1 /* Substring search in a NUL terminated string of UNIT elements,
2 using the Knuth-Morris-Pratt algorithm.
3 Copyright (C) 2005-2011 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 UNIT The element type of the needle and haystack.
22 CANON_ELEMENT(c) A macro that canonicalizes an element right after
23 it has been fetched from needle or haystack.
24 The argument is of type UNIT; the result must be
25 of type UNIT as well. */
27 /* Knuth-Morris-Pratt algorithm.
28 See http://en.wikipedia.org/wiki/Knuth-Morris-Pratt_algorithm
29 HAYSTACK is the NUL terminated string in which to search for.
30 NEEDLE is the string to search for in HAYSTACK, consisting of NEEDLE_LEN
32 Return a boolean indicating success:
33 Return true and set *RESULTP if the search was completed.
34 Return false if it was aborted because not enough memory was available. */
36 knuth_morris_pratt (const UNIT *haystack,
37 const UNIT *needle, size_t needle_len,
40 size_t m = needle_len;
42 /* Allocate the table. */
43 size_t *table = (size_t *) nmalloca (m, sizeof (size_t));
48 0 < table[i] <= i is defined such that
49 forall 0 < x < table[i]: needle[x..i-1] != needle[0..i-1-x],
50 and table[i] is as large as possible with this property.
54 needle[table[i]..i-1] = needle[0..i-1-table[i]].
56 rhaystack[0..i-1] == needle[0..i-1]
57 and exists h, i <= h < m: rhaystack[h] != needle[h]
59 forall 0 <= x < table[i]: rhaystack[x..x+m-1] != needle[0..m-1].
60 table[0] remains uninitialized. */
64 /* i = 1: Nothing to verify for x = 0. */
68 for (i = 2; i < m; i++)
70 /* Here: j = i-1 - table[i-1].
71 The inequality needle[x..i-1] != needle[0..i-1-x] is known to hold
72 for x < table[i-1], by induction.
73 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
74 UNIT b = CANON_ELEMENT (needle[i - 1]);
78 /* Invariants: The inequality needle[x..i-1] != needle[0..i-1-x]
79 is known to hold for x < i-1-j.
80 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
81 if (b == CANON_ELEMENT (needle[j]))
83 /* Set table[i] := i-1-j. */
87 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
88 for x = i-1-j, because
89 needle[i-1] != needle[j] = needle[i-1-x]. */
92 /* The inequality holds for all possible x. */
96 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
97 for i-1-j < x < i-1-j+table[j], because for these x:
99 = needle[x-(i-1-j)..j-1]
100 != needle[0..j-1-(x-(i-1-j))] (by definition of table[j])
102 hence needle[x..i-1] != needle[0..i-1-x].
104 needle[i-1-j+table[j]..i-2]
105 = needle[table[j]..j-1]
106 = needle[0..j-1-table[j]] (by definition of table[j]). */
109 /* Here: j = i - table[i]. */
113 /* Search, using the table to accelerate the processing. */
116 const UNIT *rhaystack;
117 const UNIT *phaystack;
121 rhaystack = haystack;
122 phaystack = haystack;
123 /* Invariant: phaystack = rhaystack + j. */
124 while (*phaystack != 0)
125 if (CANON_ELEMENT (needle[j]) == CANON_ELEMENT (*phaystack))
131 /* The entire needle has been found. */
132 *resultp = rhaystack;
138 /* Found a match of needle[0..j-1], mismatch at needle[j]. */
139 rhaystack += table[j];
144 /* Found a mismatch at needle[0] already. */