1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
|
// Copyright 2018 Google LLC.
// Use of this source code is governed by a BSD-style license that can be found in the LICENSE file.
#include "SkUTF.h"
#include <climits>
static constexpr inline int32_t left_shift(int32_t value, int32_t shift) {
return (int32_t) ((uint32_t) value << shift);
}
template <typename T> static constexpr bool is_align2(T x) { return 0 == (x & 1); }
template <typename T> static constexpr bool is_align4(T x) { return 0 == (x & 3); }
static constexpr inline bool utf16_is_high_surrogate(uint16_t c) { return (c & 0xFC00) == 0xD800; }
static constexpr inline bool utf16_is_low_surrogate(uint16_t c) { return (c & 0xFC00) == 0xDC00; }
/** @returns -1 iff invalid UTF8 byte,
0 iff UTF8 continuation byte,
1 iff ASCII byte,
2 iff leading byte of 2-byte sequence,
3 iff leading byte of 3-byte sequence, and
4 iff leading byte of 4-byte sequence.
I.e.: if return value > 0, then gives length of sequence.
*/
static int utf8_byte_type(uint8_t c) {
if (c < 0x80) {
return 1;
} else if (c < 0xC0) {
return 0;
} else if (c >= 0xF5 || (c & 0xFE) == 0xC0) { // "octet values c0, c1, f5 to ff never appear"
return -1;
} else {
int value = (((0xe5 << 24) >> ((unsigned)c >> 4 << 1)) & 3) + 1;
// assert(value >= 2 && value <=4);
return value;
}
}
static bool utf8_type_is_valid_leading_byte(int type) { return type > 0; }
static bool utf8_byte_is_continuation(uint8_t c) { return utf8_byte_type(c) == 0; }
////////////////////////////////////////////////////////////////////////////////
int SkUTF::CountUTF8(const char* utf8, size_t byteLength) {
if (!utf8) {
return -1;
}
int count = 0;
const char* stop = utf8 + byteLength;
while (utf8 < stop) {
int type = utf8_byte_type(*(const uint8_t*)utf8);
if (!utf8_type_is_valid_leading_byte(type) || utf8 + type > stop) {
return -1; // Sequence extends beyond end.
}
while(type-- > 1) {
++utf8;
if (!utf8_byte_is_continuation(*(const uint8_t*)utf8)) {
return -1;
}
}
++utf8;
++count;
}
return count;
}
int SkUTF::CountUTF16(const uint16_t* utf16, size_t byteLength) {
if (!utf16 || !is_align2(intptr_t(utf16)) || !is_align2(byteLength)) {
return -1;
}
const uint16_t* src = (const uint16_t*)utf16;
const uint16_t* stop = src + (byteLength >> 1);
int count = 0;
while (src < stop) {
unsigned c = *src++;
if (utf16_is_low_surrogate(c)) {
return -1;
}
if (utf16_is_high_surrogate(c)) {
if (src >= stop) {
return -1;
}
c = *src++;
if (!utf16_is_low_surrogate(c)) {
return -1;
}
}
count += 1;
}
return count;
}
int SkUTF::CountUTF32(const int32_t* utf32, size_t byteLength) {
if (!is_align4(intptr_t(utf32)) || !is_align4(byteLength) || byteLength >> 2 > INT_MAX) {
return -1;
}
const uint32_t kInvalidUnicharMask = 0xFF000000; // unichar fits in 24 bits
const uint32_t* ptr = (const uint32_t*)utf32;
const uint32_t* stop = ptr + (byteLength >> 2);
while (ptr < stop) {
if (*ptr & kInvalidUnicharMask) {
return -1;
}
ptr += 1;
}
return (int)(byteLength >> 2);
}
template <typename T>
static SkUnichar next_fail(const T** ptr, const T* end) {
*ptr = end;
return -1;
}
SkUnichar SkUTF::NextUTF8(const char** ptr, const char* end) {
if (!ptr || !end ) {
return -1;
}
const uint8_t* p = (const uint8_t*)*ptr;
if (!p || p >= (const uint8_t*)end) {
return next_fail(ptr, end);
}
int c = *p;
int hic = c << 24;
if (!utf8_type_is_valid_leading_byte(utf8_byte_type(c))) {
return next_fail(ptr, end);
}
if (hic < 0) {
uint32_t mask = (uint32_t)~0x3F;
hic = left_shift(hic, 1);
do {
++p;
if (p >= (const uint8_t*)end) {
return next_fail(ptr, end);
}
// check before reading off end of array.
uint8_t nextByte = *p;
if (!utf8_byte_is_continuation(nextByte)) {
return next_fail(ptr, end);
}
c = (c << 6) | (nextByte & 0x3F);
mask <<= 5;
} while ((hic = left_shift(hic, 1)) < 0);
c &= ~mask;
}
*ptr = (char*)p + 1;
return c;
}
SkUnichar SkUTF::NextUTF16(const uint16_t** ptr, const uint16_t* end) {
if (!ptr || !end ) {
return next_fail(ptr, end);
}
const uint16_t* src = *ptr;
if (!src || src + 1 > end || !is_align2(intptr_t(src))) {
return next_fail(ptr, end);
}
uint16_t c = *src++;
SkUnichar result = c;
if (utf16_is_low_surrogate(c)) {
return next_fail(ptr, end); // srcPtr should never point at low surrogate.
}
if (utf16_is_high_surrogate(c)) {
if (src + 1 > end) {
return next_fail(ptr, end); // Truncated string.
}
uint16_t low = *src++;
if (!utf16_is_low_surrogate(low)) {
return next_fail(ptr, end);
}
/*
[paraphrased from wikipedia]
Take the high surrogate and subtract 0xD800, then multiply by 0x400.
Take the low surrogate and subtract 0xDC00. Add these two results
together, and finally add 0x10000 to get the final decoded codepoint.
unicode = (high - 0xD800) * 0x400 + low - 0xDC00 + 0x10000
unicode = (high * 0x400) - (0xD800 * 0x400) + low - 0xDC00 + 0x10000
unicode = (high << 10) - (0xD800 << 10) + low - 0xDC00 + 0x10000
unicode = (high << 10) + low - ((0xD800 << 10) + 0xDC00 - 0x10000)
*/
result = (result << 10) + (SkUnichar)low - ((0xD800 << 10) + 0xDC00 - 0x10000);
}
*ptr = src;
return result;
}
SkUnichar SkUTF::NextUTF32(const int32_t** ptr, const int32_t* end) {
if (!ptr || !end ) {
return -1;
}
const int32_t* s = *ptr;
if (!s || s + 1 > end || !is_align4(intptr_t(s))) {
return next_fail(ptr, end);
}
int32_t value = *s;
const uint32_t kInvalidUnicharMask = 0xFF000000; // unichar fits in 24 bits
if (value & kInvalidUnicharMask) {
return next_fail(ptr, end);
}
*ptr = s + 1;
return value;
}
size_t SkUTF::ToUTF8(SkUnichar uni, char utf8[SkUTF::kMaxBytesInUTF8Sequence]) {
if ((uint32_t)uni > 0x10FFFF) {
return 0;
}
if (uni <= 127) {
if (utf8) {
*utf8 = (char)uni;
}
return 1;
}
char tmp[4];
char* p = tmp;
size_t count = 1;
while (uni > 0x7F >> count) {
*p++ = (char)(0x80 | (uni & 0x3F));
uni >>= 6;
count += 1;
}
if (utf8) {
p = tmp;
utf8 += count;
while (p < tmp + count - 1) {
*--utf8 = *p++;
}
*--utf8 = (char)(~(0xFF >> count) | uni);
}
return count;
}
size_t SkUTF::ToUTF16(SkUnichar uni, uint16_t utf16[2]) {
if ((uint32_t)uni > 0x10FFFF) {
return 0;
}
int extra = (uni > 0xFFFF);
if (utf16) {
if (extra) {
utf16[0] = (uint16_t)((0xD800 - 64) + (uni >> 10));
utf16[1] = (uint16_t)(0xDC00 | (uni & 0x3FF));
} else {
utf16[0] = (uint16_t)uni;
}
}
return 1 + extra;
}
|
