1 /* $OpenBSD: strstr.c,v 1.7 2017/04/12 16:06:12 millert Exp $ */ 2 3 /* 4 * Copyright (c) 2005-2014 Rich Felker 5 * 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: 13 * 14 * The above copyright notice and this permission notice shall be 15 * included in all copies or substantial portions of the Software. 16 * 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. 24 */ 25 26 #include <string.h> 27 #include <stdlib.h> 28 #include <stdint.h> 29 30 static char * 31 twobyte_strstr(const unsigned char *h, const unsigned char *n) 32 { 33 uint16_t nw = n[0]<<8 | n[1], hw = h[0]<<8 | h[1]; 34 for (h++; *h && hw != nw; hw = hw<<8 | *++h); 35 return *h ? (char *)h-1 : 0; 36 } 37 38 static char * 39 threebyte_strstr(const unsigned char *h, const unsigned char *n) 40 { 41 uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8; 42 uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8; 43 for (h+=2; *h && hw != nw; hw = (hw|*++h)<<8); 44 return *h ? (char *)h-2 : 0; 45 } 46 47 static char * 48 fourbyte_strstr(const unsigned char *h, const unsigned char *n) 49 { 50 uint32_t nw = n[0]<<24 | n[1]<<16 | n[2]<<8 | n[3]; 51 uint32_t hw = h[0]<<24 | h[1]<<16 | h[2]<<8 | h[3]; 52 for (h+=3; *h && hw != nw; hw = hw<<8 | *++h); 53 return *h ? (char *)h-3 : 0; 54 } 55 56 #define MAX(a,b) ((a)>(b)?(a):(b)) 57 #define MIN(a,b) ((a)<(b)?(a):(b)) 58 59 #define BITOP(a,b,op) \ 60 ((a)[(size_t)(b)/(8*sizeof *(a))] op (size_t)1<<((size_t)(b)%(8*sizeof *(a)))) 61 62 /* 63 * Maxime Crochemore and Dominique Perrin, Two-way string-matching, 64 * Journal of the ACM, 38(3):651-675, July 1991. 65 */ 66 static char * 67 twoway_strstr(const unsigned char *h, const unsigned char *n) 68 { 69 const unsigned char *z; 70 size_t l, ip, jp, k, p, ms, p0, mem, mem0; 71 size_t byteset[32 / sizeof(size_t)] = { 0 }; 72 size_t shift[256]; 73 74 /* Computing length of needle and fill shift table */ 75 for (l=0; n[l] && h[l]; l++) 76 BITOP(byteset, n[l], |=), shift[n[l]] = l+1; 77 if (n[l]) return 0; /* hit the end of h */ 78 79 /* Compute maximal suffix */ 80 ip = -1; jp = 0; k = p = 1; 81 while (jp+k<l) { 82 if (n[ip+k] == n[jp+k]) { 83 if (k == p) { 84 jp += p; 85 k = 1; 86 } else k++; 87 } else if (n[ip+k] > n[jp+k]) { 88 jp += k; 89 k = 1; 90 p = jp - ip; 91 } else { 92 ip = jp++; 93 k = p = 1; 94 } 95 } 96 ms = ip; 97 p0 = p; 98 99 /* And with the opposite comparison */ 100 ip = -1; jp = 0; k = p = 1; 101 while (jp+k<l) { 102 if (n[ip+k] == n[jp+k]) { 103 if (k == p) { 104 jp += p; 105 k = 1; 106 } else k++; 107 } else if (n[ip+k] < n[jp+k]) { 108 jp += k; 109 k = 1; 110 p = jp - ip; 111 } else { 112 ip = jp++; 113 k = p = 1; 114 } 115 } 116 if (ip+1 > ms+1) ms = ip; 117 else p = p0; 118 119 /* Periodic needle? */ 120 if (memcmp(n, n+p, ms+1)) { 121 mem0 = 0; 122 p = MAX(ms, l-ms-1) + 1; 123 } else mem0 = l-p; 124 mem = 0; 125 126 /* Initialize incremental end-of-haystack pointer */ 127 z = h; 128 129 /* Search loop */ 130 for (;;) { 131 /* Update incremental end-of-haystack pointer */ 132 if (z-h < l) { 133 /* Fast estimate for MIN(l,63) */ 134 size_t grow = l | 63; 135 const unsigned char *z2 = memchr(z, 0, grow); 136 if (z2) { 137 z = z2; 138 if (z-h < l) return 0; 139 } else z += grow; 140 } 141 142 /* Check last byte first; advance by shift on mismatch */ 143 if (BITOP(byteset, h[l-1], &)) { 144 k = l-shift[h[l-1]]; 145 #ifdef DEBUG 146 printf("adv by %zu (on %c) at [%s] (%zu;l=%zu)\n", k, h[l-1], h, shift[h[l-1]], l); 147 #endif 148 if (k) { 149 if (mem0 && mem && k < p) k = l-p; 150 h += k; 151 mem = 0; 152 continue; 153 } 154 } else { 155 h += l; 156 mem = 0; 157 continue; 158 } 159 160 /* Compare right half */ 161 for (k=MAX(ms+1,mem); n[k] && n[k] == h[k]; k++); 162 if (n[k]) { 163 h += k-ms; 164 mem = 0; 165 continue; 166 } 167 /* Compare left half */ 168 for (k=ms+1; k>mem && n[k-1] == h[k-1]; k--); 169 if (k <= mem) return (char *)h; 170 h += p; 171 mem = mem0; 172 } 173 } 174 175 char * 176 strstr(const char *h, const char *n) 177 { 178 /* Return immediately on empty needle */ 179 if (!n[0]) return (char *)h; 180 181 /* Use faster algorithms for short needles */ 182 h = strchr(h, *n); 183 if (!h || !n[1]) return (char *)h; 184 if (!h[1]) return 0; 185 if (!n[2]) return twobyte_strstr((void *)h, (void *)n); 186 if (!h[2]) return 0; 187 if (!n[3]) return threebyte_strstr((void *)h, (void *)n); 188 if (!h[3]) return 0; 189 if (!n[4]) return fourbyte_strstr((void *)h, (void *)n); 190 191 return twoway_strstr((void *)h, (void *)n); 192 } 193 DEF_STRONG(strstr); 194