utf8: make it easier to auto-update git_wcwidth()

The function git_wcwidth() returns for a given unicode code point the width on the display: -1 for control characters, 0 for combining or other non-visible code points 1 for e.g. ASCII 2 for double-width code points. This table had been originally been extracted for one Unicode version, probably 3.2. We now use two tables these days, one for zero-width and another for double-width. Make it easier to update these tables to a later version of Unicode by factoring out the table from utf8.c into unicode_width.h and add the script update_unicode.sh to update the table based on the latest Unicode specification files. Thanks to Peter Krefting <peter@softwolves.pp.se> and Kevin Bracey <kevin@bracey.fi> for helping with their Unicode knowledge. Signed-off-by: Torsten Bögershausen <tboegi@web.de> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Torsten Bögershausen committed May 9, 2014 at 23:51 UTC 9c94389c3ee02df891100b894c1790a524268d91
5 files changed +329 -59
.gitignore
+1
@@ -225,6 +225,7 @@
225 /config.mak.autogen
226 /config.mak.append
227 /configure
228 +/unicode
229 /tags
230 /TAGS
231 /cscope*
Makefile
+1
@@ -726,6 +726,7 @@ LIB_H += transport.h
726 LIB_H += tree-walk.h
727 LIB_H += tree.h
728 LIB_H += unpack-trees.h
729 +LIB_H += unicode_width.h
730 LIB_H += url.h
731 LIB_H += urlmatch.h
732 LIB_H += userdiff.h
unicode_width.h new
+288
@@ -0,0 +1,288 @@
1 +static const struct interval zero_width[] = {
2 +{ 0x0300, 0x036F },
3 +{ 0x0483, 0x0489 },
4 +{ 0x0591, 0x05BD },
5 +{ 0x05BF, 0x05BF },
6 +{ 0x05C1, 0x05C2 },
7 +{ 0x05C4, 0x05C5 },
8 +{ 0x05C7, 0x05C7 },
9 +{ 0x0600, 0x0604 },
10 +{ 0x0610, 0x061A },
11 +{ 0x061C, 0x061C },
12 +{ 0x064B, 0x065F },
13 +{ 0x0670, 0x0670 },
14 +{ 0x06D6, 0x06DD },
15 +{ 0x06DF, 0x06E4 },
16 +{ 0x06E7, 0x06E8 },
17 +{ 0x06EA, 0x06ED },
18 +{ 0x070F, 0x070F },
19 +{ 0x0711, 0x0711 },
20 +{ 0x0730, 0x074A },
21 +{ 0x07A6, 0x07B0 },
22 +{ 0x07EB, 0x07F3 },
23 +{ 0x0816, 0x0819 },
24 +{ 0x081B, 0x0823 },
25 +{ 0x0825, 0x0827 },
26 +{ 0x0829, 0x082D },
27 +{ 0x0859, 0x085B },
28 +{ 0x08E4, 0x08FE },
29 +{ 0x0900, 0x0902 },
30 +{ 0x093A, 0x093A },
31 +{ 0x093C, 0x093C },
32 +{ 0x0941, 0x0948 },
33 +{ 0x094D, 0x094D },
34 +{ 0x0951, 0x0957 },
35 +{ 0x0962, 0x0963 },
36 +{ 0x0981, 0x0981 },
37 +{ 0x09BC, 0x09BC },
38 +{ 0x09C1, 0x09C4 },
39 +{ 0x09CD, 0x09CD },
40 +{ 0x09E2, 0x09E3 },
41 +{ 0x0A01, 0x0A02 },
42 +{ 0x0A3C, 0x0A3C },
43 +{ 0x0A41, 0x0A42 },
44 +{ 0x0A47, 0x0A48 },
45 +{ 0x0A4B, 0x0A4D },
46 +{ 0x0A51, 0x0A51 },
47 +{ 0x0A70, 0x0A71 },
48 +{ 0x0A75, 0x0A75 },
49 +{ 0x0A81, 0x0A82 },
50 +{ 0x0ABC, 0x0ABC },
51 +{ 0x0AC1, 0x0AC5 },
52 +{ 0x0AC7, 0x0AC8 },
53 +{ 0x0ACD, 0x0ACD },
54 +{ 0x0AE2, 0x0AE3 },
55 +{ 0x0B01, 0x0B01 },
56 +{ 0x0B3C, 0x0B3C },
57 +{ 0x0B3F, 0x0B3F },
58 +{ 0x0B41, 0x0B44 },
59 +{ 0x0B4D, 0x0B4D },
60 +{ 0x0B56, 0x0B56 },
61 +{ 0x0B62, 0x0B63 },
62 +{ 0x0B82, 0x0B82 },
63 +{ 0x0BC0, 0x0BC0 },
64 +{ 0x0BCD, 0x0BCD },
65 +{ 0x0C3E, 0x0C40 },
66 +{ 0x0C46, 0x0C48 },
67 +{ 0x0C4A, 0x0C4D },
68 +{ 0x0C55, 0x0C56 },
69 +{ 0x0C62, 0x0C63 },
70 +{ 0x0CBC, 0x0CBC },
71 +{ 0x0CBF, 0x0CBF },
72 +{ 0x0CC6, 0x0CC6 },
73 +{ 0x0CCC, 0x0CCD },
74 +{ 0x0CE2, 0x0CE3 },
75 +{ 0x0D41, 0x0D44 },
76 +{ 0x0D4D, 0x0D4D },
77 +{ 0x0D62, 0x0D63 },
78 +{ 0x0DCA, 0x0DCA },
79 +{ 0x0DD2, 0x0DD4 },
80 +{ 0x0DD6, 0x0DD6 },
81 +{ 0x0E31, 0x0E31 },
82 +{ 0x0E34, 0x0E3A },
83 +{ 0x0E47, 0x0E4E },
84 +{ 0x0EB1, 0x0EB1 },
85 +{ 0x0EB4, 0x0EB9 },
86 +{ 0x0EBB, 0x0EBC },
87 +{ 0x0EC8, 0x0ECD },
88 +{ 0x0F18, 0x0F19 },
89 +{ 0x0F35, 0x0F35 },
90 +{ 0x0F37, 0x0F37 },
91 +{ 0x0F39, 0x0F39 },
92 +{ 0x0F71, 0x0F7E },
93 +{ 0x0F80, 0x0F84 },
94 +{ 0x0F86, 0x0F87 },
95 +{ 0x0F8D, 0x0F97 },
96 +{ 0x0F99, 0x0FBC },
97 +{ 0x0FC6, 0x0FC6 },
98 +{ 0x102D, 0x1030 },
99 +{ 0x1032, 0x1037 },
100 +{ 0x1039, 0x103A },
101 +{ 0x103D, 0x103E },
102 +{ 0x1058, 0x1059 },
103 +{ 0x105E, 0x1060 },
104 +{ 0x1071, 0x1074 },
105 +{ 0x1082, 0x1082 },
106 +{ 0x1085, 0x1086 },
107 +{ 0x108D, 0x108D },
108 +{ 0x109D, 0x109D },
109 +{ 0x1160, 0x11FF },
110 +{ 0x135D, 0x135F },
111 +{ 0x1712, 0x1714 },
112 +{ 0x1732, 0x1734 },
113 +{ 0x1752, 0x1753 },
114 +{ 0x1772, 0x1773 },
115 +{ 0x17B4, 0x17B5 },
116 +{ 0x17B7, 0x17BD },
117 +{ 0x17C6, 0x17C6 },
118 +{ 0x17C9, 0x17D3 },
119 +{ 0x17DD, 0x17DD },
120 +{ 0x180B, 0x180E },
121 +{ 0x18A9, 0x18A9 },
122 +{ 0x1920, 0x1922 },
123 +{ 0x1927, 0x1928 },
124 +{ 0x1932, 0x1932 },
125 +{ 0x1939, 0x193B },
126 +{ 0x1A17, 0x1A18 },
127 +{ 0x1A1B, 0x1A1B },
128 +{ 0x1A56, 0x1A56 },
129 +{ 0x1A58, 0x1A5E },
130 +{ 0x1A60, 0x1A60 },
131 +{ 0x1A62, 0x1A62 },
132 +{ 0x1A65, 0x1A6C },
133 +{ 0x1A73, 0x1A7C },
134 +{ 0x1A7F, 0x1A7F },
135 +{ 0x1B00, 0x1B03 },
136 +{ 0x1B34, 0x1B34 },
137 +{ 0x1B36, 0x1B3A },
138 +{ 0x1B3C, 0x1B3C },
139 +{ 0x1B42, 0x1B42 },
140 +{ 0x1B6B, 0x1B73 },
141 +{ 0x1B80, 0x1B81 },
142 +{ 0x1BA2, 0x1BA5 },
143 +{ 0x1BA8, 0x1BA9 },
144 +{ 0x1BAB, 0x1BAB },
145 +{ 0x1BE6, 0x1BE6 },
146 +{ 0x1BE8, 0x1BE9 },
147 +{ 0x1BED, 0x1BED },
148 +{ 0x1BEF, 0x1BF1 },
149 +{ 0x1C2C, 0x1C33 },
150 +{ 0x1C36, 0x1C37 },
151 +{ 0x1CD0, 0x1CD2 },
152 +{ 0x1CD4, 0x1CE0 },
153 +{ 0x1CE2, 0x1CE8 },
154 +{ 0x1CED, 0x1CED },
155 +{ 0x1CF4, 0x1CF4 },
156 +{ 0x1DC0, 0x1DE6 },
157 +{ 0x1DFC, 0x1DFF },
158 +{ 0x200B, 0x200F },
159 +{ 0x202A, 0x202E },
160 +{ 0x2060, 0x2064 },
161 +{ 0x2066, 0x206F },
162 +{ 0x20D0, 0x20F0 },
163 +{ 0x2CEF, 0x2CF1 },
164 +{ 0x2D7F, 0x2D7F },
165 +{ 0x2DE0, 0x2DFF },
166 +{ 0x302A, 0x302D },
167 +{ 0x3099, 0x309A },
168 +{ 0xA66F, 0xA672 },
169 +{ 0xA674, 0xA67D },
170 +{ 0xA69F, 0xA69F },
171 +{ 0xA6F0, 0xA6F1 },
172 +{ 0xA802, 0xA802 },
173 +{ 0xA806, 0xA806 },
174 +{ 0xA80B, 0xA80B },
175 +{ 0xA825, 0xA826 },
176 +{ 0xA8C4, 0xA8C4 },
177 +{ 0xA8E0, 0xA8F1 },
178 +{ 0xA926, 0xA92D },
179 +{ 0xA947, 0xA951 },
180 +{ 0xA980, 0xA982 },
181 +{ 0xA9B3, 0xA9B3 },
182 +{ 0xA9B6, 0xA9B9 },
183 +{ 0xA9BC, 0xA9BC },
184 +{ 0xAA29, 0xAA2E },
185 +{ 0xAA31, 0xAA32 },
186 +{ 0xAA35, 0xAA36 },
187 +{ 0xAA43, 0xAA43 },
188 +{ 0xAA4C, 0xAA4C },
189 +{ 0xAAB0, 0xAAB0 },
190 +{ 0xAAB2, 0xAAB4 },
191 +{ 0xAAB7, 0xAAB8 },
192 +{ 0xAABE, 0xAABF },
193 +{ 0xAAC1, 0xAAC1 },
194 +{ 0xAAEC, 0xAAED },
195 +{ 0xAAF6, 0xAAF6 },
196 +{ 0xABE5, 0xABE5 },
197 +{ 0xABE8, 0xABE8 },
198 +{ 0xABED, 0xABED },
199 +{ 0xFB1E, 0xFB1E },
200 +{ 0xFE00, 0xFE0F },
201 +{ 0xFE20, 0xFE26 },
202 +{ 0xFEFF, 0xFEFF },
203 +{ 0xFFF9, 0xFFFB },
204 +{ 0x101FD, 0x101FD },
205 +{ 0x10A01, 0x10A03 },
206 +{ 0x10A05, 0x10A06 },
207 +{ 0x10A0C, 0x10A0F },
208 +{ 0x10A38, 0x10A3A },
209 +{ 0x10A3F, 0x10A3F },
210 +{ 0x11001, 0x11001 },
211 +{ 0x11038, 0x11046 },
212 +{ 0x11080, 0x11081 },
213 +{ 0x110B3, 0x110B6 },
214 +{ 0x110B9, 0x110BA },
215 +{ 0x110BD, 0x110BD },
216 +{ 0x11100, 0x11102 },
217 +{ 0x11127, 0x1112B },
218 +{ 0x1112D, 0x11134 },
219 +{ 0x11180, 0x11181 },
220 +{ 0x111B6, 0x111BE },
221 +{ 0x116AB, 0x116AB },
222 +{ 0x116AD, 0x116AD },
223 +{ 0x116B0, 0x116B5 },
224 +{ 0x116B7, 0x116B7 },
225 +{ 0x16F8F, 0x16F92 },
226 +{ 0x1D167, 0x1D169 },
227 +{ 0x1D173, 0x1D182 },
228 +{ 0x1D185, 0x1D18B },
229 +{ 0x1D1AA, 0x1D1AD },
230 +{ 0x1D242, 0x1D244 },
231 +{ 0xE0001, 0xE0001 },
232 +{ 0xE0020, 0xE007F },
233 +{ 0xE0100, 0xE01EF }
234 +};
235 +static const struct interval double_width[] = {
236 +{ /* plane */ 0x0, 0x1C },
237 +{ /* plane */ 0x1C, 0x21 },
238 +{ /* plane */ 0x21, 0x22 },
239 +{ /* plane */ 0x22, 0x23 },
240 +{ /* plane */ 0x0, 0x0 },
241 +{ /* plane */ 0x0, 0x0 },
242 +{ /* plane */ 0x0, 0x0 },
243 +{ /* plane */ 0x0, 0x0 },
244 +{ /* plane */ 0x0, 0x0 },
245 +{ /* plane */ 0x0, 0x0 },
246 +{ /* plane */ 0x0, 0x0 },
247 +{ /* plane */ 0x0, 0x0 },
248 +{ /* plane */ 0x0, 0x0 },
249 +{ /* plane */ 0x0, 0x0 },
250 +{ /* plane */ 0x0, 0x0 },
251 +{ /* plane */ 0x0, 0x0 },
252 +{ /* plane */ 0x0, 0x0 },
253 +{ 0x1100, 0x115F },
254 +{ 0x2329, 0x232A },
255 +{ 0x2E80, 0x2E99 },
256 +{ 0x2E9B, 0x2EF3 },
257 +{ 0x2F00, 0x2FD5 },
258 +{ 0x2FF0, 0x2FFB },
259 +{ 0x3000, 0x303E },
260 +{ 0x3041, 0x3096 },
261 +{ 0x3099, 0x30FF },
262 +{ 0x3105, 0x312D },
263 +{ 0x3131, 0x318E },
264 +{ 0x3190, 0x31BA },
265 +{ 0x31C0, 0x31E3 },
266 +{ 0x31F0, 0x321E },
267 +{ 0x3220, 0x3247 },
268 +{ 0x3250, 0x32FE },
269 +{ 0x3300, 0x4DBF },
270 +{ 0x4E00, 0xA48C },
271 +{ 0xA490, 0xA4C6 },
272 +{ 0xA960, 0xA97C },
273 +{ 0xAC00, 0xD7A3 },
274 +{ 0xF900, 0xFAFF },
275 +{ 0xFE10, 0xFE19 },
276 +{ 0xFE30, 0xFE52 },
277 +{ 0xFE54, 0xFE66 },
278 +{ 0xFE68, 0xFE6B },
279 +{ 0xFF01, 0xFF60 },
280 +{ 0xFFE0, 0xFFE6 },
281 +{ 0x1B000, 0x1B001 },
282 +{ 0x1F200, 0x1F202 },
283 +{ 0x1F210, 0x1F23A },
284 +{ 0x1F240, 0x1F248 },
285 +{ 0x1F250, 0x1F251 },
286 +{ 0x20000, 0x2FFFD },
287 +{ 0x30000, 0x3FFFD }
288 +};
update_unicode.sh new
+37
@@ -0,0 +1,37 @@
1 +#!/bin/sh
2 +#See http://www.unicode.org/reports/tr44/
3 +#
4 +#Me Enclosing_Mark an enclosing combining mark
5 +#Mn Nonspacing_Mark a nonspacing combining mark (zero advance width)
6 +#Cf Format a format control character
7 +#
8 +UNICODEWIDTH_H=../unicode_width.h
9 +if ! test -d unicode; then
10 + mkdir unicode
11 +fi &&
12 +( cd unicode &&
13 + if ! test -f UnicodeData.txt; then
14 + wget http://www.unicode.org/Public/UCD/latest/ucd/UnicodeData.txt
15 + fi &&
16 + if ! test -f EastAsianWidth.txt; then
17 + wget http://www.unicode.org/Public/UCD/latest/ucd/EastAsianWidth.txt
18 + fi &&
19 + if ! test -d uniset; then
20 + git clone https://github.com/depp/uniset.git
21 + fi &&
22 + (
23 + cd uniset &&
24 + if ! test -x uniset; then
25 + autoreconf -i &&
26 + ./configure --enable-warnings=-Werror CFLAGS='-O0 -ggdb'
27 + fi &&
28 + make
29 + ) &&
30 + echo "static const struct interval zero_width[] = {" >$UNICODEWIDTH_H &&
31 + UNICODE_DIR=. ./uniset/uniset --32 cat:Me,Mn,Cf + U+1160..U+11FF - U+00AD |
32 + grep -v plane >>$UNICODEWIDTH_H &&
33 + echo "};" >>$UNICODEWIDTH_H &&
34 + echo "static const struct interval double_width[] = {" >>$UNICODEWIDTH_H &&
35 + UNICODE_DIR=. ./uniset/uniset --32 eaw:F,W >>$UNICODEWIDTH_H &&
36 + echo "};" >>$UNICODEWIDTH_H
37 +)
utf8.c
+2 -59
@@ -80,65 +80,8 @@ static int git_wcwidth(ucs_char_t ch)
80 {
81 /*
82 * Sorted list of non-overlapping intervals of non-spacing characters,
83 - * generated by
84 - * "uniset +cat=Me +cat=Mn +cat=Cf -00AD +1160-11FF +200B c".
83 */
86 - static const struct interval combining[] = {
87 - { 0x0300, 0x036F }, { 0x0483, 0x0489 }, { 0x0591, 0x05BD },
88 - { 0x05BF, 0x05BF }, { 0x05C1, 0x05C2 }, { 0x05C4, 0x05C5 },
89 - { 0x05C7, 0x05C7 }, { 0x0600, 0x0604 }, { 0x0610, 0x061A },
90 - { 0x064B, 0x065F }, { 0x0670, 0x0670 }, { 0x06D6, 0x06E4 },
91 - { 0x06E7, 0x06E8 }, { 0x06EA, 0x06ED }, { 0x070F, 0x070F },
92 - { 0x0711, 0x0711 }, { 0x0730, 0x074A }, { 0x07A6, 0x07B0 },
93 - { 0x0901, 0x0902 }, { 0x093C, 0x093C }, { 0x0941, 0x0948 },
94 - { 0x094D, 0x094D }, { 0x0951, 0x0954 }, { 0x0962, 0x0963 },
95 - { 0x0981, 0x0981 }, { 0x09BC, 0x09BC }, { 0x09C1, 0x09C4 },
96 - { 0x09CD, 0x09CD }, { 0x09E2, 0x09E3 }, { 0x0A01, 0x0A02 },
97 - { 0x0A3C, 0x0A3C }, { 0x0A41, 0x0A42 }, { 0x0A47, 0x0A48 },
98 - { 0x0A4B, 0x0A4D }, { 0x0A70, 0x0A71 }, { 0x0A81, 0x0A82 },
99 - { 0x0ABC, 0x0ABC }, { 0x0AC1, 0x0AC5 }, { 0x0AC7, 0x0AC8 },
100 - { 0x0ACD, 0x0ACD }, { 0x0AE2, 0x0AE3 }, { 0x0B01, 0x0B01 },
101 - { 0x0B3C, 0x0B3C }, { 0x0B3F, 0x0B3F }, { 0x0B41, 0x0B43 },
102 - { 0x0B4D, 0x0B4D }, { 0x0B56, 0x0B56 }, { 0x0B82, 0x0B82 },
103 - { 0x0BC0, 0x0BC0 }, { 0x0BCD, 0x0BCD }, { 0x0C3E, 0x0C40 },
104 - { 0x0C46, 0x0C48 }, { 0x0C4A, 0x0C4D }, { 0x0C55, 0x0C56 },
105 - { 0x0CBC, 0x0CBC }, { 0x0CBF, 0x0CBF }, { 0x0CC6, 0x0CC6 },
106 - { 0x0CCC, 0x0CCD }, { 0x0D41, 0x0D43 }, { 0x0D4D, 0x0D4D },
107 - { 0x0DCA, 0x0DCA }, { 0x0DD2, 0x0DD4 }, { 0x0DD6, 0x0DD6 },
108 - { 0x0E31, 0x0E31 }, { 0x0E34, 0x0E3A }, { 0x0E47, 0x0E4E },
109 - { 0x0EB1, 0x0EB1 }, { 0x0EB4, 0x0EB9 }, { 0x0EBB, 0x0EBC },
110 - { 0x0EC8, 0x0ECD }, { 0x0F18, 0x0F19 }, { 0x0F35, 0x0F35 },
111 - { 0x0F37, 0x0F37 }, { 0x0F39, 0x0F39 }, { 0x0F71, 0x0F7E },
112 - { 0x0F80, 0x0F84 }, { 0x0F86, 0x0F87 }, { 0x0F90, 0x0F97 },
113 - { 0x0F99, 0x0FBC }, { 0x0FC6, 0x0FC6 }, { 0x102D, 0x1030 },
114 - { 0x1032, 0x1032 }, { 0x1036, 0x1037 }, { 0x1039, 0x1039 },
115 - { 0x1058, 0x1059 }, { 0x1160, 0x11FF }, { 0x1712, 0x1714 },
116 - { 0x1732, 0x1734 }, { 0x1752, 0x1753 }, { 0x1772, 0x1773 },
117 - { 0x17B4, 0x17B5 }, { 0x17B7, 0x17BD }, { 0x17C6, 0x17C6 },
118 - { 0x17C9, 0x17D3 }, { 0x17DD, 0x17DD }, { 0x180B, 0x180D },
119 - { 0x18A9, 0x18A9 }, { 0x1920, 0x1922 }, { 0x1927, 0x1928 },
120 - { 0x1932, 0x1932 }, { 0x1939, 0x193B }, { 0x200B, 0x200F },
121 - { 0x202A, 0x202E }, { 0x2060, 0x2063 }, { 0x206A, 0x206F },
122 - { 0x20D0, 0x20EA }, { 0x302A, 0x302F }, { 0x3099, 0x309A },
123 - { 0xFB1E, 0xFB1E }, { 0xFE00, 0xFE0F }, { 0xFE20, 0xFE23 },
124 - { 0xFEFF, 0xFEFF }, { 0xFFF9, 0xFFFB }, { 0x1D167, 0x1D169 },
125 - { 0x1D173, 0x1D182 }, { 0x1D185, 0x1D18B },
126 - { 0x1D1AA, 0x1D1AD }, { 0xE0001, 0xE0001 },
127 - { 0xE0020, 0xE007F }, { 0xE0100, 0xE01EF }
128 - };
129 - static const struct interval double_width[] = {
130 - { 0x1100, 0x115F },
131 - { 0x2329, 0x232A },
132 - { 0x2E80, 0x303E },
133 - { 0x3040, 0xA4CF },
134 - { 0xAC00, 0xD7A3 },
135 - { 0xF900, 0xFAFF },
136 - { 0xFE30, 0xFE6F },
137 - { 0xFF00, 0xFF60 },
138 - { 0xFFE0, 0xFFE6 },
139 - { 0x20000, 0x2FFFD },
140 - { 0x30000, 0x3FFFD }
141 - };
84 +#include "unicode_width.h"
85
86 /* test for 8-bit control characters */
87 if (ch == 0)
@@ -147,7 +90,7 @@ static int git_wcwidth(ucs_char_t ch)
90 return -1;
91
92 /* binary search in table of non-spacing characters */
150 - if (bisearch(ch, combining, sizeof(combining)
93 + if (bisearch(ch, zero_width, sizeof(zero_width)
94 / sizeof(struct interval) - 1))
95 return 0;
96