| 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 -- 2001-01-12 -- public domain |
| 7 | */ |
| 8 | |
| 9 | #include <stddef.h> |
| 10 | |
| 11 | #include "putty.h" |
| 12 | |
| 13 | struct interval { |
| 14 | unsigned short first; |
| 15 | unsigned short last; |
| 16 | }; |
| 17 | |
| 18 | /* auxiliary function for binary search in interval table */ |
| 19 | static int bisearch(wchar_t ucs, const struct interval *table, int max) { |
| 20 | int min = 0; |
| 21 | int mid; |
| 22 | |
| 23 | if (ucs < table[0].first || ucs > table[max].last) |
| 24 | return 0; |
| 25 | while (max >= min) { |
| 26 | mid = (min + max) / 2; |
| 27 | if (ucs > table[mid].last) |
| 28 | min = mid + 1; |
| 29 | else if (ucs < table[mid].first) |
| 30 | max = mid - 1; |
| 31 | else |
| 32 | return 1; |
| 33 | } |
| 34 | |
| 35 | return 0; |
| 36 | } |
| 37 | |
| 38 | |
| 39 | /* The following functions define the column width of an ISO 10646 |
| 40 | * character as follows: |
| 41 | * |
| 42 | * - The null character (U+0000) has a column width of 0. |
| 43 | * |
| 44 | * - Other C0/C1 control characters and DEL will lead to a return |
| 45 | * value of -1. |
| 46 | * |
| 47 | * - Non-spacing and enclosing combining characters (general |
| 48 | * category code Mn or Me in the Unicode database) have a |
| 49 | * column width of 0. |
| 50 | * |
| 51 | * - Other format characters (general category code Cf in the Unicode |
| 52 | * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0. |
| 53 | * |
| 54 | * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF) |
| 55 | * have a column width of 0. |
| 56 | * |
| 57 | * - Spacing characters in the East Asian Wide (W) or East Asian |
| 58 | * FullWidth (F) category as defined in Unicode Technical |
| 59 | * Report #11 have a column width of 2. |
| 60 | * |
| 61 | * - All remaining characters (including all printable |
| 62 | * ISO 8859-1 and WGL4 characters, Unicode control characters, |
| 63 | * etc.) have a column width of 1. |
| 64 | * |
| 65 | * This implementation assumes that wchar_t characters are encoded |
| 66 | * in ISO 10646. |
| 67 | */ |
| 68 | |
| 69 | int wcwidth(wchar_t ucs) |
| 70 | { |
| 71 | /* sorted list of non-overlapping intervals of non-spacing characters */ |
| 72 | static const struct interval combining[] = { |
| 73 | { 0x0300, 0x034E }, { 0x0360, 0x0362 }, { 0x0483, 0x0486 }, |
| 74 | { 0x0488, 0x0489 }, { 0x0591, 0x05A1 }, { 0x05A3, 0x05B9 }, |
| 75 | { 0x05BB, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 }, |
| 76 | { 0x05C4, 0x05C4 }, { 0x064B, 0x0655 }, { 0x0670, 0x0670 }, |
| 77 | { 0x06D6, 0x06E4 }, { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, |
| 78 | { 0x070F, 0x070F }, { 0x0711, 0x0711 }, { 0x0730, 0x074A }, |
| 79 | { 0x07A6, 0x07B0 }, { 0x0901, 0x0902 }, { 0x093C, 0x093C }, |
| 80 | { 0x0941, 0x0948 }, { 0x094D, 0x094D }, { 0x0951, 0x0954 }, |
| 81 | { 0x0962, 0x0963 }, { 0x0981, 0x0981 }, { 0x09BC, 0x09BC }, |
| 82 | { 0x09C1, 0x09C4 }, { 0x09CD, 0x09CD }, { 0x09E2, 0x09E3 }, |
| 83 | { 0x0A02, 0x0A02 }, { 0x0A3C, 0x0A3C }, { 0x0A41, 0x0A42 }, |
| 84 | { 0x0A47, 0x0A48 }, { 0x0A4B, 0x0A4D }, { 0x0A70, 0x0A71 }, |
| 85 | { 0x0A81, 0x0A82 }, { 0x0ABC, 0x0ABC }, { 0x0AC1, 0x0AC5 }, |
| 86 | { 0x0AC7, 0x0AC8 }, { 0x0ACD, 0x0ACD }, { 0x0B01, 0x0B01 }, |
| 87 | { 0x0B3C, 0x0B3C }, { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 }, |
| 88 | { 0x0B4D, 0x0B4D }, { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 }, |
| 89 | { 0x0BC0, 0x0BC0 }, { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 }, |
| 90 | { 0x0C46, 0x0C48 }, { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 }, |
| 91 | { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 }, { 0x0CCC, 0x0CCD }, |
| 92 | { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D }, { 0x0DCA, 0x0DCA }, |
| 93 | { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 }, { 0x0E31, 0x0E31 }, |
| 94 | { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E }, { 0x0EB1, 0x0EB1 }, |
| 95 | { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC }, { 0x0EC8, 0x0ECD }, |
| 96 | { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 }, { 0x0F37, 0x0F37 }, |
| 97 | { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E }, { 0x0F80, 0x0F84 }, |
| 98 | { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 }, { 0x0F99, 0x0FBC }, |
| 99 | { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 }, { 0x1032, 0x1032 }, |
| 100 | { 0x1036, 0x1037 }, { 0x1039, 0x1039 }, { 0x1058, 0x1059 }, |
| 101 | { 0x1160, 0x11FF }, { 0x17B7, 0x17BD }, { 0x17C6, 0x17C6 }, |
| 102 | { 0x17C9, 0x17D3 }, { 0x180B, 0x180E }, { 0x18A9, 0x18A9 }, |
| 103 | { 0x200B, 0x200F }, { 0x202A, 0x202E }, { 0x206A, 0x206F }, |
| 104 | { 0x20D0, 0x20E3 }, { 0x302A, 0x302F }, { 0x3099, 0x309A }, |
| 105 | { 0xFB1E, 0xFB1E }, { 0xFE20, 0xFE23 }, { 0xFEFF, 0xFEFF }, |
| 106 | { 0xFFF9, 0xFFFB } |
| 107 | }; |
| 108 | |
| 109 | /* test for 8-bit control characters */ |
| 110 | if (ucs == 0) |
| 111 | return 0; |
| 112 | if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0)) |
| 113 | return -1; |
| 114 | |
| 115 | /* binary search in table of non-spacing characters */ |
| 116 | if (bisearch(ucs, combining, |
| 117 | sizeof(combining) / sizeof(struct interval) - 1)) |
| 118 | return 0; |
| 119 | |
| 120 | /* if we arrive here, ucs is not a combining or C0/C1 control character */ |
| 121 | |
| 122 | return 1 + |
| 123 | (ucs >= 0x1100 && |
| 124 | (ucs <= 0x115f || /* Hangul Jamo init. consonants */ |
| 125 | (ucs >= 0x2e80 && ucs <= 0xa4cf && (ucs & ~0x0011) != 0x300a && |
| 126 | ucs != 0x303f) || /* CJK ... Yi */ |
| 127 | (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */ |
| 128 | (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */ |
| 129 | (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */ |
| 130 | (ucs >= 0xff00 && ucs <= 0xff5f) || /* Fullwidth Forms */ |
| 131 | (ucs >= 0xffe0 && ucs <= 0xffe6) || |
| 132 | (ucs >= 0x20000 && ucs <= 0x2ffff))); |
| 133 | } |
| 134 | |
| 135 | |
| 136 | int wcswidth(const wchar_t *pwcs, size_t n) |
| 137 | { |
| 138 | int w, width = 0; |
| 139 | |
| 140 | for (;*pwcs && n-- > 0; pwcs++) |
| 141 | if ((w = wcwidth(*pwcs)) < 0) |
| 142 | return -1; |
| 143 | else |
| 144 | width += w; |
| 145 | |
| 146 | return width; |
| 147 | } |
| 148 | |
| 149 | #if 0 /* RDB: we don't need the cjk version */ |
| 150 | /* |
| 151 | * The following function is the same as wcwidth(), except that |
| 152 | * spacing characters in the East Asian Ambiguous (A) category as |
| 153 | * defined in Unicode Technical Report #11 have a column width of 2. |
| 154 | * This experimental variant might be useful for users of CJK legacy |
| 155 | * encodings who want to migrate to UCS. It is not otherwise |
| 156 | * recommended for general use. |
| 157 | */ |
| 158 | static int wcwidth_cjk(wchar_t ucs) |
| 159 | { |
| 160 | /* sorted list of non-overlapping intervals of East Asian Ambiguous |
| 161 | * characters */ |
| 162 | static const struct interval ambiguous[] = { |
| 163 | { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 }, |
| 164 | { 0x00AA, 0x00AA }, { 0x00AD, 0x00AD }, { 0x00B0, 0x00B4 }, |
| 165 | { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 }, |
| 166 | { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 }, |
| 167 | { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED }, |
| 168 | { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA }, |
| 169 | { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 }, |
| 170 | { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B }, |
| 171 | { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 }, |
| 172 | { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 }, |
| 173 | { 0x0148, 0x014A }, { 0x014D, 0x014D }, { 0x0152, 0x0153 }, |
| 174 | { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE }, |
| 175 | { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 }, |
| 176 | { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA }, |
| 177 | { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 }, |
| 178 | { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB }, { 0x02CD, 0x02CD }, |
| 179 | { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB }, { 0x02DD, 0x02DD }, |
| 180 | { 0x0391, 0x03A1 }, { 0x03A3, 0x03A9 }, { 0x03B1, 0x03C1 }, |
| 181 | { 0x03C3, 0x03C9 }, { 0x0401, 0x0401 }, { 0x0410, 0x044F }, |
| 182 | { 0x0451, 0x0451 }, { 0x2010, 0x2010 }, { 0x2013, 0x2016 }, |
| 183 | { 0x2018, 0x2019 }, { 0x201C, 0x201D }, { 0x2020, 0x2021 }, |
| 184 | { 0x2025, 0x2027 }, { 0x2030, 0x2030 }, { 0x2032, 0x2033 }, |
| 185 | { 0x2035, 0x2035 }, { 0x203B, 0x203B }, { 0x2074, 0x2074 }, |
| 186 | { 0x207F, 0x207F }, { 0x2081, 0x2084 }, { 0x20AC, 0x20AC }, |
| 187 | { 0x2103, 0x2103 }, { 0x2105, 0x2105 }, { 0x2109, 0x2109 }, |
| 188 | { 0x2113, 0x2113 }, { 0x2121, 0x2122 }, { 0x2126, 0x2126 }, |
| 189 | { 0x212B, 0x212B }, { 0x2154, 0x2155 }, { 0x215B, 0x215B }, |
| 190 | { 0x215E, 0x215E }, { 0x2160, 0x216B }, { 0x2170, 0x2179 }, |
| 191 | { 0x2190, 0x2199 }, { 0x21D2, 0x21D2 }, { 0x21D4, 0x21D4 }, |
| 192 | { 0x2200, 0x2200 }, { 0x2202, 0x2203 }, { 0x2207, 0x2208 }, |
| 193 | { 0x220B, 0x220B }, { 0x220F, 0x220F }, { 0x2211, 0x2211 }, |
| 194 | { 0x2215, 0x2215 }, { 0x221A, 0x221A }, { 0x221D, 0x2220 }, |
| 195 | { 0x2223, 0x2223 }, { 0x2225, 0x2225 }, { 0x2227, 0x222C }, |
| 196 | { 0x222E, 0x222E }, { 0x2234, 0x2237 }, { 0x223C, 0x223D }, |
| 197 | { 0x2248, 0x2248 }, { 0x224C, 0x224C }, { 0x2252, 0x2252 }, |
| 198 | { 0x2260, 0x2261 }, { 0x2264, 0x2267 }, { 0x226A, 0x226B }, |
| 199 | { 0x226E, 0x226F }, { 0x2282, 0x2283 }, { 0x2286, 0x2287 }, |
| 200 | { 0x2295, 0x2295 }, { 0x2299, 0x2299 }, { 0x22A5, 0x22A5 }, |
| 201 | { 0x22BF, 0x22BF }, { 0x2312, 0x2312 }, { 0x2460, 0x24BF }, |
| 202 | { 0x24D0, 0x24E9 }, { 0x2500, 0x254B }, { 0x2550, 0x2574 }, |
| 203 | { 0x2580, 0x258F }, { 0x2592, 0x2595 }, { 0x25A0, 0x25A1 }, |
| 204 | { 0x25A3, 0x25A9 }, { 0x25B2, 0x25B3 }, { 0x25B6, 0x25B7 }, |
| 205 | { 0x25BC, 0x25BD }, { 0x25C0, 0x25C1 }, { 0x25C6, 0x25C8 }, |
| 206 | { 0x25CB, 0x25CB }, { 0x25CE, 0x25D1 }, { 0x25E2, 0x25E5 }, |
| 207 | { 0x25EF, 0x25EF }, { 0x2605, 0x2606 }, { 0x2609, 0x2609 }, |
| 208 | { 0x260E, 0x260F }, { 0x261C, 0x261C }, { 0x261E, 0x261E }, |
| 209 | { 0x2640, 0x2640 }, { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, |
| 210 | { 0x2663, 0x2665 }, { 0x2667, 0x266A }, { 0x266C, 0x266D }, |
| 211 | { 0x266F, 0x266F }, { 0x300A, 0x300B }, { 0x301A, 0x301B }, |
| 212 | { 0xE000, 0xF8FF }, { 0xFFFD, 0xFFFD } |
| 213 | }; |
| 214 | |
| 215 | /* binary search in table of non-spacing characters */ |
| 216 | if (bisearch(ucs, ambiguous, |
| 217 | sizeof(ambiguous) / sizeof(struct interval) - 1)) |
| 218 | return 2; |
| 219 | |
| 220 | return wcwidth(ucs); |
| 221 | } |
| 222 | |
| 223 | |
| 224 | int wcswidth_cjk(const wchar_t *pwcs, size_t n) |
| 225 | { |
| 226 | int w, width = 0; |
| 227 | |
| 228 | for (;*pwcs && n-- > 0; pwcs++) |
| 229 | if ((w = wcwidth_cjk(*pwcs)) < 0) |
| 230 | return -1; |
| 231 | else |
| 232 | width += w; |
| 233 | |
| 234 | return width; |
| 235 | } |
| 236 | #endif |