2017-10-02 07:45:16 +08:00
|
|
|
|
/*
|
|
|
|
|
Implementation of Knuth–Morris–Pratt algorithm
|
2018-10-15 16:52:08 +08:00
|
|
|
|
Usage: see the main function for an example
|
2017-10-02 07:45:16 +08:00
|
|
|
|
*/
|
|
|
|
|
public class KMP {
|
2018-10-15 16:52:08 +08:00
|
|
|
|
//a working example
|
|
|
|
|
public static void main(String[] args) {
|
|
|
|
|
final String haystack = "AAAAABAAABA"; //This is the full string
|
|
|
|
|
final String needle = "AAAA"; //This is the substring that we want to find
|
|
|
|
|
KMPmatcher(haystack, needle);
|
|
|
|
|
}
|
|
|
|
|
// find the starting index in string haystack[] that matches the search word P[]
|
|
|
|
|
public static void KMPmatcher(final String haystack, final String needle) {
|
|
|
|
|
final int m = haystack.length();
|
|
|
|
|
final int n = needle.length();
|
|
|
|
|
final int[] pi = computePrefixFunction(needle);
|
2017-10-02 07:45:16 +08:00
|
|
|
|
int q = 0;
|
|
|
|
|
for (int i = 0; i < m; i++) {
|
2018-10-15 16:52:08 +08:00
|
|
|
|
while (q > 0 && haystack.charAt(i) != needle.charAt(q)) {
|
2017-10-02 07:45:16 +08:00
|
|
|
|
q = pi[q - 1];
|
|
|
|
|
}
|
|
|
|
|
|
2018-10-15 16:52:08 +08:00
|
|
|
|
if (haystack.charAt(i) == needle.charAt(q)) {
|
2017-10-02 07:45:16 +08:00
|
|
|
|
q++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (q == n) {
|
|
|
|
|
System.out.println("Pattern starts: " + (i + 1 - n));
|
|
|
|
|
q = pi[q - 1];
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
// return the prefix function
|
2018-10-15 16:52:08 +08:00
|
|
|
|
private static int[] computePrefixFunction(final String P) {
|
2017-10-02 07:45:16 +08:00
|
|
|
|
final int n = P.length();
|
|
|
|
|
final int[] pi = new int[n];
|
|
|
|
|
pi[0] = 0;
|
|
|
|
|
int q = 0;
|
|
|
|
|
for (int i = 1; i < n; i++) {
|
|
|
|
|
while (q > 0 && P.charAt(q) != P.charAt(i)) {
|
|
|
|
|
q = pi[q - 1];
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (P.charAt(q) == P.charAt(i)) {
|
|
|
|
|
q++;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
pi[i] = q;
|
|
|
|
|
|
|
|
|
|
}
|
|
|
|
|
return pi;
|
|
|
|
|
}
|
2018-10-15 16:52:08 +08:00
|
|
|
|
}
|