1 /* -*- buffer-read-only: t -*- vi: set ro: */
2 /* DO NOT EDIT! GENERATED AUTOMATICALLY! */
3 /* Substring search in a NUL terminated string of UNIT elements,
4 using the Knuth-Morris-Pratt algorithm.
5 Copyright (C) 2005-2011 Free Software Foundation, Inc.
6 Written by Bruno Haible <bruno@clisp.org>, 2005.
8 This program is free software; you can redistribute it and/or modify
9 it under the terms of the GNU General Public License as published by
10 the Free Software Foundation; either version 3, or (at your option)
13 This program is distributed in the hope that it will be useful,
14 but WITHOUT ANY WARRANTY; without even the implied warranty of
15 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
16 GNU General Public License for more details.
18 You should have received a copy of the GNU General Public License
19 along with this program; if not, write to the Free Software Foundation,
20 Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */
22 /* Before including this file, you need to define:
23 UNIT The element type of the needle and haystack.
24 CANON_ELEMENT(c) A macro that canonicalizes an element right after
25 it has been fetched from needle or haystack.
26 The argument is of type UNIT; the result must be
27 of type UNIT as well. */
29 /* Knuth-Morris-Pratt algorithm.
30 See http://en.wikipedia.org/wiki/Knuth-Morris-Pratt_algorithm
31 HAYSTACK is the NUL terminated string in which to search for.
32 NEEDLE is the string to search for in HAYSTACK, consisting of NEEDLE_LEN
34 Return a boolean indicating success:
35 Return true and set *RESULTP if the search was completed.
36 Return false if it was aborted because not enough memory was available. */
38 knuth_morris_pratt (const UNIT *haystack,
39 const UNIT *needle, size_t needle_len,
42 size_t m = needle_len;
44 /* Allocate the table. */
45 size_t *table = (size_t *) nmalloca (m, sizeof (size_t));
50 0 < table[i] <= i is defined such that
51 forall 0 < x < table[i]: needle[x..i-1] != needle[0..i-1-x],
52 and table[i] is as large as possible with this property.
56 needle[table[i]..i-1] = needle[0..i-1-table[i]].
58 rhaystack[0..i-1] == needle[0..i-1]
59 and exists h, i <= h < m: rhaystack[h] != needle[h]
61 forall 0 <= x < table[i]: rhaystack[x..x+m-1] != needle[0..m-1].
62 table[0] remains uninitialized. */
66 /* i = 1: Nothing to verify for x = 0. */
70 for (i = 2; i < m; i++)
72 /* Here: j = i-1 - table[i-1].
73 The inequality needle[x..i-1] != needle[0..i-1-x] is known to hold
74 for x < table[i-1], by induction.
75 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
76 UNIT b = CANON_ELEMENT (needle[i - 1]);
80 /* Invariants: The inequality needle[x..i-1] != needle[0..i-1-x]
81 is known to hold for x < i-1-j.
82 Furthermore, if j>0: needle[i-1-j..i-2] = needle[0..j-1]. */
83 if (b == CANON_ELEMENT (needle[j]))
85 /* Set table[i] := i-1-j. */
89 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
90 for x = i-1-j, because
91 needle[i-1] != needle[j] = needle[i-1-x]. */
94 /* The inequality holds for all possible x. */
98 /* The inequality needle[x..i-1] != needle[0..i-1-x] also holds
99 for i-1-j < x < i-1-j+table[j], because for these x:
101 = needle[x-(i-1-j)..j-1]
102 != needle[0..j-1-(x-(i-1-j))] (by definition of table[j])
104 hence needle[x..i-1] != needle[0..i-1-x].
106 needle[i-1-j+table[j]..i-2]
107 = needle[table[j]..j-1]
108 = needle[0..j-1-table[j]] (by definition of table[j]). */
111 /* Here: j = i - table[i]. */
115 /* Search, using the table to accelerate the processing. */
118 const UNIT *rhaystack;
119 const UNIT *phaystack;
123 rhaystack = haystack;
124 phaystack = haystack;
125 /* Invariant: phaystack = rhaystack + j. */
126 while (*phaystack != 0)
127 if (CANON_ELEMENT (needle[j]) == CANON_ELEMENT (*phaystack))
133 /* The entire needle has been found. */
134 *resultp = rhaystack;
140 /* Found a match of needle[0..j-1], mismatch at needle[j]. */
141 rhaystack += table[j];
146 /* Found a mismatch at needle[0] already. */