1
Subject: Solaris 9 patch
2
From: Ewgenij Starostin <estar@cs.tu-berlin.de>
3
Date: Wed, 22 Jun 2005 15:30:22 +0200
7
this is a patch to compile rxvt-unicode-5.5 on Solaris 9. Changes:
8
- wcwidth() replaced by mk_wcwidth()
9
- workaround for bug in <iso/wchar_iso.h>
14
This patch will probably break things on other systems.
16
diff -Naur rxvt-unicode-5.5/src/Makefile.in rxvt-unicode-5.5-mod/src/Makefile.in
17
--- rxvt-unicode-5.5/src/Makefile.in 2005-02-16 22:35:00.000000000 +0100
18
+++ rxvt-unicode-5.5-mod/src/Makefile.in 2005-06-22 14:47:36.237424000 +0200
23
- command.o rxvtfont.o init.o logging.o \
24
+ wcwidth.o command.o rxvtfont.o init.o logging.o \
25
main.o menubar.o misc.o netdisp.o ptytty.o screen.o \
26
scrollbar.o scrollbar-next.o scrollbar-rxvt.o scrollbar-xterm.o scrollbar-plain.o \
27
strings.o xdefaults.o xpm.o encoding.o rxvttoolkit.o rxvtutil.o iom.o keyboard.o
28
diff -Naur rxvt-unicode-5.5/src/rxvtlib.h.in rxvt-unicode-5.5-mod/src/rxvtlib.h.in
29
--- rxvt-unicode-5.5/src/rxvtlib.h.in 2005-04-17 21:46:40.000000000 +0200
30
+++ rxvt-unicode-5.5-mod/src/rxvtlib.h.in 2005-06-22 14:53:57.187365000 +0200
35
+/* work around problem in Solaris 9 wchar_iso.h */
37
+#define BCK_XOPEN_SOURCE _XOPEN_SOURCE
40
+/* ... now do the includes ... */
46
+/* now put it all back like it was before */
47
+#ifdef BCK_XOPEN_SOURCE
48
+#define _XOPEN_SOURCE BCK_XOPEN_SOURCE
49
+#undef BCK_XOPEN_SOURCE
51
+/* end of workaround */
52
// stdlib.h might provide it
54
+/* avoid buggy Solaris 9 wcwidth() using Markus's mk_wcwidth() */
55
+int mk_wcwidth( wchar_t c );
56
+# define wcwidth(x) mk_wcwidth(x)
60
diff -Naur rxvt-unicode-5.5/src/wcwidth.C rxvt-unicode-5.5-mod/src/wcwidth.C
61
--- rxvt-unicode-5.5/src/wcwidth.C 1970-01-01 01:00:00.000000000 +0100
62
+++ rxvt-unicode-5.5-mod/src/wcwidth.C 2005-06-22 14:47:21.289766000 +0200
65
+ * This is an implementation of wcwidth() and wcswidth() (defined in
66
+ * IEEE Std 1002.1-2001) for Unicode.
68
+ * http://www.opengroup.org/onlinepubs/007904975/functions/wcwidth.html
69
+ * http://www.opengroup.org/onlinepubs/007904975/functions/wcswidth.html
71
+ * In fixed-width output devices, Latin characters all occupy a single
72
+ * "cell" position of equal width, whereas ideographic CJK characters
73
+ * occupy two such cells. Interoperability between terminal-line
74
+ * applications and (teletype-style) character terminals using the
75
+ * UTF-8 encoding requires agreement on which character should advance
76
+ * the cursor by how many cell positions. No established formal
77
+ * standards exist at present on which Unicode character shall occupy
78
+ * how many cell positions on character terminals. These routines are
79
+ * a first attempt of defining such behavior based on simple rules
80
+ * applied to data provided by the Unicode Consortium.
82
+ * For some graphical characters, the Unicode standard explicitly
83
+ * defines a character-cell width via the definition of the East Asian
84
+ * FullWidth (F), Wide (W), Half-width (H), and Narrow (Na) classes.
85
+ * In all these cases, there is no ambiguity about which width a
86
+ * terminal shall use. For characters in the East Asian Ambiguous (A)
87
+ * class, the width choice depends purely on a preference of backward
88
+ * compatibility with either historic CJK or Western practice.
89
+ * Choosing single-width for these characters is easy to justify as
90
+ * the appropriate long-term solution, as the CJK practice of
91
+ * displaying these characters as double-width comes from historic
92
+ * implementation simplicity (8-bit encoded characters were displayed
93
+ * single-width and 16-bit ones double-width, even for Greek,
94
+ * Cyrillic, etc.) and not any typographic considerations.
96
+ * Much less clear is the choice of width for the Not East Asian
97
+ * (Neutral) class. Existing practice does not dictate a width for any
98
+ * of these characters. It would nevertheless make sense
99
+ * typographically to allocate two character cells to characters such
100
+ * as for instance EM SPACE or VOLUME INTEGRAL, which cannot be
101
+ * represented adequately with a single-width glyph. The following
102
+ * routines at present merely assign a single-cell width to all
103
+ * neutral characters, in the interest of simplicity. This is not
104
+ * entirely satisfactory and should be reconsidered before
105
+ * establishing a formal standard in this area. At the moment, the
106
+ * decision which Not East Asian (Neutral) characters should be
107
+ * represented by double-width glyphs cannot yet be answered by
108
+ * applying a simple rule from the Unicode database content. Setting
109
+ * up a proper standard for the behavior of UTF-8 character terminals
110
+ * will require a careful analysis not only of each Unicode character,
111
+ * but also of each presentation form, something the author of these
112
+ * routines has avoided to do so far.
114
+ * http://www.unicode.org/unicode/reports/tr11/
116
+ * Markus Kuhn -- 2003-05-20 (Unicode 4.0)
118
+ * Permission to use, copy, modify, and distribute this software
119
+ * for any purpose and without fee is hereby granted. The author
120
+ * disclaims all warranties with regard to this software.
122
+ * Latest version: http://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c
132
+/* auxiliary function for binary search in interval table */
133
+static int bisearch(wchar_t ucs, const struct interval *table, int max) {
137
+ if (ucs < table[0].first || ucs > table[max].last)
139
+ while (max >= min) {
140
+ mid = (min + max) / 2;
141
+ if (ucs > table[mid].last)
143
+ else if (ucs < table[mid].first)
153
+/* The following two functions define the column width of an ISO 10646
154
+ * character as follows:
156
+ * - The null character (U+0000) has a column width of 0.
158
+ * - Other C0/C1 control characters and DEL will lead to a return
161
+ * - Non-spacing and enclosing combining characters (general
162
+ * category code Mn or Me in the Unicode database) have a
163
+ * column width of 0.
165
+ * - SOFT HYPHEN (U+00AD) has a column width of 1.
167
+ * - Other format characters (general category code Cf in the Unicode
168
+ * database) and ZERO WIDTH SPACE (U+200B) have a column width of 0.
170
+ * - Hangul Jamo medial vowels and final consonants (U+1160-U+11FF)
171
+ * have a column width of 0.
173
+ * - Spacing characters in the East Asian Wide (W) or East Asian
174
+ * Full-width (F) category as defined in Unicode Technical
175
+ * Report #11 have a column width of 2.
177
+ * - All remaining characters (including all printable
178
+ * ISO 8859-1 and WGL4 characters, Unicode control characters,
179
+ * etc.) have a column width of 1.
181
+ * This implementation assumes that wchar_t characters are encoded
185
+int mk_wcwidth(wchar_t ucs)
187
+ /* sorted list of non-overlapping intervals of non-spacing characters */
188
+ /* generated by "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c" */
189
+ static const struct interval combining[] = {
190
+ { 0x0300, 0x0357 }, { 0x035D, 0x036F }, { 0x0483, 0x0486 },
191
+ { 0x0488, 0x0489 }, { 0x0591, 0x05A1 }, { 0x05A3, 0x05B9 },
192
+ { 0x05BB, 0x05BD }, { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 },
193
+ { 0x05C4, 0x05C4 }, { 0x0600, 0x0603 }, { 0x0610, 0x0615 },
194
+ { 0x064B, 0x0658 }, { 0x0670, 0x0670 }, { 0x06D6, 0x06E4 },
195
+ { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, { 0x070F, 0x070F },
196
+ { 0x0711, 0x0711 }, { 0x0730, 0x074A }, { 0x07A6, 0x07B0 },
197
+ { 0x0901, 0x0902 }, { 0x093C, 0x093C }, { 0x0941, 0x0948 },
198
+ { 0x094D, 0x094D }, { 0x0951, 0x0954 }, { 0x0962, 0x0963 },
199
+ { 0x0981, 0x0981 }, { 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 },
200
+ { 0x09CD, 0x09CD }, { 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 },
201
+ { 0x0A3C, 0x0A3C }, { 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 },
202
+ { 0x0A4B, 0x0A4D }, { 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 },
203
+ { 0x0ABC, 0x0ABC }, { 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 },
204
+ { 0x0ACD, 0x0ACD }, { 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 },
205
+ { 0x0B3C, 0x0B3C }, { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 },
206
+ { 0x0B4D, 0x0B4D }, { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 },
207
+ { 0x0BC0, 0x0BC0 }, { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 },
208
+ { 0x0C46, 0x0C48 }, { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 },
209
+ { 0x0CBC, 0x0CBC }, { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 },
210
+ { 0x0CCC, 0x0CCD }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D },
211
+ { 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 },
212
+ { 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E },
213
+ { 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC },
214
+ { 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 },
215
+ { 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E },
216
+ { 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 },
217
+ { 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 },
218
+ { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 },
219
+ { 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x1712, 0x1714 },
220
+ { 0x1732, 0x1734 }, { 0x1752, 0x1753 }, { 0x1772, 0x1773 },
221
+ { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD }, { 0x17C6, 0x17C6 },
222
+ { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD }, { 0x180B, 0x180D },
223
+ { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 }, { 0x1927, 0x1928 },
224
+ { 0x1932, 0x1932 }, { 0x1939, 0x193B }, { 0x200B, 0x200F },
225
+ { 0x202A, 0x202E }, { 0x2060, 0x2063 }, { 0x206A, 0x206F },
226
+ { 0x20D0, 0x20EA }, { 0x302A, 0x302F }, { 0x3099, 0x309A },
227
+ { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F }, { 0xFE20, 0xFE23 },
228
+ { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB }, { 0x1D167, 0x1D169 },
229
+ { 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B }, { 0x1D1AA, 0x1D1AD },
230
+ { 0xE0001, 0xE0001 }, { 0xE0020, 0xE007F }, { 0xE0100, 0xE01EF }
233
+ /* test for 8-bit control characters */
236
+ if (ucs < 32 || (ucs >= 0x7f && ucs < 0xa0))
239
+ /* binary search in table of non-spacing characters */
240
+ if (bisearch(ucs, combining,
241
+ sizeof(combining) / sizeof(struct interval) - 1))
244
+ /* if we arrive here, ucs is not a combining or C0/C1 control character */
248
+ (ucs <= 0x115f || /* Hangul Jamo init. consonants */
249
+ ucs == 0x2329 || ucs == 0x232a ||
250
+ (ucs >= 0x2e80 && ucs <= 0xa4cf &&
251
+ ucs != 0x303f) || /* CJK ... Yi */
252
+ (ucs >= 0xac00 && ucs <= 0xd7a3) || /* Hangul Syllables */
253
+ (ucs >= 0xf900 && ucs <= 0xfaff) || /* CJK Compatibility Ideographs */
254
+ (ucs >= 0xfe30 && ucs <= 0xfe6f) || /* CJK Compatibility Forms */
255
+ (ucs >= 0xff00 && ucs <= 0xff60) || /* Fullwidth Forms */
256
+ (ucs >= 0xffe0 && ucs <= 0xffe6) ||
257
+ (ucs >= 0x20000 && ucs <= 0x2fffd) ||
258
+ (ucs >= 0x30000 && ucs <= 0x3fffd)));
262
+int mk_wcswidth(const wchar_t *pwcs, size_t n)
266
+ for (;*pwcs && n-- > 0; pwcs++)
267
+ if ((w = mk_wcwidth(*pwcs)) < 0)
277
+ * The following functions are the same as mk_wcwidth() and
278
+ * mk_wcwidth_cjk(), except that spacing characters in the East Asian
279
+ * Ambiguous (A) category as defined in Unicode Technical Report #11
280
+ * have a column width of 2. This variant might be useful for users of
281
+ * CJK legacy encodings who want to migrate to UCS without changing
282
+ * the traditional terminal character-width behaviour. It is not
283
+ * otherwise recommended for general use.
285
+static int mk_wcwidth_cjk(wchar_t ucs)
287
+ /* sorted list of non-overlapping intervals of East Asian Ambiguous
288
+ * characters, generated by "uniset +WIDTH-A -cat=Me -cat=Mn -cat=Cf c" */
289
+ static const struct interval ambiguous[] = {
290
+ { 0x00A1, 0x00A1 }, { 0x00A4, 0x00A4 }, { 0x00A7, 0x00A8 },
291
+ { 0x00AA, 0x00AA }, { 0x00AE, 0x00AE }, { 0x00B0, 0x00B4 },
292
+ { 0x00B6, 0x00BA }, { 0x00BC, 0x00BF }, { 0x00C6, 0x00C6 },
293
+ { 0x00D0, 0x00D0 }, { 0x00D7, 0x00D8 }, { 0x00DE, 0x00E1 },
294
+ { 0x00E6, 0x00E6 }, { 0x00E8, 0x00EA }, { 0x00EC, 0x00ED },
295
+ { 0x00F0, 0x00F0 }, { 0x00F2, 0x00F3 }, { 0x00F7, 0x00FA },
296
+ { 0x00FC, 0x00FC }, { 0x00FE, 0x00FE }, { 0x0101, 0x0101 },
297
+ { 0x0111, 0x0111 }, { 0x0113, 0x0113 }, { 0x011B, 0x011B },
298
+ { 0x0126, 0x0127 }, { 0x012B, 0x012B }, { 0x0131, 0x0133 },
299
+ { 0x0138, 0x0138 }, { 0x013F, 0x0142 }, { 0x0144, 0x0144 },
300
+ { 0x0148, 0x014B }, { 0x014D, 0x014D }, { 0x0152, 0x0153 },
301
+ { 0x0166, 0x0167 }, { 0x016B, 0x016B }, { 0x01CE, 0x01CE },
302
+ { 0x01D0, 0x01D0 }, { 0x01D2, 0x01D2 }, { 0x01D4, 0x01D4 },
303
+ { 0x01D6, 0x01D6 }, { 0x01D8, 0x01D8 }, { 0x01DA, 0x01DA },
304
+ { 0x01DC, 0x01DC }, { 0x0251, 0x0251 }, { 0x0261, 0x0261 },
305
+ { 0x02C4, 0x02C4 }, { 0x02C7, 0x02C7 }, { 0x02C9, 0x02CB },
306
+ { 0x02CD, 0x02CD }, { 0x02D0, 0x02D0 }, { 0x02D8, 0x02DB },
307
+ { 0x02DD, 0x02DD }, { 0x02DF, 0x02DF }, { 0x0391, 0x03A1 },
308
+ { 0x03A3, 0x03A9 }, { 0x03B1, 0x03C1 }, { 0x03C3, 0x03C9 },
309
+ { 0x0401, 0x0401 }, { 0x0410, 0x044F }, { 0x0451, 0x0451 },
310
+ { 0x2010, 0x2010 }, { 0x2013, 0x2016 }, { 0x2018, 0x2019 },
311
+ { 0x201C, 0x201D }, { 0x2020, 0x2022 }, { 0x2024, 0x2027 },
312
+ { 0x2030, 0x2030 }, { 0x2032, 0x2033 }, { 0x2035, 0x2035 },
313
+ { 0x203B, 0x203B }, { 0x203E, 0x203E }, { 0x2074, 0x2074 },
314
+ { 0x207F, 0x207F }, { 0x2081, 0x2084 }, { 0x20AC, 0x20AC },
315
+ { 0x2103, 0x2103 }, { 0x2105, 0x2105 }, { 0x2109, 0x2109 },
316
+ { 0x2113, 0x2113 }, { 0x2116, 0x2116 }, { 0x2121, 0x2122 },
317
+ { 0x2126, 0x2126 }, { 0x212B, 0x212B }, { 0x2153, 0x2154 },
318
+ { 0x215B, 0x215E }, { 0x2160, 0x216B }, { 0x2170, 0x2179 },
319
+ { 0x2190, 0x2199 }, { 0x21B8, 0x21B9 }, { 0x21D2, 0x21D2 },
320
+ { 0x21D4, 0x21D4 }, { 0x21E7, 0x21E7 }, { 0x2200, 0x2200 },
321
+ { 0x2202, 0x2203 }, { 0x2207, 0x2208 }, { 0x220B, 0x220B },
322
+ { 0x220F, 0x220F }, { 0x2211, 0x2211 }, { 0x2215, 0x2215 },
323
+ { 0x221A, 0x221A }, { 0x221D, 0x2220 }, { 0x2223, 0x2223 },
324
+ { 0x2225, 0x2225 }, { 0x2227, 0x222C }, { 0x222E, 0x222E },
325
+ { 0x2234, 0x2237 }, { 0x223C, 0x223D }, { 0x2248, 0x2248 },
326
+ { 0x224C, 0x224C }, { 0x2252, 0x2252 }, { 0x2260, 0x2261 },
327
+ { 0x2264, 0x2267 }, { 0x226A, 0x226B }, { 0x226E, 0x226F },
328
+ { 0x2282, 0x2283 }, { 0x2286, 0x2287 }, { 0x2295, 0x2295 },
329
+ { 0x2299, 0x2299 }, { 0x22A5, 0x22A5 }, { 0x22BF, 0x22BF },
330
+ { 0x2312, 0x2312 }, { 0x2460, 0x24E9 }, { 0x24EB, 0x254B },
331
+ { 0x2550, 0x2573 }, { 0x2580, 0x258F }, { 0x2592, 0x2595 },
332
+ { 0x25A0, 0x25A1 }, { 0x25A3, 0x25A9 }, { 0x25B2, 0x25B3 },
333
+ { 0x25B6, 0x25B7 }, { 0x25BC, 0x25BD }, { 0x25C0, 0x25C1 },
334
+ { 0x25C6, 0x25C8 }, { 0x25CB, 0x25CB }, { 0x25CE, 0x25D1 },
335
+ { 0x25E2, 0x25E5 }, { 0x25EF, 0x25EF }, { 0x2605, 0x2606 },
336
+ { 0x2609, 0x2609 }, { 0x260E, 0x260F }, { 0x2614, 0x2615 },
337
+ { 0x261C, 0x261C }, { 0x261E, 0x261E }, { 0x2640, 0x2640 },
338
+ { 0x2642, 0x2642 }, { 0x2660, 0x2661 }, { 0x2663, 0x2665 },
339
+ { 0x2667, 0x266A }, { 0x266C, 0x266D }, { 0x266F, 0x266F },
340
+ { 0x273D, 0x273D }, { 0x2776, 0x277F }, { 0xE000, 0xF8FF },
341
+ { 0xFFFD, 0xFFFD }, { 0xF0000, 0xFFFFD }, { 0x100000, 0x10FFFD }
344
+ /* binary search in table of non-spacing characters */
345
+ if (bisearch(ucs, ambiguous,
346
+ sizeof(ambiguous) / sizeof(struct interval) - 1))
349
+ return mk_wcwidth(ucs);
353
+int mk_wcswidth_cjk(const wchar_t *pwcs, size_t n)
357
+ for (;*pwcs && n-- > 0; pwcs++)
358
+ if ((w = mk_wcwidth_cjk(*pwcs)) < 0)
added
removed
doc/solaris9.patch
