2 * This is an implementation of wcwidth() and wcswidth() (defined in
3 * IEEE Std 1002.1-2001) for Unicode.
5 * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html
6 * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html
8 * In fixed-width output devices, Latin characters all occupy a single
9 * "cell" position of equal width, whereas ideographic CJK characters
10 * occupy two such cells. Interoperability between terminal-line
11 * applications and (teletype-style) character terminals using the
12 * UTF-8 encoding requires agreement on which character should advance
13 * the cursor by how many cell positions. No established formal
14 * standards exist at present on which Unicode character shall occupy
15 * how many cell positions on character terminals. These routines are
16 * a first attempt of defining such behavior based on simple rules
17 * applied to data provided by the Unicode Consortium.
19 * For some graphical characters, the Unicode standard explicitly
20 * defines a character-cell width via the definition of the East Asian
21 * FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes.
22 * In all these cases, there is no ambiguity about which width a
23 * terminal shall use. For characters in the East Asian Ambiguous (A)
24 * class, the width choice depends purely on a preference of backward
25 * compatibility with either historic CJK or Western practice.
26 * Choosing single-width for these characters is easy to justify as
27 * the appropriate long-term solution, as the CJK practice of
28 * displaying these characters as double-width comes from historic
29 * implementation simplicity (8-bit encoded characters were displayed
30 * single-width and 16-bit ones double-width, even for Greek,
31 * Cyrillic, etc.) and not any typographic considerations.
33 * Much less clear is the choice of width for the Not East Asian
34 * (Neutral) class. Existing practice does not dictate a width for any
35 * of these characters. It would nevertheless make sense
36 * typographically to allocate two character cells to characters such
37 * as for instance EM SPACE or VOLUME INTEGRAL, which cannot be
38 * represented adequately with a single-width glyph. The following
39 * routines at present merely assign a single-cell width to all
40 * neutral characters, in the interest of simplicity. This is not
41 * entirely satisfactory and should be reconsidered before
42 * establishing a formal standard in this area. At the moment, the
43 * decision which Not East Asian (Neutral) characters should be
44 * represented by double-width glyphs cannot yet be answered by
45 * applying a simple rule from the Unicode database content. Setting
46 * up a proper standard for the behavior of UTF-8 character terminals
47 * will require a careful analysis not only of each Unicode character,
48 * but also of each presentation form, something the author of these
49 * routines has avoided to do so far.
51 * http://www.unicode.org/unicode/reports/tr11/
53 * Markus Kuhn -- 2003-05-20 (Unicode 4.0)
55 * Permission to use, copy, modify, and distribute this software
56 * for any purpose and without fee is hereby granted. The author
57 * disclaims all warranties with regard to this software.
59 * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c
78 /* auxiliary function for binary search in interval table */
79 static int bisearch (wchar_t ucs, const struct interval *table, int max)
84 if (ucs < table[0].first || ucs > table[max].last)
87 mid = (min + max) / 2;
88 if (ucs > table[mid].last)
90 else if (ucs < table[mid].first)
100 /* The following two functions define the column width of an ISO 10646
101 * character as follows:
103 * - The null character (U+0000) has a column width of 0.
105 * - Other C0/C1 control characters and DEL will lead to a return
108 * - Non-spacing and enclosing combining characters (general
109 * category code Mn or Me in the Unicode database) have a
112 * - SOFT HYPHEN (U+00AD) has a column width of 1.
114 * - Other format characters (general category code Cf in the Unicode
115 * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0.
117 * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF)
118 * have a column width of 0.
120 * - Spacing characters in the East Asian Wide (W) or East Asian
121 * Full-width (F) category as defined in Unicode Technical
122 * Report #11 have a column width of 2.
124 * - All remaining characters (including all printable
125 * ISO 8859-1 and WGL4 characters, Unicode control characters,
126 * etc.) have a column width of 1.
128 * This implementation assumes that wchar_t characters are encoded
132 int wcwidth_ucs (wchar_t ucs)
134 /* sorted list of non-overlapping intervals of non-spacing characters */
135 /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */
136 static const struct interval combining[] = {
137 {0x0300, 0x0357}, {0x035D, 0x036F}, {0x0483, 0x0486},
138 {0x0488, 0x0489}, {0x0591, 0x05A1}, {0x05A3, 0x05B9},
139 {0x05BB, 0x05BD}, {0x05BF, 0x05BF}, {0x05C1, 0x05C2},
140 {0x05C4, 0x05C4}, {0x0600, 0x0603}, {0x0610, 0x0615},
141 {0x064B, 0x0658}, {0x0670, 0x0670}, {0x06D6, 0x06E4},
142 {0x06E7, 0x06E8}, {0x06EA, 0x06ED}, {0x070F, 0x070F},
143 {0x0711, 0x0711}, {0x0730, 0x074A}, {0x07A6, 0x07B0},
144 {0x0901, 0x0902}, {0x093C, 0x093C}, {0x0941, 0x0948},
145 {0x094D, 0x094D}, {0x0951, 0x0954}, {0x0962, 0x0963},
146 {0x0981, 0x0981}, {0x09BC, 0x09BC}, {0x09C1, 0x09C4},
147 {0x09CD, 0x09CD}, {0x09E2, 0x09E3}, {0x0A01, 0x0A02},
148 {0x0A3C, 0x0A3C}, {0x0A41, 0x0A42}, {0x0A47, 0x0A48},
149 {0x0A4B, 0x0A4D}, {0x0A70, 0x0A71}, {0x0A81, 0x0A82},
150 {0x0ABC, 0x0ABC}, {0x0AC1, 0x0AC5}, {0x0AC7, 0x0AC8},
151 {0x0ACD, 0x0ACD}, {0x0AE2, 0x0AE3}, {0x0B01, 0x0B01},
152 {0x0B3C, 0x0B3C}, {0x0B3F, 0x0B3F}, {0x0B41, 0x0B43},
153 {0x0B4D, 0x0B4D}, {0x0B56, 0x0B56}, {0x0B82, 0x0B82},
154 {0x0BC0, 0x0BC0}, {0x0BCD, 0x0BCD}, {0x0C3E, 0x0C40},
155 {0x0C46, 0x0C48}, {0x0C4A, 0x0C4D}, {0x0C55, 0x0C56},
156 {0x0CBC, 0x0CBC}, {0x0CBF, 0x0CBF}, {0x0CC6, 0x0CC6},
157 {0x0CCC, 0x0CCD}, {0x0D41, 0x0D43}, {0x0D4D, 0x0D4D},
158 {0x0DCA, 0x0DCA}, {0x0DD2, 0x0DD4}, {0x0DD6, 0x0DD6},
159 {0x0E31, 0x0E31}, {0x0E34, 0x0E3A}, {0x0E47, 0x0E4E},
160 {0x0EB1, 0x0EB1}, {0x0EB4, 0x0EB9}, {0x0EBB, 0x0EBC},
161 {0x0EC8, 0x0ECD}, {0x0F18, 0x0F19}, {0x0F35, 0x0F35},
162 {0x0F37, 0x0F37}, {0x0F39, 0x0F39}, {0x0F71, 0x0F7E},
163 {0x0F80, 0x0F84}, {0x0F86, 0x0F87}, {0x0F90, 0x0F97},
164 {0x0F99, 0x0FBC}, {0x0FC6, 0x0FC6}, {0x102D, 0x1030},
165 {0x1032, 0x1032}, {0x1036, 0x1037}, {0x1039, 0x1039},
166 {0x1058, 0x1059}, {0x1160, 0x11FF}, {0x1712, 0x1714},
167 {0x1732, 0x1734}, {0x1752, 0x1753}, {0x1772, 0x1773},
168 {0x17B4, 0x17B5}, {0x17B7, 0x17BD}, {0x17C6, 0x17C6},
169 {0x17C9, 0x17D3}, {0x17DD, 0x17DD}, {0x180B, 0x180D},
170 {0x18A9, 0x18A9}, {0x1920, 0x1922}, {0x1927, 0x1928},
171 {0x1932, 0x1932}, {0x1939, 0x193B}, {0x200B, 0x200F},
172 {0x202A, 0x202E}, {0x2060, 0x2063}, {0x206A, 0x206F},
173 {0x20D0, 0x20EA}, {0x302A, 0x302F}, {0x3099, 0x309A},
174 {0xFB1E, 0xFB1E}, {0xFE00, 0xFE0F}, {0xFE20, 0xFE23},
175 {0xFEFF, 0xFEFF}, {0xFFF9, 0xFFFB}, {0x1D167, 0x1D169},
176 {0x1D173, 0x1D182}, {0x1D185, 0x1D18B}, {0x1D1AA, 0x1D1AD},
177 {0xE0001, 0xE0001}, {0xE0020, 0xE007F}, {0xE0100, 0xE01EF}
180 /* test for 8-bit control characters */
183 if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0))
186 /* binary search in table of non-spacing characters */
187 if (bisearch (ucs, combining,
188 sizeof (combining) / sizeof (struct interval) - 1))
191 /* if we arrive here, ucs is not a combining or C0/C1 control character */
193 return 1 + (ucs >= 0x1100 && (ucs <= 0x115f || /* Hangul Jamo init. consonants */
194 ucs == 0x2329 || ucs == 0x232a || (ucs >= 0x2e80 && ucs <= 0xa4cf && ucs != 0x303f) || /* CJK ... Yi */
195 (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */
196 (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */
197 (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */
198 (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */
199 (ucs >= 0xffe0 && ucs <= 0xffe6) ||
200 (ucs >= 0x20000 && ucs <= 0x2fffd) ||
201 (ucs >= 0x30000 && ucs <= 0x3fffd)));
204 #endif /* !HAVE_WC_FUNCS */