java - 我的 Boggle 程序找不到所有有效单词 - Java

Question

所以我正在用java编写一个Boggle程序，但我很难让它找到每个可能的单词。

它几乎找到了所有这些，但我一生都无法弄清楚为什么它不能找到所有这些。

private void findWords( TreeSet<String> foundWords, String word, Tile t ){
    int i=t.getRow();
    int j=t.getCol();

    //Make sure the tile isn't visited
    if(visited[i][j]) return;

    //Set the current tile to visited
    visited[i][j]=true;

    //Decide what the current word is
    if(t.getLetter().equalsIgnoreCase("q")) word=word+"qu";
    else word=word+t.getLetter();

    //If the string is a word
    if(Boggle.dictionary.search(word)==1){
        //If the word length is greater than or equal to the prefix length
        if(word.length()>=Dictionary.prefixLength){
            //If the word has not already been found
            if(!foundWords.contains(word)){
                //Add the word to the found list
                foundWords.add(word);   
            }
        }
    }

    //Recurse through all neighbor tiles
    for(int curRow=0; curRow<=nRows; curRow++){
        for(int curCol=0; curCol<=nCols; curCol++){
            //Make sure it is not out of bounds
            if((i+curRow<nRows)&&(j+curCol<nCols)){
                findWords(foundWords, word, board[i + curRow][j + curCol]); 
                findWords(foundWords, word, board[i - curRow][j - curCol]); 

                findWords(foundWords, word, board[i + curRow][curCol]); 
                findWords(foundWords, word, board[i - curRow][curCol]); 

                findWords(foundWords, word, board[curRow][j + curCol]); 
                findWords(foundWords, word, board[curRow][j - curCol]); 

                findWords(foundWords, word, board[i + curRow][j - curCol]);
                findWords(foundWords, word, board[i - curRow][j + curCol]);
            }
        }
    }

    //Reset the tile to be not visited
    visited[i][j] = false;
}

这是有问题的方法；它会递归地在 Boggle 板上查找单词。

有人知道为什么它只找到大约 75% 的单词吗？

如果需要，我可以附加更多代码。

Answer 1

我建议编写一些测试用例 - 我总是发现这样做可以立即找到问题，或者至少允许您使用调试器单步执行代码，并找到现实与您的期望不同的地方。

编辑:另外，你的两个 for 循环看起来很奇怪。您不应该查看每个方向上 -1,0 和 1 的偏移量(并折扣 0,0)，而不是 0 -> nRows 吗？看来您只朝一个方向看。