| 1 |
diff --git a/src/Makefile.in b/src/Makefile.in |
| 2 |
index d374ae83..08a26e65 100644 |
| 3 |
--- a/src/Makefile.in |
| 4 |
+++ b/src/Makefile.in |
| 5 |
@@ -40,7 +40,7 @@ COMMON = \ |
| 6 |
screen.o scrollbar.o scrollbar-next.o scrollbar-rxvt.o \ |
| 7 |
scrollbar-xterm.o scrollbar-plain.o xdefaults.o encoding.o \ |
| 8 |
rxvttoolkit.o rxvtutil.o keyboard.o rxvtimg.o \ |
| 9 |
- ev_cpp.o @PERL_O@ |
| 10 |
+ ev_cpp.o @PERL_O@ wcwidth.o |
| 11 |
|
| 12 |
COMMON_DAEMON = rxvtdaemon.o |
| 13 |
|
| 14 |
diff --git a/src/rxvt.h b/src/rxvt.h |
| 15 |
index 043b7c18..5bcfdb01 100644 |
| 16 |
--- a/src/rxvt.h |
| 17 |
+++ b/src/rxvt.h |
| 18 |
@@ -652,6 +652,12 @@ typedef struct _mwmhints |
| 19 |
// speed hack, copy some member variable into a local variable of the same name |
| 20 |
#define dLocal(type,name) type const name = this->name |
| 21 |
|
| 22 |
+int mk_wcwidth(wchar_t c); |
| 23 |
+int mk_wcswidth(const wchar_t *s, size_t n); |
| 24 |
+ |
| 25 |
+#define wcwidth mk_wcwidth |
| 26 |
+#define wcswidth mk_wcswidth |
| 27 |
+ |
| 28 |
// for speed reasons, we assume that all codepoints 32 to 126 are |
| 29 |
// single-width. |
| 30 |
#define WCWIDTH(c) (IN_RANGE_INC (c, 0x20, 0x7e) ? 1 : wcwidth (c)) |
| 31 |
diff --git a/src/wcwidth.C b/src/wcwidth.C |
| 32 |
new file mode 100644 |
| 33 |
index 00000000..c3d90a6a |
| 34 |
--- /dev/null |
| 35 |
+++ b/src/wcwidth.C |
| 36 |
@@ -0,0 +1,218 @@ |
| 37 |
+/* |
| 38 |
+ * This is an implementation of wcwidth() and wcswidth() (defined in |
| 39 |
+ * IEEE Std 1002.1-2001) for Unicode. |
| 40 |
+ * |
| 41 |
+ * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html |
| 42 |
+ * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html |
| 43 |
+ * |
| 44 |
+ * In fixed-width output devices, Latin characters all occupy a single |
| 45 |
+ * "cell" position of equal width, whereas ideographic CJK characters |
| 46 |
+ * occupy two such cells. Interoperability between terminal-line |
| 47 |
+ * applications and (teletype-style) character terminals using the |
| 48 |
+ * UTF-8 encoding requires agreement on which character should advance |
| 49 |
+ * the cursor by how many cell positions. No established formal |
| 50 |
+ * standards exist at present on which Unicode character shall occupy |
| 51 |
+ * how many cell positions on character terminals. These routines are |
| 52 |
+ * a first attempt of defining such behavior based on simple rules |
| 53 |
+ * applied to data provided by the Unicode Consortium. |
| 54 |
+ * |
| 55 |
+ * For some graphical characters, the Unicode standard explicitly |
| 56 |
+ * defines a character-cell width via the definition of the East Asian |
| 57 |
+ * FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes. |
| 58 |
+ * In all these cases, there is no ambiguity about which width a |
| 59 |
+ * terminal shall use. For characters in the East Asian Ambiguous (A) |
| 60 |
+ * class, the width choice depends purely on a preference of backward |
| 61 |
+ * compatibility with either historic CJK or Western practice. |
| 62 |
+ * Choosing single-width for these characters is easy to justify as |
| 63 |
+ * the appropriate long-term solution, as the CJK practice of |
| 64 |
+ * displaying these characters as double-width comes from historic |
| 65 |
+ * implementation simplicity (8-bit encoded characters were displayed |
| 66 |
+ * single-width and 16-bit ones double-width, even for Greek, |
| 67 |
+ * Cyrillic, etc.) and not any typographic considerations. |
| 68 |
+ * |
| 69 |
+ * Much less clear is the choice of width for the Not East Asian |
| 70 |
+ * (Neutral) class. Existing practice does not dictate a width for any |
| 71 |
+ * of these characters. It would nevertheless make sense |
| 72 |
+ * typographically to allocate two character cells to characters such |
| 73 |
+ * as for instance EM SPACE or VOLUME INTEGRAL, which cannot be |
| 74 |
+ * represented adequately with a single-width glyph. The following |
| 75 |
+ * routines at present merely assign a single-cell width to all |
| 76 |
+ * neutral characters, in the interest of simplicity. This is not |
| 77 |
+ * entirely satisfactory and should be reconsidered before |
| 78 |
+ * establishing a formal standard in this area. At the moment, the |
| 79 |
+ * decision which Not East Asian (Neutral) characters should be |
| 80 |
+ * represented by double-width glyphs cannot yet be answered by |
| 81 |
+ * applying a simple rule from the Unicode database content. Setting |
| 82 |
+ * up a proper standard for the behavior of UTF-8 character terminals |
| 83 |
+ * will require a careful analysis not only of each Unicode character, |
| 84 |
+ * but also of each presentation form, something the author of these |
| 85 |
+ * routines has avoided to do so far. |
| 86 |
+ * |
| 87 |
+ * http://www.unicode.org/unicode/reports/tr11/ |
| 88 |
+ * |
| 89 |
+ * Markus Kuhn -- 2007-05-26 (Unicode 5.0) |
| 90 |
+ * |
| 91 |
+ * Permission to use, copy, modify, and distribute this software |
| 92 |
+ * for any purpose and without fee is hereby granted. The author |
| 93 |
+ * disclaims all warranties with regard to this software. |
| 94 |
+ * |
| 95 |
+ * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c |
| 96 |
+ */ |
| 97 |
+ |
| 98 |
+#include <wchar.h> |
| 99 |
+ |
| 100 |
+struct interval { |
| 101 |
+ int first; |
| 102 |
+ int last; |
| 103 |
+}; |
| 104 |
+ |
| 105 |
+/* auxiliary function for binary search in interval table */ |
| 106 |
+static int bisearch(wchar_t ucs, const struct interval *table, int max) { |
| 107 |
+ int min = 0; |
| 108 |
+ int mid; |
| 109 |
+ |
| 110 |
+ if (ucs < table[0].first || ucs > table[max].last) |
| 111 |
+ return 0; |
| 112 |
+ while (max >= min) { |
| 113 |
+ mid = (min + max) / 2; |
| 114 |
+ if (ucs > table[mid].last) |
| 115 |
+ min = mid + 1; |
| 116 |
+ else if (ucs < table[mid].first) |
| 117 |
+ max = mid - 1; |
| 118 |
+ else |
| 119 |
+ return 1; |
| 120 |
+ } |
| 121 |
+ |
| 122 |
+ return 0; |
| 123 |
+} |
| 124 |
+ |
| 125 |
+ |
| 126 |
+/* The following two functions define the column width of an ISO 10646 |
| 127 |
+ * character as follows: |
| 128 |
+ * |
| 129 |
+ * - The null character (U+0000) has a column width of 0. |
| 130 |
+ * |
| 131 |
+ * - Other C0/C1 control characters and DEL will lead to a return |
| 132 |
+ * value of -1. |
| 133 |
+ * |
| 134 |
+ * - Non-spacing and enclosing combining characters (general |
| 135 |
+ * category code Mn or Me in the Unicode database) have a |
| 136 |
+ * column width of 0. |
| 137 |
+ * |
| 138 |
+ * - SOFT HYPHEN (U+00AD) has a column width of 1. |
| 139 |
+ * |
| 140 |
+ * - Other format characters (general category code Cf in the Unicode |
| 141 |
+ * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0. |
| 142 |
+ * |
| 143 |
+ * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF) |
| 144 |
+ * have a column width of 0. |
| 145 |
+ * |
| 146 |
+ * - Spacing characters in the East Asian Wide (W) or East Asian |
| 147 |
+ * Full-width (F) category as defined in Unicode Technical |
| 148 |
+ * Report #11 have a column width of 2. |
| 149 |
+ * |
| 150 |
+ * - All remaining characters (including all printable |
| 151 |
+ * ISO 8859-1 and WGL4 characters, Unicode control characters, |
| 152 |
+ * etc.) have a column width of 1. |
| 153 |
+ * |
| 154 |
+ * This implementation assumes that wchar_t characters are encoded |
| 155 |
+ * in ISO 10646. |
| 156 |
+ */ |
| 157 |
+ |
| 158 |
+int mk_wcwidth(wchar_t ucs) |
| 159 |
+{ |
| 160 |
+ /* sorted list of non-overlapping intervals of non-spacing characters */ |
| 161 |
+ /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */ |
| 162 |
+ static const struct interval combining[] = { |
| 163 |
+ { 0x0300, 0x036F }, { 0x0483, 0x0486 }, { 0x0488, 0x0489 }, |
| 164 |
+ { 0x0591, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 }, |
| 165 |
+ { 0x05C4, 0x05C5 }, { 0x05C7, 0x05C7 }, { 0x0600, 0x0603 }, |
| 166 |
+ { 0x0610, 0x0615 }, { 0x064B, 0x065E }, { 0x0670, 0x0670 }, |
| 167 |
+ { 0x06D6, 0x06E4 }, { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, |
| 168 |
+ { 0x070F, 0x070F }, { 0x0711, 0x0711 }, { 0x0730, 0x074A }, |
| 169 |
+ { 0x07A6, 0x07B0 }, { 0x07EB, 0x07F3 }, { 0x0901, 0x0902 }, |
| 170 |
+ { 0x093C, 0x093C }, { 0x0941, 0x0948 }, { 0x094D, 0x094D }, |
| 171 |
+ { 0x0951, 0x0954 }, { 0x0962, 0x0963 }, { 0x0981, 0x0981 }, |
| 172 |
+ { 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 }, { 0x09CD, 0x09CD }, |
| 173 |
+ { 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 }, { 0x0A3C, 0x0A3C }, |
| 174 |
+ { 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 }, { 0x0A4B, 0x0A4D }, |
| 175 |
+ { 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 }, { 0x0ABC, 0x0ABC }, |
| 176 |
+ { 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 }, { 0x0ACD, 0x0ACD }, |
| 177 |
+ { 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 }, { 0x0B3C, 0x0B3C }, |
| 178 |
+ { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 }, { 0x0B4D, 0x0B4D }, |
| 179 |
+ { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 }, { 0x0BC0, 0x0BC0 }, |
| 180 |
+ { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 }, { 0x0C46, 0x0C48 }, |
| 181 |
+ { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 }, { 0x0CBC, 0x0CBC }, |
| 182 |
+ { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 }, { 0x0CCC, 0x0CCD }, |
| 183 |
+ { 0x0CE2, 0x0CE3 }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D }, |
| 184 |
+ { 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 }, |
| 185 |
+ { 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E }, |
| 186 |
+ { 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC }, |
| 187 |
+ { 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 }, |
| 188 |
+ { 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E }, |
| 189 |
+ { 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 }, |
| 190 |
+ { 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 }, |
| 191 |
+ { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 }, |
| 192 |
+ { 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x135F, 0x135F }, |
| 193 |
+ { 0x1712, 0x1714 }, { 0x1732, 0x1734 }, { 0x1752, 0x1753 }, |
| 194 |
+ { 0x1772, 0x1773 }, { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD }, |
| 195 |
+ { 0x17C6, 0x17C6 }, { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD }, |
| 196 |
+ { 0x180B, 0x180D }, { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 }, |
| 197 |
+ { 0x1927, 0x1928 }, { 0x1932, 0x1932 }, { 0x1939, 0x193B }, |
| 198 |
+ { 0x1A17, 0x1A18 }, { 0x1B00, 0x1B03 }, { 0x1B34, 0x1B34 }, |
| 199 |
+ { 0x1B36, 0x1B3A }, { 0x1B3C, 0x1B3C }, { 0x1B42, 0x1B42 }, |
| 200 |
+ { 0x1B6B, 0x1B73 }, { 0x1DC0, 0x1DCA }, { 0x1DFE, 0x1DFF }, |
| 201 |
+ { 0x200B, 0x200F }, { 0x202A, 0x202E }, { 0x2060, 0x2063 }, |
| 202 |
+ { 0x206A, 0x206F }, { 0x20D0, 0x20EF }, { 0x302A, 0x302F }, |
| 203 |
+ { 0x3099, 0x309A }, { 0xA806, 0xA806 }, { 0xA80B, 0xA80B }, |
| 204 |
+ { 0xA825, 0xA826 }, { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F }, |
| 205 |
+ { 0xFE20, 0xFE23 }, { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB }, |
| 206 |
+ { 0x10A01, 0x10A03 }, { 0x10A05, 0x10A06 }, { 0x10A0C, 0x10A0F }, |
| 207 |
+ { 0x10A38, 0x10A3A }, { 0x10A3F, 0x10A3F }, { 0x1D167, 0x1D169 }, |
| 208 |
+ { 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B }, { 0x1D1AA, 0x1D1AD }, |
| 209 |
+ { 0x1D242, 0x1D244 }, { 0xE0001, 0xE0001 }, { 0xE0020, 0xE007F }, |
| 210 |
+ { 0xE0100, 0xE01EF } |
| 211 |
+ }; |
| 212 |
+ |
| 213 |
+ /* test for 8-bit control characters */ |
| 214 |
+ if (ucs == 0) |
| 215 |
+ return 0; |
| 216 |
+ if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0)) |
| 217 |
+ return -1; |
| 218 |
+ |
| 219 |
+ /* binary search in table of non-spacing characters */ |
| 220 |
+ if (bisearch(ucs, combining, |
| 221 |
+ sizeof(combining) / sizeof(struct interval) - 1)) |
| 222 |
+ return 0; |
| 223 |
+ |
| 224 |
+ /* if we arrive here, ucs is not a combining or C0/C1 control character */ |
| 225 |
+ |
| 226 |
+ return 1 + |
| 227 |
+ (ucs >= 0x1100 && |
| 228 |
+ (ucs <= 0x115f || /* Hangul Jamo init. consonants */ |
| 229 |
+ ucs == 0x2329 || ucs == 0x232a || |
| 230 |
+ (ucs >= 0x2e80 && ucs <= 0xa4cf && |
| 231 |
+ ucs != 0x303f) || /* CJK ... Yi */ |
| 232 |
+ (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */ |
| 233 |
+ (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */ |
| 234 |
+ (ucs >= 0xfe10 && ucs <= 0xfe19) || /* Vertical forms */ |
| 235 |
+ (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */ |
| 236 |
+ (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */ |
| 237 |
+ (ucs >= 0xffe0 && ucs <= 0xffe6) || |
| 238 |
+ (ucs >= 0x20000 && ucs <= 0x2fffd) || |
| 239 |
+ (ucs >= 0x30000 && ucs <= 0x3fffd))); |
| 240 |
+} |
| 241 |
+ |
| 242 |
+ |
| 243 |
+int mk_wcswidth(const wchar_t *pwcs, size_t n) |
| 244 |
+{ |
| 245 |
+ int w, width = 0; |
| 246 |
+ |
| 247 |
+ for (;*pwcs && n-- > 0; pwcs++) |
| 248 |
+ if ((w = mk_wcwidth(*pwcs)) < 0) |
| 249 |
+ return -1; |
| 250 |
+ else |
| 251 |
+ width += w; |
| 252 |
+ |
| 253 |
+ return width; |
| 254 |
+} |
