1 /* Copyright (c) 2008-2009 Bjoern Hoehrmann <bjoern@hoehrmann.de>
2  *
3  * Permission is hereby granted, free of charge, to any person
4  * obtaining a copy of this software and associated documentation
5  * files (the "Software"), to deal in the Software without
6  * restriction, including without limitation the rights to use, copy,
7  * modify, merge, publish, distribute, sublicense, and/or sell copies
8  * of the Software, and to permit persons to whom the Software is
9  * furnished to do so, subject to the following conditions:
10  *
11  * The above copyright notice and this permission notice shall be
12  * included in all copies or substantial portions of the Software.
13  *
14  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
15  * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
16  * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
17  * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
18  * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
19  * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
20  * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21  * SOFTWARE.
22  */
23 
24 #include "compat.h"
25 
26 #include <assert.h>
27 #include <stddef.h>
28 #include <stdint.h>
29 #include <wchar.h>
30 
31 #include "utf8.h"
32 
33 #define UTF8_ACCEPT 0
34 #define UTF8_REJECT 1
35 
36 static const uint8_t utf8d[] = {
37 	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 00..1f
38 	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 20..3f
39 	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 40..5f
40 	0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0, // 60..7f
41 	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, // 80..9f
42 	7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, // a0..bf
43 	8,8,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2, // c0..df
44 	0xa,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x3,0x4,0x3,0x3, // e0..ef
45 	0xb,0x6,0x6,0x6,0x5,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8,0x8, // f0..ff
46 	0x0,0x1,0x2,0x3,0x5,0x8,0x7,0x1,0x1,0x1,0x4,0x6,0x1,0x1,0x1,0x1, // s0..s0
47 	1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,0,1,1,1,1,1,0,1,0,1,1,1,1,1,1, // s1..s2
48 	1,2,1,1,1,1,1,2,1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1, // s3..s4
49 	1,2,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,1,1,1,1,1,3,1,3,1,1,1,1,1,1, // s5..s6
50 	1,3,1,1,1,1,1,3,1,3,1,1,1,1,1,1,1,3,1,1,1,1,1,1,1,1,1,1,1,1,1,1, // s7..s8
51 };
52 
53 static inline uint32_t
54 decode(uint32_t* restrict state, uint32_t* restrict codep, uint8_t byte)
55 {
56 	uint32_t type = utf8d[byte];
57 
58 	*codep = (*state != UTF8_ACCEPT) ?
59 		(byte & 0x3fu) | (*codep << 6) :
60 		(0xff >> type) & (byte);
61 
62 	*state = utf8d[256 + *state*16 + type];
63 	return *state;
64 }
65 
66 
67 /* end of the converter, utility functions ahead */
68 
69 #define ZERO_WIDTH_SPACE 0x200B
70 
71 /* public version of decode */
72 uint32_t
73 utf8_decode(uint32_t* restrict state, uint32_t* restrict codep, uint8_t byte)
74 {
75 	return decode(state, codep, byte);
76 }
77 
78 /*
79  * returns 0, 1, 2 or less than 8 for tabs.  assumes that
80  * sizeof(wchar_t) == 4
81  */
82 static size_t
83 utf8_chwidth(uint32_t cp, int col)
84 {
85 	/* XXX: if we're running on a platform where sizeof(wchar_t)
86 	 * == 2 what to do?  The manpage for wcwidth and wcs isn't
87 	 * clear about the encoding, but if it's 16 bit wide I assume
88 	 * it must use UTF-16... right? */
89 	assert(sizeof(wchar_t) == 4);
90 
91 	/*
92 	 * Tabs are wide until the next multiple of eight.
93 	 */
94 	if (cp == '\t')
95 		return (((col + 8) / 8) * 8) - col;
96 
97 	return wcwidth((wchar_t)cp);
98 }
99 
100 size_t
101 utf8_snwidth(const char *s, size_t off, int col)
102 {
103 	size_t i, tot;
104 	uint32_t cp = 0, state = 0;
105 	int width;
106 
107 	tot = 0;
108 	for (i = 0; i < off; ++i)
109 		if (!decode(&state, &cp, s[i])) {
110 			width = utf8_chwidth(cp, col);
111 			tot += width;
112 			col += width;
113 		}
114 
115 	return tot;
116 }
117 
118 size_t
119 utf8_swidth(const char *s, int col )
120 {
121 	size_t tot;
122 	uint32_t cp = 0, state = 0;
123 	int width;
124 
125 	tot = 0;
126 	for (; *s; ++s)
127 		if (!decode(&state, &cp, *s)) {
128 			width = utf8_chwidth(cp, col);
129 			tot += width;
130 			col += width;
131 		}
132 
133 	return tot;
134 }
135 
136 size_t
137 utf8_swidth_between(const char *str, const char *end, int col)
138 {
139 	size_t tot;
140 	uint32_t cp = 0, state = 0;
141 	int width;
142 
143 	tot = 0;
144 	for (; *str && str < end; ++str)
145 		if (!decode(&state, &cp, *str)) {
146 			width = utf8_chwidth(cp, col);
147 			tot += width;
148 			col += width;
149 		}
150 	return tot;
151 }
152 
153 /*
154  * XXX: This is not correct.  There are codepoints classified as
155  * "emoji", but these can be joined together to form more complex
156  * emoji.  There is an official list of what these valid combinations
157  * are, but it would require a costly lookup (a trie can be used to
158  * reduce the times, but...).  The following approach is conceptually
159  * simpler: if there is a sequence of "emoji codepoints" (or ZWS) and
160  * then a space, consider everything before the space a single emoji.
161  * It needs a special check for numbers (yes, 0..9 and # are
162  * technically speaking emojis) but otherwise seems to work well in
163  * practice.
164  */
165 int
166 emojied_line(const char *s, const char **space_ret)
167 {
168 	uint32_t cp = 0, state = 0;
169 	int only_numbers = 1;
170 
171 	for (; *s; ++s) {
172 		if (!decode(&state, &cp, *s)) {
173 			if (cp == ZERO_WIDTH_SPACE)
174 				continue;
175 			if (cp == ' ') {
176 				*space_ret = s;
177 				return !only_numbers;
178 			}
179 			if (!is_emoji(cp))
180 				return 0;
181 			if (cp < '0' || cp > '9')
182 				only_numbers = 0;
183 		}
184 	}
185 
186 	return 0;
187 }