根据VasyaNovikov的回答。改进了一些Java魔术:泛型和函数接口。

/**
 * Print a tree structure in a pretty ASCII fromat.
 * @param prefix Currnet previx. Use "" in initial call!
 * @param node The current node. Pass the root node of your tree in initial call.
 * @param getChildrenFunc A {@link Function} that returns the children of a given node.
 * @param isTail Is node the last of its sibblings. Use true in initial call. (This is needed for pretty printing.)
 * @param <T> The type of your nodes. Anything that has a toString can be used.
 */
private <T> void printTreeRec(String prefix, T node, Function<T, List<T>> getChildrenFunc, boolean isTail) {
    String nodeName = node.toString();
    String nodeConnection = isTail ? "└── " : "├── ";
    log.debug(prefix + nodeConnection + nodeName);
    List<T> children = getChildrenFunc.apply(node);
    for (int i = 0; i < children.size(); i++) {
        String newPrefix = prefix + (isTail ? "    " : "│   ");
        printTreeRec(newPrefix, children.get(i), getChildrenFunc, i == children.size()-1);
    }
}

初始调用示例:

Function<ChecksumModel, List<ChecksumModel>> getChildrenFunc = node -> getChildrenOf(node)
printTreeRec("", rootNode, getChildrenFunc, true);

将输出如下内容

└── rootNode
    ├── childNode1
    ├── childNode2
    │   ├── childNode2.1
    │   ├── childNode2.2
    │   └── childNode2.3
    ├── childNode3
    └── childNode4

public void printPreety() {
    List<TreeNode> list = new ArrayList<TreeNode>();
    list.add(head);
    printTree(list, getHeight(head));
}

public int getHeight(TreeNode head) {

    if (head == null) {
        return 0;
    } else {
        return 1 + Math.max(getHeight(head.left), getHeight(head.right));
    }
}

/**
 * pass head node in list and height of the tree 
 * 
 * @param levelNodes
 * @param level
 */
private void printTree(List<TreeNode> levelNodes, int level) {

    List<TreeNode> nodes = new ArrayList<TreeNode>();

    //indentation for first node in given level
    printIndentForLevel(level);

    for (TreeNode treeNode : levelNodes) {

        //print node data
        System.out.print(treeNode == null?" ":treeNode.data);

        //spacing between nodes
        printSpacingBetweenNodes(level);

        //if its not a leaf node
        if(level>1){
            nodes.add(treeNode == null? null:treeNode.left);
            nodes.add(treeNode == null? null:treeNode.right);
        }
    }
    System.out.println();

    if(level>1){        
        printTree(nodes, level-1);
    }
}

private void printIndentForLevel(int level){
    for (int i = (int) (Math.pow(2,level-1)); i >0; i--) {
        System.out.print(" ");
    }
}

private void printSpacingBetweenNodes(int level){
    //spacing between nodes
    for (int i = (int) ((Math.pow(2,level-1))*2)-1; i >0; i--) {
        System.out.print(" ");
    }
}


Prints Tree in following format:
                4                               
        3               7               
    1               5       8       
      2                       10   
                             9   

改编自Vasya Novikov的答案，使其更二进制，并使用StringBuilder提高效率(在Java中将String对象连接在一起通常效率很低)。

public StringBuilder toString(StringBuilder prefix, boolean isTail, StringBuilder sb) {
    if(right!=null) {
        right.toString(new StringBuilder().append(prefix).append(isTail ? "│   " : "    "), false, sb);
    }
    sb.append(prefix).append(isTail ? "└── " : "┌── ").append(value.toString()).append("\n");
    if(left!=null) {
        left.toString(new StringBuilder().append(prefix).append(isTail ? "    " : "│   "), true, sb);
    }
    return sb;
}

@Override
public String toString() {
    return this.toString(new StringBuilder(), true, new StringBuilder()).toString();
}

输出:

│       ┌── 7
│   ┌── 6
│   │   └── 5
└── 4
    │   ┌── 3
    └── 2
        └── 1
            └── 0

在控制台打印:

                                                500
                       700                                             300   
    200                                   400                                                                                          

简单代码:

public int getHeight()
    {
        if(rootNode == null) return -1;
        return getHeight(rootNode);
    }

    private int getHeight(Node node)
    {
        if(node == null) return -1;

        return Math.max(getHeight(node.left), getHeight(node.right)) + 1;
    }

    public void printBinaryTree(Node rootNode)
    {
        Queue<Node> rootsQueue = new LinkedList<Node>();
        Queue<Node> levelQueue = new LinkedList<Node>();
        levelQueue.add(rootNode);
        int treeHeight = getHeight();
        int firstNodeGap;
        int internalNodeGap;
        int copyinternalNodeGap;
        while(true)
        {
            System.out.println("");
            internalNodeGap = (int)(Math.pow(2, treeHeight + 1) -1);  
            copyinternalNodeGap = internalNodeGap;
            firstNodeGap = internalNodeGap/2;

            boolean levelFirstNode = true;

            while(!levelQueue.isEmpty())
            {
                internalNodeGap = copyinternalNodeGap;
                Node currNode = levelQueue.poll();
                if(currNode != null)
                {
                    if(levelFirstNode)
                    {
                        while(firstNodeGap > 0)
                        {
                            System.out.format("%s", "   ");
                            firstNodeGap--; 
                        }
                        levelFirstNode =false;
                    }
                    else
                    {
                        while(internalNodeGap>0)
                        {
                            internalNodeGap--;
                            System.out.format("%s", "   ");
                        }
                    }
                    System.out.format("%3d",currNode.data);
                    rootsQueue.add(currNode);
                }
            }

            --treeHeight;

            while(!rootsQueue.isEmpty())
            {
                Node currNode = rootsQueue.poll();
                if(currNode != null)
                {
                    levelQueue.add(currNode.left);
                    levelQueue.add(currNode.right);
                }
            }

            if(levelQueue.isEmpty()) break;
        }

    }

using map...
{
Map<Integer,String> m = new LinkedHashMap<>();

         tn.printNodeWithLvl(node,l,m);

        for(Entry<Integer, String> map :m.entrySet()) {
            System.out.println(map.getValue());
        }
then....method


   private  void printNodeWithLvl(Node node,int l,Map<Integer,String> m) {
       if(node==null) {
           return;
       }
      if(m.containsKey(l)) {
          m.put(l, new StringBuilder(m.get(l)).append(node.value).toString());
      }else {
          m.put(l, node.value+"");
      }
      l++;
      printNodeWithLvl( node.left,l,m);
      printNodeWithLvl(node.right,l,m);

    }
}

你的树每一层需要两倍的距离:

       a
      / \
     /   \
    /     \
   /       \
   b       c
  / \     / \
 /   \   /   \
 d   e   f   g
/ \ / \ / \ / \
h i j k l m n o

你可以将你的树保存在一个数组的数组中，每个数组对应一个深度:

[[a],[b,c],[d,e,f,g],[h,i,j,k,l,m,n,o]]

如果你的树没有满，你需要在数组中包含空值:

       a
      / \
     /   \
    /     \
   /       \
   b       c
  / \     / \
 /   \   /   \
 d   e   f   g
/ \   \ / \   \
h i   k l m   o
[[a],[b,c],[d,e,f,g],[h,i, ,k,l,m, ,o]]

然后你可以遍历数组来打印你的树，根据深度打印第一个元素之前和元素之间的空格，根据下一层数组中对应的元素是否被填充打印行。 如果您的值可以超过一个字符长，您需要在创建数组表示时找到最长的值，并相应地乘以所有宽度和行数。