 0x07B0 }, { 0x0901, 0x0902 }, { 0x093C, 0x093C }, | |
78 | { 0x0941, 0x0948 }, { 0x094D, 0x094D }, { 0x0951, 0x0954 }, | |
79 | { 0x0962, 0x0963 }, { 0x0981, 0x0981 }, { 0x09BC, 0x09BC }, | |
80 | { 0x09C1, 0x09C4 }, { 0x09CD, 0x09CD }, { 0x09E2, 0x09E3 }, | |
81 | { 0x0A02, 0x0A02 }, { 0x0A3C, 0x0A3C }, { 0x0A41, 0x0A42 }, | |
82 | { 0x0A47, 0x0A48 }, { 0x0A4B, 0x0A4D }, { 0x0A70, 0x0A71 }, | |
83 | { 0x0A81, 0x0A82 }, { 0x0ABC, 0x0ABC }, { 0x0AC1, 0x0AC5 }, | |
84 | { 0x0AC7, 0x0AC8 }, { 0x0ACD, 0x0ACD }, { 0x0B01, 0x0B01 }, | |
85 | { 0x0B3C, 0x0B3C }, { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 }, | |
86 | { 0x0B4D, 0x0B4D }, { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 }, | |
87 | { 0x0BC0, 0x0BC0 }, { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 }, | |
88 | { 0x0C46, 0x0C48 }, { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 }, | |
89 | { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 }, { 0x0CCC, 0x0CCD }, | |
90 | { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D }, { 0x0DCA, 0x0DCA }, | |
91 | { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 }, { 0x0E31, 0x0E31 }, | |
92 | { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E }, { 0x0EB1, 0x0EB1 }, | |
93 | { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC }, { 0x0EC8, 0x0ECD }, | |
94 | { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 }, { 0x0F37, 0x0F37 }, | |
95 | { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E }, { 0x0F80, 0x0F84 }, | |
96 | { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 }, { 0x0F99, 0x0FBC }, | |
97 | { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 }, { 0x1032, 0x1032 }, | |
98 | { 0x1036, 0x1037 }, { 0x1039, 0x1039 }, { 0x1058, 0x1059 }, | |
99 | { 0x1160, 0x11FF }, { 0x17B7, 0x17BD }, { 0x17C6, 0x17C6 }, | |
100 | { 0x17C9, 0x17D3 }, { 0x180B, 0x180E }, { 0x18A9, 0x18A9 }, | |
101 | { 0x200B, 0x200F }, { 0x202A, 0x202E }, { 0x206A, 0x206F }, | |
102 | { 0x20D0, 0x20E3 }, { 0x302A, 0x302F }, { 0x3099, 0x309A }, | |
103 | { 0xFB1E, 0xFB1E }, { 0xFE20, 0xFE23 }, { 0xFEFF, 0xFEFF }, | |
104 | { 0xFFF9, 0xFFFB } | |
105 | }; | |
106 | ||
107 | /* test for 8-bit control characters */ | |
108 | if (ucs == 0) | |
109 | return 0; | |
110 | if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0)) | |
111 | return -1; | |
112 | ||
113 | /* binary search in table of non-spacing characters */ | |
114 | if (bisearch(ucs, combining, | |
115 | sizeof(combining) / sizeof(struct interval) - 1)) | |
116 | return 0; | |
117 | ||
118 | /* if we arrive here, ucs is not a combining or C0/C1 control character */ | |
119 | ||
120 | return 1 + | |
121 | (ucs >= 0x1100 && | |
122 | (ucs <= 0x115f || /* Hangul Jamo init. consonants */ | |
123 | (ucs >= 0x2e80 && ucs <= 0xa4cf && (ucs & ~0x0011) != 0x300a && | |
124 | ucs != 0x303f) || /* CJK ... Yi */ | |
125 | (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */ | |
126 | (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */ | |
127 | (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */ | |
128 | (ucs >= 0xff00 && ucs <= 0xff5f) || /* Fullwidth Forms */ | |
129 | (ucs >= 0xffe0 && ucs <= 0xffe6) || | |
130 | (ucs >= 0x20000 && ucs <= 0x2ffff))); | |
131 | } | |
132 | ||
133 | ||
134 | int wcswidth(const wchar_t *pwcs, size_t n) | |
135 | { | |
136 | int w, width = 0; | |
137 | ||
138 | for (;*pwcs && n-- > 0; pwcs++) | |
139 | if ((w = wcwidth(*pwcs)) < 0) | |
140 | return -1; | |
141 | else | |
142 | width += w; | |
143 | ||
144 | return width; | |
145 | } | |
146 | ||
147 | ||
148 | /* | |
149 | * The following function is the same as wcwidth(), except that | |
150 | * spacing characters in the East Asian Ambiguous (A) category as | |
151 | * defined in Unicode Technical Report #11 have a column width of 2. | |
152 | * This experimental variant might be useful for users of CJK legacy | |
153 | * encodings who want to migrate to UCS. It is not otherwise | |
154 | * recommended for general use. | |
155 | */ | |
156 | static int wcwidth_cjk(wchar_t ucs) | |
157 | { | |
158 | /* sorted list of non-overlapping intervals of East Asian Ambiguous | |
159 | * characters */ | |
160 | static const struct interval ambiguous[] = { | |
161 | { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 }, | |
162 | { 0x00AA, 0x00AA }, { 0x00AD, 0x00AE }, { 0x00B0, 0x00B4 }, | |
163 | { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 }, | |
164 | { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 }, | |
165 | { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED }, | |
166 | { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA }, | |
167 | { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 }, | |
168 | { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B }, | |
169 | { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 }, | |
170 | { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 }, | |
171 | { 0x0148, 0x014B }, { 0x014D, 0x014D }, { 0x0152, 0x0153 }, | |
172 | { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE }, | |
173 | { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 }, | |
174 | { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA }, | |
175 | { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 }, | |
176 | { 0x02C4, 0x02C4 }, { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB }, | |
177 | { 0x02CD, 0x02CD }, { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB }, | |
178 | { 0x02DD, 0x02DD }, { 0x02DF, 0x02DF }, { 0x0300, 0x034E }, | |
179 | { 0x0360, 0x0362 }, { 0x0391, 0x03A1 }, { 0x03A3, 0x03A9 }, | |
180 | { 0x03B1, 0x03C1 }, { 0x03C3, 0x03C9 }, { 0x0401, 0x0401 }, | |
181 | { 0x0410, 0x044F }, { 0x0451, 0x0451 }, { 0x2010, 0x2010 }, | |
182 | { 0x2013, 0x2016 }, { 0x2018, 0x2019 }, { 0x201C, 0x201D }, | |
183 | { 0x2020, 0x2022 }, { 0x2024, 0x2027 }, { 0x2030, 0x2030 }, | |
184 | { 0x2032, 0x2033 }, { 0x2035, 0x2035 }, { 0x203B, 0x203B }, | |
185 | { 0x203E, 0x203E }, { 0x2074, 0x2074 }, { 0x207F, 0x207F }, | |
186 | { 0x2081, 0x2084 }, { 0x20AC, 0x20AC }, { 0x2103, 0x2103 }, | |
187 | { 0x2105, 0x2105 }, { 0x2109, 0x2109 }, { 0x2113, 0x2113 }, | |
188 | { 0x2116, 0x2116 }, { 0x2121, 0x2122 }, { 0x2126, 0x2126 }, | |
189 | { 0x212B, 0x212B }, { 0x2153, 0x2155 }, { 0x215B, 0x215E }, | |
190 | { 0x2160, 0x216B }, { 0x2170, 0x2179 }, { 0x2190, 0x2199 }, | |
191 | { 0x21B8, 0x21B9 }, { 0x21D2, 0x21D2 }, { 0x21D4, 0x21D4 }, | |
192 | { 0x21E7, 0x21E7 }, { 0x2200, 0x2200 }, { 0x2202, 0x2203 }, | |
193 | { 0x2207, 0x2208 }, { 0x220B, 0x220B }, { 0x220F, 0x220F }, | |
194 | { 0x2211, 0x2211 }, { 0x2215, 0x2215 }, { 0x221A, 0x221A }, | |
195 | { 0x221D, 0x2220 }, { 0x2223, 0x2223 }, { 0x2225, 0x2225 }, | |
196 | { 0x2227, 0x222C }, { 0x222E, 0x222E }, { 0x2234, 0x2237 }, | |
197 | { 0x223C, 0x223D }, { 0x2248, 0x2248 }, { 0x224C, 0x224C }, | |
198 | { 0x2252, 0x2252 }, { 0x2260, 0x2261 }, { 0x2264, 0x2267 }, | |
199 | { 0x226A, 0x226B }, { 0x226E, 0x226F }, { 0x2282, 0x2283 }, | |
200 | { 0x2286, 0x2287 }, { 0x2295, 0x2295 }, { 0x2299, 0x2299 }, | |
201 | { 0x22A5, 0x22A5 }, { 0x22BF, 0x22BF }, { 0x2312, 0x2312 }, | |
202 | { 0x2329, 0x232A }, { 0x2460, 0x24BF }, { 0x24D0, 0x24E9 }, | |
203 | { 0x2500, 0x254B }, { 0x2550, 0x2574 }, { 0x2580, 0x258F }, | |
204 | { 0x2592, 0x2595 }, { 0x25A0, 0x25A1 }, { 0x25A3, 0x25A9 }, | |
205 | { 0x25B2, 0x25B3 }, { 0x25B6, 0x25B7 }, { 0x25BC, 0x25BD }, | |
206 | { 0x25C0, 0x25C1 }, { 0x25C6, 0x25C8 }, { 0x25CB, 0x25CB }, | |
207 | { 0x25CE, 0x25D1 }, { 0x25E2, 0x25E5 }, { 0x25EF, 0x25EF }, | |
208 | { 0x2605, 0x2606 }, { 0x2609, 0x2609 }, { 0x260E, 0x260F }, | |
209 | { 0x261C, 0x261C }, { 0x261E, 0x261E }, { 0x2640, 0x2640 }, | |
210 | { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, { 0x2663, 0x2665 }, | |
211 | { 0x2667, 0x266A }, { 0x266C, 0x266D }, { 0x266F, 0x266F }, | |
212 | { 0x273D, 0x273D }, { 0x3008, 0x300B }, { 0x3014, 0x3015 }, | |
213 | { 0x3018, 0x301B }, { 0xFFFD, 0xFFFD } | |
214 | }; | |
215 | ||
216 | /* binary search in table of non-spacing characters */ | |
217 | if (bisearch(ucs, ambiguous, | |
218 | sizeof(ambiguous) / sizeof(struct interval) - 1)) | |
219 | return 2; | |
220 | ||
221 | return wcwidth(ucs); | |
222 | } | |
223 | ||
224 | ||
225 | int wcswidth_cjk(const wchar_t *pwcs, size_t n) | |
226 | { | |
227 | int w, width = 0; | |
228 | ||
229 | for (;*pwcs && n-- > 0; pwcs++) | |
230 | if ((w = wcwidth_cjk(*pwcs)) < 0) | |
231 | return -1; | |
232 | else | |
233 | width += w; | |
234 | ||
235 | return width; | |
236 | } |