2 * SPDX-License-Identifier: BSD-3-Clause
4 * Copyright (c) 2005-2014 Rich Felker, et al.
6 * Permission is hereby granted, free of charge, to any person obtaining
7 * a copy of this software and associated documentation files (the
8 * "Software"), to deal in the Software without restriction, including
9 * without limitation the rights to use, copy, modify, merge, publish,
10 * distribute, sublicense, and/or sell copies of the Software, and to
11 * permit persons to whom the Software is furnished to do so, subject to
12 * the following conditions:
14 * The above copyright notice and this permission notice shall be
15 * included in all copies or substantial portions of the Software.
17 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
18 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
19 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
20 * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
21 * CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
22 * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
23 * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
25 #include <sys/cdefs.h>
26 __FBSDID("$FreeBSD$");
32 twobyte_strstr(const unsigned char *h, const unsigned char *n)
34 uint16_t nw = n[0] << 8 | n[1], hw = h[0] << 8 | h[1];
35 for (h++; *h && hw != nw; hw = hw << 8 | *++h)
37 return *h ? (char *)h - 1 : 0;
41 threebyte_strstr(const unsigned char *h, const unsigned char *n)
43 uint32_t nw = (uint32_t)n[0] << 24 | n[1] << 16 | n[2] << 8;
44 uint32_t hw = (uint32_t)h[0] << 24 | h[1] << 16 | h[2] << 8;
45 for (h += 2; *h && hw != nw; hw = (hw | *++h) << 8)
47 return *h ? (char *)h - 2 : 0;
51 fourbyte_strstr(const unsigned char *h, const unsigned char *n)
53 uint32_t nw = (uint32_t)n[0] << 24 | n[1] << 16 | n[2] << 8 | n[3];
54 uint32_t hw = (uint32_t)h[0] << 24 | h[1] << 16 | h[2] << 8 | h[3];
55 for (h += 3; *h && hw != nw; hw = hw << 8 | *++h)
57 return *h ? (char *)h - 3 : 0;
60 #define MAX(a, b) ((a) > (b) ? (a) : (b))
61 #define MIN(a, b) ((a) < (b) ? (a) : (b))
63 #define BITOP(a, b, op) \
64 ((a)[(size_t)(b) / (8 * sizeof *(a))] op \
65 (size_t)1 << ((size_t)(b) % (8 * sizeof *(a))))
68 * Two Way string search algorithm, with a bad shift table applied to the last
69 * byte of the window. A bit array marks which entries in the shift table are
70 * initialized to avoid fully initializing a 1kb/2kb table.
72 * Reference: CROCHEMORE M., PERRIN D., 1991, Two-way string-matching,
73 * Journal of the ACM 38(3):651-675
76 twoway_strstr(const unsigned char *h, const unsigned char *n)
78 const unsigned char *z;
79 size_t l, ip, jp, k, p, ms, p0, mem, mem0;
80 size_t byteset[32 / sizeof(size_t)] = { 0 };
83 /* Computing length of needle and fill shift table */
84 for (l = 0; n[l] && h[l]; l++)
85 BITOP(byteset, n[l], |=), shift[n[l]] = l + 1;
87 return 0; /* hit the end of h */
89 /* Compute maximal suffix */
94 if (n[ip + k] == n[jp + k]) {
100 } else if (n[ip + k] > n[jp + k]) {
112 /* And with the opposite comparison */
117 if (n[ip + k] == n[jp + k]) {
123 } else if (n[ip + k] < n[jp + k]) {
137 /* Periodic needle? */
138 if (memcmp(n, n + p, ms + 1)) {
140 p = MAX(ms, l - ms - 1) + 1;
145 /* Initialize incremental end-of-haystack pointer */
150 /* Update incremental end-of-haystack pointer */
152 /* Fast estimate for MIN(l,63) */
153 size_t grow = l | 63;
154 const unsigned char *z2 = memchr(z, 0, grow);
163 /* Check last byte first; advance by shift on mismatch */
164 if (BITOP(byteset, h[l - 1], &)) {
165 k = l - shift[h[l - 1]];
179 /* Compare right half */
180 for (k = MAX(ms + 1, mem); n[k] && n[k] == h[k]; k++)
187 /* Compare left half */
188 for (k = ms + 1; k > mem && n[k - 1] == h[k - 1]; k--)
198 strstr(const char *h, const char *n)
200 /* Return immediately on empty needle */
204 /* Use faster algorithms for short needles */
211 return twobyte_strstr((void *)h, (void *)n);
215 return threebyte_strstr((void *)h, (void *)n);
219 return fourbyte_strstr((void *)h, (void *)n);
221 return twoway_strstr((void *)h, (void *)n);