1 #ifndef UNICODE_H
2 #define UNICODE_H
3
4 #include <assert.h>
5 #include "adt/obst.h"
6
7 typedef unsigned int utf32;
8 #define UTF32_PRINTF_FORMAT "%u"
9
10 /**
11 * "parse" an utf8 character from a string.
12 * Warning: This function only works for valid utf-8 inputs. The behaviour
13 * is undefined for invalid utf-8 input.
14 *
15 * @param p A pointer to a pointer into the string. The pointer
16 * is incremented for each consumed char
17 */
read_utf8_char(const char ** p)18 static inline utf32 read_utf8_char(const char **p)
19 {
20 const unsigned char *c = (const unsigned char *) *p;
21 utf32 result;
22
23 if ((*c & 0x80) == 0) {
24 /* 1 character encoding: 0b0??????? */
25 result = *c++;
26 } else if ((*c & 0xE0) == 0xC0) {
27 /* 2 character encoding: 0b110?????, 0b10?????? */
28 result = *c++ & 0x1F;
29 result = (result << 6) | (*c++ & 0x3F);
30 } else if ((*c & 0xF0) == 0xE0) {
31 /* 3 character encoding: 0b1110????, 0b10??????, 0b10?????? */
32 result = *c++ & 0x0F;
33 result = (result << 6) | (*c++ & 0x3F);
34 result = (result << 6) | (*c++ & 0x3F);
35 } else {
36 /* 4 character enc.: 0b11110???, 0b10??????, 0b10??????, 0b10?????? */
37 assert((*c & 0xF8) == 0xF0);
38 result = *c++ & 0x07;
39 result = (result << 6) | (*c++ & 0x3F);
40 result = (result << 6) | (*c++ & 0x3F);
41 result = (result << 6) | (*c++ & 0x3F);
42 }
43
44 *p = (const char*) c;
45 return result;
46 }
47
obstack_grow_utf8(struct obstack * const obst,utf32 const c)48 static inline void obstack_grow_utf8(struct obstack *const obst, utf32 const c)
49 {
50 if (c < 0x80U) {
51 obstack_1grow(obst, c);
52 } else if (c < 0x800) {
53 obstack_1grow(obst, 0xC0 | (c >> 6));
54 goto one_more;
55 } else if (c < 0x10000) {
56 obstack_1grow(obst, 0xE0 | (c >> 12));
57 goto two_more;
58 } else {
59 obstack_1grow(obst, 0xF0 | (c >> 18));
60 obstack_1grow(obst, 0x80 | ((c >> 12) & 0x3F));
61 two_more:
62 obstack_1grow(obst, 0x80 | ((c >> 6) & 0x3F));
63 one_more:
64 obstack_1grow(obst, 0x80 | ( c & 0x3F));
65 }
66 }
67
68 #endif
69