1 /* See LICENSE file for copyright and license details. */
2 #include <stddef.h>
3 #include <stdint.h>
4 
5 #include "../grapheme.h"
6 #include "util.h"
7 
8 #define BETWEEN(c, l, u) ((c) >= (l) && (c) <= (u))
9 
10 /* lookup-table for the types of sequence first bytes */
11 static const struct {
12 	uint_least8_t  lower; /* lower bound of sequence first byte */
13 	uint_least8_t  upper; /* upper bound of sequence first byte */
14 	uint_least32_t mincp; /* smallest non-overlong encoded codepoint */
15 	uint_least32_t maxcp; /* largest encodable codepoint */
16 	/*
17 	 * implicit: table-offset represents the number of following
18 	 * bytes of the form 10xxxxxx (6 bits capacity each)
19 	 */
20 } lut[] = {
21 	[0] = {
22 		/* 0xxxxxxx */
23 		.lower = 0x00, /* 00000000 */
24 		.upper = 0x7F, /* 01111111 */
25 		.mincp = (uint_least32_t)0,
26 		.maxcp = ((uint_least32_t)1 << 7) - 1, /* 7 bits capacity */
27 	},
28 	[1] = {
29 		/* 110xxxxx */
30 		.lower = 0xC0, /* 11000000 */
31 		.upper = 0xDF, /* 11011111 */
32 		.mincp = (uint_least32_t)1 << 7,
33 		.maxcp = ((uint_least32_t)1 << 11) - 1, /* 5+6=11 bits capacity */
34 	},
35 	[2] = {
36 		/* 1110xxxx */
37 		.lower = 0xE0, /* 11100000 */
38 		.upper = 0xEF, /* 11101111 */
39 		.mincp = (uint_least32_t)1 << 11,
40 		.maxcp = ((uint_least32_t)1 << 16) - 1, /* 4+6+6=16 bits capacity */
41 	},
42 	[3] = {
43 		/* 11110xxx */
44 		.lower = 0xF0, /* 11110000 */
45 		.upper = 0xF7, /* 11110111 */
46 		.mincp = (uint_least32_t)1 << 16,
47 		.maxcp = ((uint_least32_t)1 << 21) - 1, /* 3+6+6+6=21 bits capacity */
48 	},
49 };
50 
51 size_t
52 grapheme_decode_utf8(const char *str, size_t len, uint_least32_t *cp)
53 {
54 	size_t off, i;
55 	uint_least32_t tmp;
56 
57 	if (cp == NULL) {
58 		/*
59 		 * instead of checking every time if cp is NULL within
60 		 * the decoder, simply point it at a dummy variable here.
61 		 */
62 		cp = &tmp;
63 	}
64 
65 	if (str == NULL || len == 0) {
66 		/* a sequence must be at least 1 byte long */
67 		*cp = GRAPHEME_INVALID_CODEPOINT;
68 		return 0;
69 	}
70 
71 	/* identify sequence type with the first byte */
72 	for (off = 0; off < LEN(lut); off++) {
73 		if (BETWEEN(((const unsigned char *)str)[0], lut[off].lower,
74 		            lut[off].upper)) {
75 			/*
76 			 * first byte is within the bounds; fill
77 			 * p with the the first bits contained in
78 			 * the first byte (by subtracting the high bits)
79 			 */
80 			*cp = ((const unsigned char *)str)[0] - lut[off].lower;
81 			break;
82 		}
83 	}
84 	if (off == LEN(lut)) {
85 		/*
86 		 * first byte does not match a sequence type;
87 		 * set cp as invalid and return 1 byte processed
88 		 *
89 		 * this also includes the cases where bits higher than
90 		 * the 8th are set on systems with CHAR_BIT > 8
91 		 */
92 		*cp = GRAPHEME_INVALID_CODEPOINT;
93 		return 1;
94 	}
95 	if (1 + off > len) {
96 		/*
97 		 * input is not long enough, set cp as invalid
98 		 */
99 		*cp = GRAPHEME_INVALID_CODEPOINT;
100 
101 		/*
102 		 * count the following continuation bytes, but nothing
103 		 * else in case we have a "rogue" case where e.g. such a
104 		 * sequence starter occurs right before a NUL-byte.
105 		 */
106 		for (i = 0; 1 + i < len; i++) {
107 			if(!BETWEEN(((const unsigned char *)str)[1 + i],
108 			            0x80, 0xBF)) {
109 				break;
110 			}
111 		}
112 
113 		/*
114 		 * if the continuation bytes do not continue until
115 		 * the end, return the incomplete sequence length.
116 		 * Otherwise return the number of bytes we actually
117 		 * expected, which is larger than n.
118 		 */
119 		return ((1 + i) < len) ? (1 + i) : (1 + off);
120 	}
121 
122 	/*
123 	 * process 'off' following bytes, each of the form 10xxxxxx
124 	 * (i.e. between 0x80 (10000000) and 0xBF (10111111))
125 	 */
126 	for (i = 1; i <= off; i++) {
127 		if(!BETWEEN(((const unsigned char *)str)[i], 0x80, 0xBF)) {
128 			/*
129 			 * byte does not match format; return
130 			 * number of bytes processed excluding the
131 			 * unexpected character as recommended since
132 			 * Unicode 6 (chapter 3)
133 			 *
134 			 * this also includes the cases where bits
135 			 * higher than the 8th are set on systems
136 			 * with CHAR_BIT > 8
137 			 */
138 			*cp = GRAPHEME_INVALID_CODEPOINT;
139 			return 1 + (i - 1);
140 		}
141 		/*
142 		 * shift codepoint by 6 bits and add the 6 stored bits
143 		 * in s[i] to it using the bitmask 0x3F (00111111)
144 		 */
145 		*cp = (*cp << 6) | (((const unsigned char *)str)[i] & 0x3F);
146 	}
147 
148 	if (*cp < lut[off].mincp ||
149 	    BETWEEN(*cp, UINT32_C(0xD800), UINT32_C(0xDFFF)) ||
150 	    *cp > UINT32_C(0x10FFFF)) {
151 		/*
152 		 * codepoint is overlong encoded in the sequence, is a
153 		 * high or low UTF-16 surrogate half (0xD800..0xDFFF) or
154 		 * not representable in UTF-16 (>0x10FFFF) (RFC-3629
155 		 * specifies the latter two conditions)
156 		 */
157 		*cp = GRAPHEME_INVALID_CODEPOINT;
158 	}
159 
160 	return 1 + off;
161 }
162 
163 size_t
164 grapheme_encode_utf8(uint_least32_t cp, char *str, size_t len)
165 {
166 	size_t off, i;
167 
168 	if (BETWEEN(cp, UINT32_C(0xD800), UINT32_C(0xDFFF)) ||
169 	    cp > UINT32_C(0x10FFFF)) {
170 		/*
171 		 * codepoint is a high or low UTF-16 surrogate half
172 		 * (0xD800..0xDFFF) or not representable in UTF-16
173 		 * (>0x10FFFF), which RFC-3629 deems invalid for UTF-8.
174 		 */
175 		cp = GRAPHEME_INVALID_CODEPOINT;
176 	}
177 
178 	/* determine necessary sequence type */
179 	for (off = 0; off < LEN(lut); off++) {
180 		if (cp <= lut[off].maxcp) {
181 			break;
182 		}
183 	}
184 	if (1 + off > len || str == NULL || len == 0) {
185 		/*
186 		 * specified buffer is too small to store sequence or
187 		 * the caller just wanted to know how many bytes the
188 		 * codepoint needs by passing a NULL-buffer.
189 		 */
190 		return 1 + off;
191 	}
192 
193 	/* build sequence by filling cp-bits into each byte */
194 
195 	/*
196 	 * lut[off].lower is the bit-format for the first byte and
197 	 * the bits to fill into it are determined by shifting the
198 	 * cp 6 times the number of following bytes, as each
199 	 * following byte stores 6 bits, yielding the wanted bits.
200 	 *
201 	 * We do not overwrite the mask because we guaranteed earlier
202 	 * that there are no bits higher than the mask allows.
203 	 */
204 	((unsigned char *)str)[0] = lut[off].lower |
205 	                            (uint_least8_t)(cp >> (6 * off));
206 
207 	for (i = 1; i <= off; i++) {
208 		/*
209 		 * the bit-format for following bytes is 10000000 (0x80)
210 		 * and it each stores 6 bits in the 6 low bits that we
211 		 * extract from the properly-shifted value using the
212 		 * mask 00111111 (0x3F)
213 		 */
214 		((unsigned char *)str)[i] = 0x80 |
215 		                            ((cp >> (6 * (off - i))) & 0x3F);
216 	}
217 
218 	return 1 + off;
219 }