1dae0a4e0SAdrien Destugues - PulkoMandy#include "vterm_internal.h"
2dae0a4e0SAdrien Destugues - PulkoMandy
3dae0a4e0SAdrien Destugues - PulkoMandy// ### The following from http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c
4dae0a4e0SAdrien Destugues - PulkoMandy// With modifications:
5dae0a4e0SAdrien Destugues - PulkoMandy//   made functions static
6dae0a4e0SAdrien Destugues - PulkoMandy//   moved 'combining' table to file scope, so other functions can see it
7dae0a4e0SAdrien Destugues - PulkoMandy// ###################################################################
8dae0a4e0SAdrien Destugues - PulkoMandy
9dae0a4e0SAdrien Destugues - PulkoMandy/*
10dae0a4e0SAdrien Destugues - PulkoMandy * This is an implementation of wcwidth() and wcswidth() (defined in
11dae0a4e0SAdrien Destugues - PulkoMandy * IEEE Std 1002.1-2001) for Unicode.
12dae0a4e0SAdrien Destugues - PulkoMandy *
13dae0a4e0SAdrien Destugues - PulkoMandy * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html
14dae0a4e0SAdrien Destugues - PulkoMandy * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html
15dae0a4e0SAdrien Destugues - PulkoMandy *
16dae0a4e0SAdrien Destugues - PulkoMandy * In fixed-width output devices, Latin characters all occupy a single
17dae0a4e0SAdrien Destugues - PulkoMandy * "cell" position of equal width, whereas ideographic CJK characters
18dae0a4e0SAdrien Destugues - PulkoMandy * occupy two such cells. Interoperability between terminal-line
19dae0a4e0SAdrien Destugues - PulkoMandy * applications and (teletype-style) character terminals using the
20dae0a4e0SAdrien Destugues - PulkoMandy * UTF-8 encoding requires agreement on which character should advance
21dae0a4e0SAdrien Destugues - PulkoMandy * the cursor by how many cell positions. No established formal
22dae0a4e0SAdrien Destugues - PulkoMandy * standards exist at present on which Unicode character shall occupy
23dae0a4e0SAdrien Destugues - PulkoMandy * how many cell positions on character terminals. These routines are
24dae0a4e0SAdrien Destugues - PulkoMandy * a first attempt of defining such behavior based on simple rules
25dae0a4e0SAdrien Destugues - PulkoMandy * applied to data provided by the Unicode Consortium.
26dae0a4e0SAdrien Destugues - PulkoMandy *
27dae0a4e0SAdrien Destugues - PulkoMandy * For some graphical characters, the Unicode standard explicitly
28dae0a4e0SAdrien Destugues - PulkoMandy * defines a character-cell width via the definition of the East Asian
29dae0a4e0SAdrien Destugues - PulkoMandy * FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes.
30dae0a4e0SAdrien Destugues - PulkoMandy * In all these cases, there is no ambiguity about which width a
31dae0a4e0SAdrien Destugues - PulkoMandy * terminal shall use. For characters in the East Asian Ambiguous (A)
32dae0a4e0SAdrien Destugues - PulkoMandy * class, the width choice depends purely on a preference of backward
33dae0a4e0SAdrien Destugues - PulkoMandy * compatibility with either historic CJK or Western practice.
34dae0a4e0SAdrien Destugues - PulkoMandy * Choosing single-width for these characters is easy to justify as
35dae0a4e0SAdrien Destugues - PulkoMandy * the appropriate long-term solution, as the CJK practice of
36dae0a4e0SAdrien Destugues - PulkoMandy * displaying these characters as double-width comes from historic
37dae0a4e0SAdrien Destugues - PulkoMandy * implementation simplicity (8-bit encoded characters were displayed
38dae0a4e0SAdrien Destugues - PulkoMandy * single-width and 16-bit ones double-width, even for Greek,
39dae0a4e0SAdrien Destugues - PulkoMandy * Cyrillic, etc.) and not any typographic considerations.
40dae0a4e0SAdrien Destugues - PulkoMandy *
41dae0a4e0SAdrien Destugues - PulkoMandy * Much less clear is the choice of width for the Not East Asian
42dae0a4e0SAdrien Destugues - PulkoMandy * (Neutral) class. Existing practice does not dictate a width for any
43dae0a4e0SAdrien Destugues - PulkoMandy * of these characters. It would nevertheless make sense
44dae0a4e0SAdrien Destugues - PulkoMandy * typographically to allocate two character cells to characters such
45dae0a4e0SAdrien Destugues - PulkoMandy * as for instance EM SPACE or VOLUME INTEGRAL, which cannot be
46dae0a4e0SAdrien Destugues - PulkoMandy * represented adequately with a single-width glyph. The following
47dae0a4e0SAdrien Destugues - PulkoMandy * routines at present merely assign a single-cell width to all
48dae0a4e0SAdrien Destugues - PulkoMandy * neutral characters, in the interest of simplicity. This is not
49dae0a4e0SAdrien Destugues - PulkoMandy * entirely satisfactory and should be reconsidered before
50dae0a4e0SAdrien Destugues - PulkoMandy * establishing a formal standard in this area. At the moment, the
51dae0a4e0SAdrien Destugues - PulkoMandy * decision which Not East Asian (Neutral) characters should be
52dae0a4e0SAdrien Destugues - PulkoMandy * represented by double-width glyphs cannot yet be answered by
53dae0a4e0SAdrien Destugues - PulkoMandy * applying a simple rule from the Unicode database content. Setting
54dae0a4e0SAdrien Destugues - PulkoMandy * up a proper standard for the behavior of UTF-8 character terminals
55dae0a4e0SAdrien Destugues - PulkoMandy * will require a careful analysis not only of each Unicode character,
56dae0a4e0SAdrien Destugues - PulkoMandy * but also of each presentation form, something the author of these
57dae0a4e0SAdrien Destugues - PulkoMandy * routines has avoided to do so far.
58dae0a4e0SAdrien Destugues - PulkoMandy *
59dae0a4e0SAdrien Destugues - PulkoMandy * http://www.unicode.org/unicode/reports/tr11/
60dae0a4e0SAdrien Destugues - PulkoMandy *
61dae0a4e0SAdrien Destugues - PulkoMandy * Markus Kuhn -- 2007-05-26 (Unicode 5.0)
62dae0a4e0SAdrien Destugues - PulkoMandy *
63dae0a4e0SAdrien Destugues - PulkoMandy * Permission to use, copy, modify, and distribute this software
64dae0a4e0SAdrien Destugues - PulkoMandy * for any purpose and without fee is hereby granted. The author
65dae0a4e0SAdrien Destugues - PulkoMandy * disclaims all warranties with regard to this software.
66dae0a4e0SAdrien Destugues - PulkoMandy *
67dae0a4e0SAdrien Destugues - PulkoMandy * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c
68dae0a4e0SAdrien Destugues - PulkoMandy */
69dae0a4e0SAdrien Destugues - PulkoMandy
70dae0a4e0SAdrien Destugues - PulkoMandy#include <wchar.h>
71dae0a4e0SAdrien Destugues - PulkoMandy
72dae0a4e0SAdrien Destugues - PulkoMandystruct interval {
73dae0a4e0SAdrien Destugues - PulkoMandy  int first;
74dae0a4e0SAdrien Destugues - PulkoMandy  int last;
75dae0a4e0SAdrien Destugues - PulkoMandy};
76dae0a4e0SAdrien Destugues - PulkoMandy
77dae0a4e0SAdrien Destugues - PulkoMandy/* sorted list of non-overlapping intervals of non-spacing characters */
78dae0a4e0SAdrien Destugues - PulkoMandy/* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */
79dae0a4e0SAdrien Destugues - PulkoMandystatic const struct interval combining[] = {
80dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0300, 0x036F }, { 0x0483, 0x0486 }, { 0x0488, 0x0489 },
81dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0591, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 },
82dae0a4e0SAdrien Destugues - PulkoMandy  { 0x05C4, 0x05C5 }, { 0x05C7, 0x05C7 }, { 0x0600, 0x0603 },
83dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0610, 0x0615 }, { 0x064B, 0x065E }, { 0x0670, 0x0670 },
84dae0a4e0SAdrien Destugues - PulkoMandy  { 0x06D6, 0x06E4 }, { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED },
85dae0a4e0SAdrien Destugues - PulkoMandy  { 0x070F, 0x070F }, { 0x0711, 0x0711 }, { 0x0730, 0x074A },
86dae0a4e0SAdrien Destugues - PulkoMandy  { 0x07A6, 0x07B0 }, { 0x07EB, 0x07F3 }, { 0x0901, 0x0902 },
87dae0a4e0SAdrien Destugues - PulkoMandy  { 0x093C, 0x093C }, { 0x0941, 0x0948 }, { 0x094D, 0x094D },
88dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0951, 0x0954 }, { 0x0962, 0x0963 }, { 0x0981, 0x0981 },
89dae0a4e0SAdrien Destugues - PulkoMandy  { 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 }, { 0x09CD, 0x09CD },
90dae0a4e0SAdrien Destugues - PulkoMandy  { 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 }, { 0x0A3C, 0x0A3C },
91dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 }, { 0x0A4B, 0x0A4D },
92dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 }, { 0x0ABC, 0x0ABC },
93dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 }, { 0x0ACD, 0x0ACD },
94dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 }, { 0x0B3C, 0x0B3C },
95dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 }, { 0x0B4D, 0x0B4D },
96dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 }, { 0x0BC0, 0x0BC0 },
97dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 }, { 0x0C46, 0x0C48 },
98dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 }, { 0x0CBC, 0x0CBC },
99dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 }, { 0x0CCC, 0x0CCD },
100dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0CE2, 0x0CE3 }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D },
101dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 },
102dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E },
103dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC },
104dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 },
105dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E },
106dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 },
107dae0a4e0SAdrien Destugues - PulkoMandy  { 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 },
108dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 },
109dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x135F, 0x135F },
110dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1712, 0x1714 }, { 0x1732, 0x1734 }, { 0x1752, 0x1753 },
111dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1772, 0x1773 }, { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD },
112dae0a4e0SAdrien Destugues - PulkoMandy  { 0x17C6, 0x17C6 }, { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD },
113dae0a4e0SAdrien Destugues - PulkoMandy  { 0x180B, 0x180D }, { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 },
114dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1927, 0x1928 }, { 0x1932, 0x1932 }, { 0x1939, 0x193B },
115dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1A17, 0x1A18 }, { 0x1B00, 0x1B03 }, { 0x1B34, 0x1B34 },
116dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1B36, 0x1B3A }, { 0x1B3C, 0x1B3C }, { 0x1B42, 0x1B42 },
117dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1B6B, 0x1B73 }, { 0x1DC0, 0x1DCA }, { 0x1DFE, 0x1DFF },
118dae0a4e0SAdrien Destugues - PulkoMandy  { 0x200B, 0x200F }, { 0x202A, 0x202E }, { 0x2060, 0x2063 },
119dae0a4e0SAdrien Destugues - PulkoMandy  { 0x206A, 0x206F }, { 0x20D0, 0x20EF }, { 0x302A, 0x302F },
120dae0a4e0SAdrien Destugues - PulkoMandy  { 0x3099, 0x309A }, { 0xA806, 0xA806 }, { 0xA80B, 0xA80B },
121dae0a4e0SAdrien Destugues - PulkoMandy  { 0xA825, 0xA826 }, { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F },
122dae0a4e0SAdrien Destugues - PulkoMandy  { 0xFE20, 0xFE23 }, { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB },
123dae0a4e0SAdrien Destugues - PulkoMandy  { 0x10A01, 0x10A03 }, { 0x10A05, 0x10A06 }, { 0x10A0C, 0x10A0F },
124dae0a4e0SAdrien Destugues - PulkoMandy  { 0x10A38, 0x10A3A }, { 0x10A3F, 0x10A3F }, { 0x1D167, 0x1D169 },
125dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B }, { 0x1D1AA, 0x1D1AD },
126dae0a4e0SAdrien Destugues - PulkoMandy  { 0x1D242, 0x1D244 }, { 0xE0001, 0xE0001 }, { 0xE0020, 0xE007F },
127dae0a4e0SAdrien Destugues - PulkoMandy  { 0xE0100, 0xE01EF }
128dae0a4e0SAdrien Destugues - PulkoMandy};
129dae0a4e0SAdrien Destugues - PulkoMandy
130dae0a4e0SAdrien Destugues - PulkoMandy
131dae0a4e0SAdrien Destugues - PulkoMandy/* auxiliary function for binary search in interval table */
132dae0a4e0SAdrien Destugues - PulkoMandystatic int bisearch(wchar_t ucs, const struct interval *table, int max) {
133dae0a4e0SAdrien Destugues - PulkoMandy  int min = 0;
134dae0a4e0SAdrien Destugues - PulkoMandy  int mid;
135dae0a4e0SAdrien Destugues - PulkoMandy
136dae0a4e0SAdrien Destugues - PulkoMandy  if (ucs < table[0].first || ucs > table[max].last)
137dae0a4e0SAdrien Destugues - PulkoMandy    return 0;
138dae0a4e0SAdrien Destugues - PulkoMandy  while (max >= min) {
139dae0a4e0SAdrien Destugues - PulkoMandy    mid = (min + max) / 2;
140dae0a4e0SAdrien Destugues - PulkoMandy    if (ucs > table[mid].last)
141dae0a4e0SAdrien Destugues - PulkoMandy      min = mid + 1;
142dae0a4e0SAdrien Destugues - PulkoMandy    else if (ucs < table[mid].first)
143dae0a4e0SAdrien Destugues - PulkoMandy      max = mid - 1;
144dae0a4e0SAdrien Destugues - PulkoMandy    else
145dae0a4e0SAdrien Destugues - PulkoMandy      return 1;
146dae0a4e0SAdrien Destugues - PulkoMandy  }
147dae0a4e0SAdrien Destugues - PulkoMandy
148dae0a4e0SAdrien Destugues - PulkoMandy  return 0;
149dae0a4e0SAdrien Destugues - PulkoMandy}
150dae0a4e0SAdrien Destugues - PulkoMandy
151dae0a4e0SAdrien Destugues - PulkoMandy
152dae0a4e0SAdrien Destugues - PulkoMandy/* The following two functions define the column width of an ISO 10646
153dae0a4e0SAdrien Destugues - PulkoMandy * character as follows:
154dae0a4e0SAdrien Destugues - PulkoMandy *
155dae0a4e0SAdrien Destugues - PulkoMandy *    - The null character (U+0000) has a column width of 0.
156dae0a4e0SAdrien Destugues - PulkoMandy *
157dae0a4e0SAdrien Destugues - PulkoMandy *    - Other C0/C1 control characters and DEL will lead to a return
158dae0a4e0SAdrien Destugues - PulkoMandy *      value of -1.
159dae0a4e0SAdrien Destugues - PulkoMandy *
160dae0a4e0SAdrien Destugues - PulkoMandy *    - Non-spacing and enclosing combining characters (general
161dae0a4e0SAdrien Destugues - PulkoMandy *      category code Mn or Me in the Unicode database) have a
162dae0a4e0SAdrien Destugues - PulkoMandy *      column width of 0.
163dae0a4e0SAdrien Destugues - PulkoMandy *
164dae0a4e0SAdrien Destugues - PulkoMandy *    - SOFT HYPHEN (U+00AD) has a column width of 1.
165dae0a4e0SAdrien Destugues - PulkoMandy *
166dae0a4e0SAdrien Destugues - PulkoMandy *    - Other format characters (general category code Cf in the Unicode
167dae0a4e0SAdrien Destugues - PulkoMandy *      database) and ZERO WIDTH SPACE (U+200B) have a column width of 0.
168dae0a4e0SAdrien Destugues - PulkoMandy *
169dae0a4e0SAdrien Destugues - PulkoMandy *    - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF)
170dae0a4e0SAdrien Destugues - PulkoMandy *      have a column width of 0.
171dae0a4e0SAdrien Destugues - PulkoMandy *
172dae0a4e0SAdrien Destugues - PulkoMandy *    - Spacing characters in the East Asian Wide (W) or East Asian
173dae0a4e0SAdrien Destugues - PulkoMandy *      Full-width (F) category as defined in Unicode Technical
174dae0a4e0SAdrien Destugues - PulkoMandy *      Report #11 have a column width of 2.
175dae0a4e0SAdrien Destugues - PulkoMandy *
176dae0a4e0SAdrien Destugues - PulkoMandy *    - All remaining characters (including all printable
177dae0a4e0SAdrien Destugues - PulkoMandy *      ISO 8859-1 and WGL4 characters, Unicode control characters,
178dae0a4e0SAdrien Destugues - PulkoMandy *      etc.) have a column width of 1.
179dae0a4e0SAdrien Destugues - PulkoMandy *
180dae0a4e0SAdrien Destugues - PulkoMandy * This implementation assumes that wchar_t characters are encoded
181dae0a4e0SAdrien Destugues - PulkoMandy * in ISO 10646.
182dae0a4e0SAdrien Destugues - PulkoMandy */
183dae0a4e0SAdrien Destugues - PulkoMandy
184dae0a4e0SAdrien Destugues - PulkoMandy
185dae0a4e0SAdrien Destugues - PulkoMandystatic int mk_wcwidth(wchar_t ucs)
186dae0a4e0SAdrien Destugues - PulkoMandy{
187dae0a4e0SAdrien Destugues - PulkoMandy  /* test for 8-bit control characters */
188dae0a4e0SAdrien Destugues - PulkoMandy  if (ucs == 0)
189dae0a4e0SAdrien Destugues - PulkoMandy    return 0;
190dae0a4e0SAdrien Destugues - PulkoMandy  if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0))
191dae0a4e0SAdrien Destugues - PulkoMandy    return -1;
192dae0a4e0SAdrien Destugues - PulkoMandy
193dae0a4e0SAdrien Destugues - PulkoMandy  /* binary search in table of non-spacing characters */
194dae0a4e0SAdrien Destugues - PulkoMandy  if (bisearch(ucs, combining,
195dae0a4e0SAdrien Destugues - PulkoMandy               sizeof(combining) / sizeof(struct interval) - 1))
196dae0a4e0SAdrien Destugues - PulkoMandy    return 0;
197dae0a4e0SAdrien Destugues - PulkoMandy
198dae0a4e0SAdrien Destugues - PulkoMandy  /* if we arrive here, ucs is not a combining or C0/C1 control character */
199dae0a4e0SAdrien Destugues - PulkoMandy
200dae0a4e0SAdrien Destugues - PulkoMandy  return 1 +
201dae0a4e0SAdrien Destugues - PulkoMandy    (ucs >= 0x1100 &&
202dae0a4e0SAdrien Destugues - PulkoMandy     (ucs <= 0x115f ||                    /* Hangul Jamo init. consonants */
203dae0a4e0SAdrien Destugues - PulkoMandy      ucs == 0x2329 || ucs == 0x232a ||
204dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0x2e80 && ucs <= 0xa4cf &&
205dae0a4e0SAdrien Destugues - PulkoMandy       ucs != 0x303f) ||                  /* CJK ... Yi */
206dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */
207dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */
208dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xfe10 && ucs <= 0xfe19) || /* Vertical forms */
209dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */
210dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */
211dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0xffe0 && ucs <= 0xffe6) ||
212dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0x20000 && ucs <= 0x2fffd) ||
213dae0a4e0SAdrien Destugues - PulkoMandy      (ucs >= 0x30000 && ucs <= 0x3fffd)));
214dae0a4e0SAdrien Destugues - PulkoMandy}
215dae0a4e0SAdrien Destugues - PulkoMandy
2161ec25517SAdrien Destugues
2176a545a8eSAdrien Destugues#if 0
218dae0a4e0SAdrien Destugues - PulkoMandystatic int mk_wcswidth(const wchar_t *pwcs, size_t n)
219dae0a4e0SAdrien Destugues - PulkoMandy{
220dae0a4e0SAdrien Destugues - PulkoMandy  int w, width = 0;
221dae0a4e0SAdrien Destugues - PulkoMandy
222dae0a4e0SAdrien Destugues - PulkoMandy  for (;*pwcs && n-- > 0; pwcs++)
223dae0a4e0SAdrien Destugues - PulkoMandy    if ((w = mk_wcwidth(*pwcs)) < 0)
224dae0a4e0SAdrien Destugues - PulkoMandy      return -1;
225dae0a4e0SAdrien Destugues - PulkoMandy    else
226dae0a4e0SAdrien Destugues - PulkoMandy      width += w;
227dae0a4e0SAdrien Destugues - PulkoMandy
228dae0a4e0SAdrien Destugues - PulkoMandy  return width;
229dae0a4e0SAdrien Destugues - PulkoMandy}
230dae0a4e0SAdrien Destugues - PulkoMandy
2311ec25517SAdrien Destugues
232dae0a4e0SAdrien Destugues - PulkoMandy/*
233dae0a4e0SAdrien Destugues - PulkoMandy * The following functions are the same as mk_wcwidth() and
234dae0a4e0SAdrien Destugues - PulkoMandy * mk_wcswidth(), except that spacing characters in the East Asian
235dae0a4e0SAdrien Destugues - PulkoMandy * Ambiguous (A) category as defined in Unicode Technical Report #11
236dae0a4e0SAdrien Destugues - PulkoMandy * have a column width of 2. This variant might be useful for users of
237dae0a4e0SAdrien Destugues - PulkoMandy * CJK legacy encodings who want to migrate to UCS without changing
238dae0a4e0SAdrien Destugues - PulkoMandy * the traditional terminal character-width behaviour. It is not
239dae0a4e0SAdrien Destugues - PulkoMandy * otherwise recommended for general use.
240dae0a4e0SAdrien Destugues - PulkoMandy */
241dae0a4e0SAdrien Destugues - PulkoMandystatic int mk_wcwidth_cjk(wchar_t ucs)
242dae0a4e0SAdrien Destugues - PulkoMandy{
243dae0a4e0SAdrien Destugues - PulkoMandy  /* sorted list of non-overlapping intervals of East Asian Ambiguous
244dae0a4e0SAdrien Destugues - PulkoMandy   * characters, generated by "uniset +WIDTH-A -cat=Me -cat=Mn -cat=Cf c" */
245dae0a4e0SAdrien Destugues - PulkoMandy  static const struct interval ambiguous[] = {
246dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 },
247dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00AA, 0x00AA }, { 0x00AE, 0x00AE }, { 0x00B0, 0x00B4 },
248dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 },
249dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 },
250dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED },
251dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA },
252dae0a4e0SAdrien Destugues - PulkoMandy    { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 },
253dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B },
254dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 },
255dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 },
256dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0148, 0x014B }, { 0x014D, 0x014D }, { 0x0152, 0x0153 },
257dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE },
258dae0a4e0SAdrien Destugues - PulkoMandy    { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 },
259dae0a4e0SAdrien Destugues - PulkoMandy    { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA },
260dae0a4e0SAdrien Destugues - PulkoMandy    { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 },
261dae0a4e0SAdrien Destugues - PulkoMandy    { 0x02C4, 0x02C4 }, { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB },
262dae0a4e0SAdrien Destugues - PulkoMandy    { 0x02CD, 0x02CD }, { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB },
263dae0a4e0SAdrien Destugues - PulkoMandy    { 0x02DD, 0x02DD }, { 0x02DF, 0x02DF }, { 0x0391, 0x03A1 },
264dae0a4e0SAdrien Destugues - PulkoMandy    { 0x03A3, 0x03A9 }, { 0x03B1, 0x03C1 }, { 0x03C3, 0x03C9 },
265dae0a4e0SAdrien Destugues - PulkoMandy    { 0x0401, 0x0401 }, { 0x0410, 0x044F }, { 0x0451, 0x0451 },
266dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2010, 0x2010 }, { 0x2013, 0x2016 }, { 0x2018, 0x2019 },
267dae0a4e0SAdrien Destugues - PulkoMandy    { 0x201C, 0x201D }, { 0x2020, 0x2022 }, { 0x2024, 0x2027 },
268dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2030, 0x2030 }, { 0x2032, 0x2033 }, { 0x2035, 0x2035 },
269dae0a4e0SAdrien Destugues - PulkoMandy    { 0x203B, 0x203B }, { 0x203E, 0x203E }, { 0x2074, 0x2074 },
270dae0a4e0SAdrien Destugues - PulkoMandy    { 0x207F, 0x207F }, { 0x2081, 0x2084 }, { 0x20AC, 0x20AC },
271dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2103, 0x2103 }, { 0x2105, 0x2105 }, { 0x2109, 0x2109 },
272dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2113, 0x2113 }, { 0x2116, 0x2116 }, { 0x2121, 0x2122 },
273dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2126, 0x2126 }, { 0x212B, 0x212B }, { 0x2153, 0x2154 },
274dae0a4e0SAdrien Destugues - PulkoMandy    { 0x215B, 0x215E }, { 0x2160, 0x216B }, { 0x2170, 0x2179 },
275dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2190, 0x2199 }, { 0x21B8, 0x21B9 }, { 0x21D2, 0x21D2 },
276dae0a4e0SAdrien Destugues - PulkoMandy    { 0x21D4, 0x21D4 }, { 0x21E7, 0x21E7 }, { 0x2200, 0x2200 },
277dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2202, 0x2203 }, { 0x2207, 0x2208 }, { 0x220B, 0x220B },
278dae0a4e0SAdrien Destugues - PulkoMandy    { 0x220F, 0x220F }, { 0x2211, 0x2211 }, { 0x2215, 0x2215 },
279dae0a4e0SAdrien Destugues - PulkoMandy    { 0x221A, 0x221A }, { 0x221D, 0x2220 }, { 0x2223, 0x2223 },
280dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2225, 0x2225 }, { 0x2227, 0x222C }, { 0x222E, 0x222E },
281dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2234, 0x2237 }, { 0x223C, 0x223D }, { 0x2248, 0x2248 },
282dae0a4e0SAdrien Destugues - PulkoMandy    { 0x224C, 0x224C }, { 0x2252, 0x2252 }, { 0x2260, 0x2261 },
283dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2264, 0x2267 }, { 0x226A, 0x226B }, { 0x226E, 0x226F },
284dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2282, 0x2283 }, { 0x2286, 0x2287 }, { 0x2295, 0x2295 },
285dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2299, 0x2299 }, { 0x22A5, 0x22A5 }, { 0x22BF, 0x22BF },
286dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2312, 0x2312 }, { 0x2460, 0x24E9 }, { 0x24EB, 0x254B },
287dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2550, 0x2573 }, { 0x2580, 0x258F }, { 0x2592, 0x2595 },
288dae0a4e0SAdrien Destugues - PulkoMandy    { 0x25A0, 0x25A1 }, { 0x25A3, 0x25A9 }, { 0x25B2, 0x25B3 },
289dae0a4e0SAdrien Destugues - PulkoMandy    { 0x25B6, 0x25B7 }, { 0x25BC, 0x25BD }, { 0x25C0, 0x25C1 },
290dae0a4e0SAdrien Destugues - PulkoMandy    { 0x25C6, 0x25C8 }, { 0x25CB, 0x25CB }, { 0x25CE, 0x25D1 },
291dae0a4e0SAdrien Destugues - PulkoMandy    { 0x25E2, 0x25E5 }, { 0x25EF, 0x25EF }, { 0x2605, 0x2606 },
292dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2609, 0x2609 }, { 0x260E, 0x260F }, { 0x2614, 0x2615 },
293dae0a4e0SAdrien Destugues - PulkoMandy    { 0x261C, 0x261C }, { 0x261E, 0x261E }, { 0x2640, 0x2640 },
294dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, { 0x2663, 0x2665 },
295dae0a4e0SAdrien Destugues - PulkoMandy    { 0x2667, 0x266A }, { 0x266C, 0x266D }, { 0x266F, 0x266F },
296dae0a4e0SAdrien Destugues - PulkoMandy    { 0x273D, 0x273D }, { 0x2776, 0x277F }, { 0xE000, 0xF8FF },
297dae0a4e0SAdrien Destugues - PulkoMandy    { 0xFFFD, 0xFFFD }, { 0xF0000, 0xFFFFD }, { 0x100000, 0x10FFFD }
298dae0a4e0SAdrien Destugues - PulkoMandy  };
299dae0a4e0SAdrien Destugues - PulkoMandy
300dae0a4e0SAdrien Destugues - PulkoMandy  /* binary search in table of non-spacing characters */
301dae0a4e0SAdrien Destugues - PulkoMandy  if (bisearch(ucs, ambiguous,
302dae0a4e0SAdrien Destugues - PulkoMandy               sizeof(ambiguous) / sizeof(struct interval) - 1))
303dae0a4e0SAdrien Destugues - PulkoMandy    return 2;
304dae0a4e0SAdrien Destugues - PulkoMandy
305dae0a4e0SAdrien Destugues - PulkoMandy  return mk_wcwidth(ucs);
306dae0a4e0SAdrien Destugues - PulkoMandy}
307dae0a4e0SAdrien Destugues - PulkoMandy
308dae0a4e0SAdrien Destugues - PulkoMandystatic int mk_wcswidth_cjk(const wchar_t *pwcs, size_t n)
309dae0a4e0SAdrien Destugues - PulkoMandy{
310dae0a4e0SAdrien Destugues - PulkoMandy  int w, width = 0;
311dae0a4e0SAdrien Destugues - PulkoMandy
312dae0a4e0SAdrien Destugues - PulkoMandy  for (;*pwcs && n-- > 0; pwcs++)
313dae0a4e0SAdrien Destugues - PulkoMandy    if ((w = mk_wcwidth_cjk(*pwcs)) < 0)
314dae0a4e0SAdrien Destugues - PulkoMandy      return -1;
315dae0a4e0SAdrien Destugues - PulkoMandy    else
316dae0a4e0SAdrien Destugues - PulkoMandy      width += w;
317dae0a4e0SAdrien Destugues - PulkoMandy
318dae0a4e0SAdrien Destugues - PulkoMandy  return width;
319dae0a4e0SAdrien Destugues - PulkoMandy}
3206a545a8eSAdrien Destugues#endif
321dae0a4e0SAdrien Destugues - PulkoMandy
322dae0a4e0SAdrien Destugues - PulkoMandy// ################################
323dae0a4e0SAdrien Destugues - PulkoMandy// ### The rest added by Paul Evans
324dae0a4e0SAdrien Destugues - PulkoMandy
3251ec25517SAdrien DestuguesINTERNAL int vterm_unicode_width(int codepoint)
326dae0a4e0SAdrien Destugues - PulkoMandy{
327dae0a4e0SAdrien Destugues - PulkoMandy  return mk_wcwidth(codepoint);
328dae0a4e0SAdrien Destugues - PulkoMandy}
329dae0a4e0SAdrien Destugues - PulkoMandy
3301ec25517SAdrien DestuguesINTERNAL int vterm_unicode_is_combining(int codepoint)
331dae0a4e0SAdrien Destugues - PulkoMandy{
332dae0a4e0SAdrien Destugues - PulkoMandy  return bisearch(codepoint, combining, sizeof(combining) / sizeof(struct interval) - 1);
333dae0a4e0SAdrien Destugues - PulkoMandy}
334