blob: 9e344a0de95e708e0992c20e7a33124d14cbf420 (
plain)
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
|
/* Finding the optimal width of unicode characters in a buffer */
#if STRINGLIB_IS_UNICODE
/* Mask to check or force alignment of a pointer to C 'long' boundaries */
#define LONG_PTR_MASK (size_t) (SIZEOF_LONG - 1)
/* Mask to quickly check whether a C 'long' contains a
non-ASCII, UTF8-encoded char. */
#if (SIZEOF_LONG == 8)
# define UCS1_ASCII_CHAR_MASK 0x8080808080808080UL
#elif (SIZEOF_LONG == 4)
# define UCS1_ASCII_CHAR_MASK 0x80808080UL
#else
# error C 'long' size should be either 4 or 8!
#endif
#if STRINGLIB_SIZEOF_CHAR == 1
Py_LOCAL_INLINE(Py_UCS4)
STRINGLIB(find_max_char)(const STRINGLIB_CHAR *begin, const STRINGLIB_CHAR *end)
{
const unsigned char *p = (const unsigned char *) begin;
const unsigned char *aligned_end = (const unsigned char *) ((size_t) end & ~LONG_PTR_MASK);
while (p < end) {
if (!((size_t) p & LONG_PTR_MASK)) {
/* Help register allocation */
register const unsigned char *_p = p;
while (_p < aligned_end) {
unsigned long value = *(unsigned long *) _p;
if (value & UCS1_ASCII_CHAR_MASK)
return 255;
_p += SIZEOF_LONG;
}
p = _p;
if (p == end)
break;
}
if (*p++ & 0x80)
return 255;
}
return 127;
}
#undef LONG_PTR_MASK
#undef ASCII_CHAR_MASK
#else /* STRINGLIB_SIZEOF_CHAR == 1 */
#define MASK_ASCII 0xFFFFFF80
#define MASK_UCS1 0xFFFFFF00
#define MASK_UCS2 0xFFFF0000
#define MAX_CHAR_ASCII 0x7f
#define MAX_CHAR_UCS1 0xff
#define MAX_CHAR_UCS2 0xffff
#define MAX_CHAR_UCS4 0x10ffff
Py_LOCAL_INLINE(Py_UCS4)
STRINGLIB(find_max_char)(const STRINGLIB_CHAR *begin, const STRINGLIB_CHAR *end)
{
#if STRINGLIB_SIZEOF_CHAR == 2
const Py_UCS4 mask_limit = MASK_UCS1;
const Py_UCS4 max_char_limit = MAX_CHAR_UCS2;
#elif STRINGLIB_SIZEOF_CHAR == 4
const Py_UCS4 mask_limit = MASK_UCS2;
const Py_UCS4 max_char_limit = MAX_CHAR_UCS4;
#else
#error Invalid STRINGLIB_SIZEOF_CHAR (must be 1, 2 or 4)
#endif
register Py_UCS4 mask;
Py_ssize_t n = end - begin;
const STRINGLIB_CHAR *p = begin;
const STRINGLIB_CHAR *unrolled_end = begin + (n & ~ (Py_ssize_t) 3);
Py_UCS4 max_char;
max_char = MAX_CHAR_ASCII;
mask = MASK_ASCII;
while (p < unrolled_end) {
STRINGLIB_CHAR bits = p[0] | p[1] | p[2] | p[3];
if (bits & mask) {
if (mask == mask_limit) {
/* Limit reached */
return max_char_limit;
}
if (mask == MASK_ASCII) {
max_char = MAX_CHAR_UCS1;
mask = MASK_UCS1;
}
else {
/* mask can't be MASK_UCS2 because of mask_limit above */
assert(mask == MASK_UCS1);
max_char = MAX_CHAR_UCS2;
mask = MASK_UCS2;
}
/* We check the new mask on the same chars in the next iteration */
continue;
}
p += 4;
}
while (p < end) {
if (p[0] & mask) {
if (mask == mask_limit) {
/* Limit reached */
return max_char_limit;
}
if (mask == MASK_ASCII) {
max_char = MAX_CHAR_UCS1;
mask = MASK_UCS1;
}
else {
/* mask can't be MASK_UCS2 because of mask_limit above */
assert(mask == MASK_UCS1);
max_char = MAX_CHAR_UCS2;
mask = MASK_UCS2;
}
/* We check the new mask on the same chars in the next iteration */
continue;
}
p++;
}
return max_char;
}
#undef MASK_ASCII
#undef MASK_UCS1
#undef MASK_UCS2
#undef MAX_CHAR_ASCII
#undef MAX_CHAR_UCS1
#undef MAX_CHAR_UCS2
#undef MAX_CHAR_UCS4
#endif /* STRINGLIB_SIZEOF_CHAR == 1 */
#endif /* STRINGLIB_IS_UNICODE */
|
