字符串匹配算法之：有限状态自动机

有限状态机的详细说明参见此链接

public class Test {

    private static int alphaSize = 3; //3代表只有abc
    private static HashMap<Integer, HashMap<Character, Integer>> jumpTable = new HashMap<>(); //跳转表
    private String P;

    public Test(String p) {
        P = p;
        makeJumpTable();
    }
		//创建跳转表
    private void makeJumpTable() {
        for (int i = 0; i < P.length(); i++) {
            for (int j = 0; j < alphaSize; j++) {
                char c = (char) ('a' + j);
                String Pq = P.substring(0, i) + c;  //pq+s
                int nextState = findSuffix(Pq);
                System.out.println("from state " + i + " receive input char " + c + " jump to state " + nextState);
                HashMap<Character, Integer> map = jumpTable.get(i);
                if (map == null) {
                    map = new HashMap<>();
                }
                map.put(c, nextState);
                jumpTable.put(i, map);
            }
        }

    }
    //字符串P从第一个字符开始，连续几个字符所构成的字符串可以成为S的后缀
    private int findSuffix(String s) {
        int suffixLen = 0;
        int i = 0;
        while (i < P.length() && i < s.length()) {
            if (P.charAt(i) != s.charAt(i)) {
                break;
            }
            i++;

        }
        suffixLen = i;
        return suffixLen;
    }
  //查表是否匹配
    public int match(String Text) {
        Integer q = 0;
        HashMap<Character, Integer> map = jumpTable.get(q);
        for (int i = 0; i < Text.length(); i++) {
            int oldState = q;
            q = map.get(Text.charAt(i));
            if (q == null) {
                return -1;
            }
            System.out.println("In state " + oldState + " receive input " + Text.charAt(i) + " jump to state " + q);
            if (q == P.length()) {
                return q;
            }
        }
        return -1;

    }

    public static void main(String[] args) {
		//要匹配的串
        StringAutomaton p = new StringAutomaton("abaa");
        //文本
        int i = p.match("bbaabaas");
        System.out.println(i); //4  不代表0-4匹配
    }
}