1 /* Substring search in a NUL terminated string of 'char' elements,
2 using the Knuth-Morris-Pratt algorithm.
3 Copyright (C) 2005-2010 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:
29 Return true and set *RESULTP if the search was completed.
30 Return false if it was aborted because not enough memory was available. */
32 knuth_morris_pratt_unibyte (const char *haystack, const char *needle,
35 size_t m = strlen (needle);
37 /* Allocate the table. */
38 size_t *table = (size_t *) nmalloca (m, sizeof (size_t));
43 0 < table[i] <= i is defined such that
44 forall 0 < x < table[i]: needle[x..i-1] != needle[0..i-1-x],
45 and table[i] is as large as possible with this property.
49 needle[table[i]..i-1] = needle[0..i-1-table[i]].
51 rhaystack[0..i-1] == needle[0..i-1]
52 and exists h, i <= h < m: rhaystack[h] != needle[h]
54 forall 0 <= x < table[i]: rhaystack[x..x+m-1] != needle[0..m-1].
55 table[0] remains uninitialized. */
59 /* i = 1: Nothing to verify for x = 0. */
63 for (i = 2; i < m; i++)
65 /* Here: j = i-1 - table[i-1].
66 The inequality needle[x..i-1] != needle[0..i-1-x] is known to hold
67 for x < table[i-1], by induction.
68 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
69 unsigned char b = CANON_ELEMENT ((unsigned char) needle[i - 1]);
73 /* Invariants: The inequality needle[x..i-1] != needle[0..i-1-x]
74 is known to hold for x < i-1-j.
75 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
76 if (b == CANON_ELEMENT ((unsigned char) needle[j]))
78 /* Set table[i] := i-1-j. */
82 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
83 for x = i-1-j, because
84 needle[i-1] != needle[j] = needle[i-1-x]. */
87 /* The inequality holds for all possible x. */
91 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
92 for i-1-j < x < i-1-j+table[j], because for these x:
94 = needle[x-(i-1-j)..j-1]
95 != needle[0..j-1-(x-(i-1-j))] (by definition of table[j])
97 hence needle[x..i-1] != needle[0..i-1-x].
99 needle[i-1-j+table[j]..i-2]
100 = needle[table[j]..j-1]
101 = needle[0..j-1-table[j]] (by definition of table[j]). */
104 /* Here: j = i - table[i]. */
108 /* Search, using the table to accelerate the processing. */
111 const char *rhaystack;
112 const char *phaystack;
116 rhaystack = haystack;
117 phaystack = haystack;
118 /* Invariant: phaystack = rhaystack + j. */
119 while (*phaystack != '\0')
120 if (CANON_ELEMENT ((unsigned char) needle[j])
121 == CANON_ELEMENT ((unsigned char) *phaystack))
127 /* The entire needle has been found. */
128 *resultp = rhaystack;
134 /* Found a match of needle[0..j-1], mismatch at needle[j]. */
135 rhaystack += table[j];
140 /* Found a mismatch at needle[0] already. */