1 /* Substring search in a NUL terminated string of 'char' elements,
2 using the Knuth-Morris-Pratt algorithm.
3 Copyright (C) 2005-2007 Free Software Foundation, Inc.
4 Written by Bruno Haible <bruno@clisp.org>, 2005.
6 This program is free software; you can redistribute it and/or modify
7 it under the terms of the GNU General Public License as published by
8 the Free Software Foundation; either version 2, or (at your option)
11 This program is distributed in the hope that it will be useful,
12 but WITHOUT ANY WARRANTY; without even the implied warranty of
13 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
14 GNU General Public License for more details.
16 You should have received a copy of the GNU General Public License
17 along with this program; if not, write to the Free Software Foundation,
18 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
20 /* Before including this file, you need to define:
21 CANON_ELEMENT(c) A macro that canonicalizes an element right after
22 it has been fetched from one of the two strings.
23 The argument is an 'unsigned char'; the result
24 must be an 'unsigned char' as well. */
26 /* Knuth-Morris-Pratt algorithm.
27 See http://en.wikipedia.org/wiki/Knuth-Morris-Pratt_algorithm
28 Return a boolean indicating success. */
30 knuth_morris_pratt_unibyte (const char *haystack, const char *needle,
33 size_t m = strlen (needle);
35 /* Allocate the table. */
36 size_t *table = (size_t *) nmalloca (m, sizeof (size_t));
41 0 < table[i] <= i is defined such that
42 forall 0 < x < table[i]: needle[x..i-1] != needle[0..i-1-x],
43 and table[i] is as large as possible with this property.
47 needle[table[i]..i-1] = needle[0..i-1-table[i]].
49 rhaystack[0..i-1] == needle[0..i-1]
50 and exists h, i <= h < m: rhaystack[h] != needle[h]
52 forall 0 <= x < table[i]: rhaystack[x..x+m-1] != needle[0..m-1].
53 table[0] remains uninitialized. */
57 /* i = 1: Nothing to verify for x = 0. */
61 for (i = 2; i < m; i++)
63 /* Here: j = i-1 - table[i-1].
64 The inequality needle[x..i-1] != needle[0..i-1-x] is known to hold
65 for x < table[i-1], by induction.
66 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
67 unsigned char b = CANON_ELEMENT ((unsigned char) needle[i - 1]);
71 /* Invariants: The inequality needle[x..i-1] != needle[0..i-1-x]
72 is known to hold for x < i-1-j.
73 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
74 if (b == CANON_ELEMENT ((unsigned char) needle[j]))
76 /* Set table[i] := i-1-j. */
80 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
81 for x = i-1-j, because
82 needle[i-1] != needle[j] = needle[i-1-x]. */
85 /* The inequality holds for all possible x. */
89 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
90 for i-1-j < x < i-1-j+table[j], because for these x:
92 = needle[x-(i-1-j)..j-1]
93 != needle[0..j-1-(x-(i-1-j))] (by definition of table[j])
95 hence needle[x..i-1] != needle[0..i-1-x].
97 needle[i-1-j+table[j]..i-2]
98 = needle[table[j]..j-1]
99 = needle[0..j-1-table[j]] (by definition of table[j]). */
102 /* Here: j = i - table[i]. */
106 /* Search, using the table to accelerate the processing. */
109 const char *rhaystack;
110 const char *phaystack;
114 rhaystack = haystack;
115 phaystack = haystack;
116 /* Invariant: phaystack = rhaystack + j. */
117 while (*phaystack != '\0')
118 if (CANON_ELEMENT ((unsigned char) needle[j])
119 == CANON_ELEMENT ((unsigned char) *phaystack))
125 /* The entire needle has been found. */
126 *resultp = rhaystack;
132 /* Found a match of needle[0..j-1], mismatch at needle[j]. */
133 rhaystack += table[j];
138 /* Found a mismatch at needle[0] already. */