# Next tile on tempai

I tried this problem.

Your program receives the hand of mahjong. Returns the “Waiting style” of this hand. But there are some conditions as below.

• There are no Jihai, only Manzu
• Make Juntsu, Kohtsu and Head
• Regard different order waiting as the same
• If there are no waiting, no output

My source code are pushed this repository

import java.util.Scanner;

/**
* Created by Kai Sasaki on 3/19/14.
*/
public class Main {
public static void search(int tiles[], boolean isHead, String ans, int order) {
for (int i = 0; i < 9; i++) {
if (tiles[i] >= 3) {
// In order to remove practical same hands,
// this operation should be done before bigger values
if (i + 1 < order) {
return;
}

// In order to remove practical same hands,
// this operation should be done before finding head
return;
}

// Find *Kohtsu*
int tmp[] = tiles.clone();
tmp[i] -= 3;
String tmpAns = ans + String.format("(%d%d%d)", i + 1, i + 1, i + 1);
search(tmp, isHead, tmpAns, i + 1);
}
}

for (int i = 0; i < 7; i++) {
if (tiles[i] >= 1 && tiles[i + 1] >= 1 && tiles[i + 2] >= 1) {
// In order to remove practical same hands,
// this operation should be done before bigger values
if (i + 1 < order) {
return;
}

// In order to remove practical same hands,
// this operation should be done before finding head
return;
}

// Find *Juntsu*
int tmp[] = tiles.clone();
tmp[i] -= 1;
tmp[i + 1] -= 1;
tmp[i + 2] -= 1;
String tmpAns = ans + String.format("(%d%d%d)", i + 1, i + 2, i + 3);
search(tmp, isHead, tmpAns, i + 1);
}
}

for (int i = 0; i < 9; i++) {
if (tiles[i] >= 2 && !isHead) {
if (i + 1 < order) {
return;
}

int tmp[] = tiles.clone();
tmp[i] -= 2;
String tmpAns = ans + String.format("(%d%d)", i + 1, i + 1);
search(tmp, true, tmpAns, i + 1);
}
}

// No more mentsu
int oneCount = 0;
int twoCount = 0;
int sum = 0;
for (int i = 0; i < 9; i++) {
sum += tiles[i];
if (tiles[i] == 1) {
oneCount += 1;
} else if (tiles[i] == 2) {
twoCount += 1;
}
}

// 000100000
if (oneCount == 1 && sum == 1) {
for (int i = 0; i < 9; i++) {
if (tiles[i] == 1) {
ans += String.format("[%d]", i + 1);
System.out.println(ans);
return;
}
}
}

// 000001100
if (oneCount == 2 && sum == 2) {
for (int i = 0; i < 8; i++) {
if (tiles == 1 && tiles == 1) {
ans += "";
System.out.println(ans);
return;
} else if (tiles == 1 && tiles == 1) {
ans += "";
System.out.println(ans);
return;
} else if (tiles[i] == 1 && tiles[i + 1] == 1) {
ans += String.format("[%d%d]", i + 1, i + 2);
System.out.println(ans);
return;
}
}
}

if (twoCount == 1 && sum == 2) {
for (int i = 0; i < 9; i++) {
if (tiles[i] == 2) {
ans += String.format("[%d%d]", i + 1, i + 1);
System.out.println(ans);
return;
}
}
}
return;
}

public static void main(String[] args) {
Scanner sc = new Scanner(System.in);

// Receive string that represents hand values
String hand = sc.next();

int tiles[] = new int;

// Initialization
for (int i = 0; i < 9; i++) {
tiles[i] = 0;
}

// Setting tiles array
for (int i = 0; i < 13; i++) {
Integer tile = Integer.parseInt("" + hand.charAt(i));
tiles[tile - 1] += 1;
}

search(tiles, false, "", 1);
}
}


This is the simple depth first search algorithm. Ths main point of this code is in the main method. I expressed the data structure that represents Hand as the interger array. Each integer corresponds to the count of each tile. So in order to calculate the waiting tile, in this case, all you have to know is the count of each tile. With this data structure, you don’t need to retain complex structure. And also the operation such as finding Juntsu and so on is easy to execute bacause only increment or decrement of each value of this array.

It took me a long time but thanks to this training, a search algorithm such as DFS is no more alien to me. It’s friend!

Thank you.