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。

一个Scala解决方案，改编自Vasya Novikov的答案，专门用于二叉树:

/** An immutable Binary Tree. */
case class BTree[T](value: T, left: Option[BTree[T]], right: Option[BTree[T]]) {

  /* Adapted from: http://stackoverflow.com/a/8948691/643684 */
  def pretty: String = {
    def work(tree: BTree[T], prefix: String, isTail: Boolean): String = {
      val (line, bar) = if (isTail) ("└── ", " ") else ("├── ", "│")

      val curr = s"${prefix}${line}${tree.value}"

      val rights = tree.right match {
        case None    => s"${prefix}${bar}   ├── ∅"
        case Some(r) => work(r, s"${prefix}${bar}   ", false)
      }

      val lefts = tree.left match {
        case None    => s"${prefix}${bar}   └── ∅"
        case Some(l) => work(l, s"${prefix}${bar}   ", true)
      }

      s"${curr}\n${rights}\n${lefts}"

    }

    work(this, "", true)
  }
}

我为此做了一个改进的算法，可以很好地处理不同大小的节点。它使用行自上而下地打印。

package alg;

import java.util.ArrayList;
import java.util.List;


/**
 * Binary tree printer
 * 
 * @author MightyPork
 */
public class TreePrinter
{
    /** Node that can be printed */
    public interface PrintableNode
    {
        /** Get left child */
        PrintableNode getLeft();


        /** Get right child */
        PrintableNode getRight();


        /** Get text to be printed */
        String getText();
    }


    /**
     * Print a tree
     * 
     * @param root
     *            tree root node
     */
    public static void print(PrintableNode root)
    {
        List<List<String>> lines = new ArrayList<List<String>>();

        List<PrintableNode> level = new ArrayList<PrintableNode>();
        List<PrintableNode> next = new ArrayList<PrintableNode>();

        level.add(root);
        int nn = 1;

        int widest = 0;

        while (nn != 0) {
            List<String> line = new ArrayList<String>();

            nn = 0;

            for (PrintableNode n : level) {
                if (n == null) {
                    line.add(null);

                    next.add(null);
                    next.add(null);
                } else {
                    String aa = n.getText();
                    line.add(aa);
                    if (aa.length() > widest) widest = aa.length();

                    next.add(n.getLeft());
                    next.add(n.getRight());

                    if (n.getLeft() != null) nn++;
                    if (n.getRight() != null) nn++;
                }
            }

            if (widest % 2 == 1) widest++;

            lines.add(line);

            List<PrintableNode> tmp = level;
            level = next;
            next = tmp;
            next.clear();
        }

        int perpiece = lines.get(lines.size() - 1).size() * (widest + 4);
        for (int i = 0; i < lines.size(); i++) {
            List<String> line = lines.get(i);
            int hpw = (int) Math.floor(perpiece / 2f) - 1;

            if (i > 0) {
                for (int j = 0; j < line.size(); j++) {

                    // split node
                    char c = ' ';
                    if (j % 2 == 1) {
                        if (line.get(j - 1) != null) {
                            c = (line.get(j) != null) ? '┴' : '┘';
                        } else {
                            if (j < line.size() && line.get(j) != null) c = '└';
                        }
                    }
                    System.out.print(c);

                    // lines and spaces
                    if (line.get(j) == null) {
                        for (int k = 0; k < perpiece - 1; k++) {
                            System.out.print(" ");
                        }
                    } else {

                        for (int k = 0; k < hpw; k++) {
                            System.out.print(j % 2 == 0 ? " " : "─");
                        }
                        System.out.print(j % 2 == 0 ? "┌" : "┐");
                        for (int k = 0; k < hpw; k++) {
                            System.out.print(j % 2 == 0 ? "─" : " ");
                        }
                    }
                }
                System.out.println();
            }

            // print line of numbers
            for (int j = 0; j < line.size(); j++) {

                String f = line.get(j);
                if (f == null) f = "";
                int gap1 = (int) Math.ceil(perpiece / 2f - f.length() / 2f);
                int gap2 = (int) Math.floor(perpiece / 2f - f.length() / 2f);

                // a number
                for (int k = 0; k < gap1; k++) {
                    System.out.print(" ");
                }
                System.out.print(f);
                for (int k = 0; k < gap2; k++) {
                    System.out.print(" ");
                }
            }
            System.out.println();

            perpiece /= 2;
        }
    }
}

要在树中使用它，让Node类实现PrintableNode。

示例输出:

                                         2952:0                                             
                    ┌───────────────────────┴───────────────────────┐                       
                 1249:-1                                         5866:0                     
        ┌───────────┴───────────┐                       ┌───────────┴───────────┐           
     491:-1                  1572:0                  4786:1                  6190:0         
  ┌─────┘                                               └─────┐           ┌─────┴─────┐     
339:0                                                      5717:0      6061:0      6271:0   

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);

    }
}

这是水平视图最简单的解决方案。我举了很多例子。很适合我的目的。更新自@ ntin -k的回答。

public void print(String prefix, BTNode n, boolean isLeft) {
    if (n != null) {
        print(prefix + "     ", n.right, false);
        System.out.println (prefix + ("|-- ") + n.data);
        print(prefix + "     ", n.left, true);
    }
}

电话:

bst.print("", bst.root, false);

解决方案:

                         |-- 80
                    |-- 70
               |-- 60
          |-- 50
     |-- 40
|-- 30
     |-- 20
          |-- 10

按行打印[大]树。

输出的例子:

z
├── c
│   ├── a
│   └── b
├── d
├── e
│   └── asdf
└── f

代码:

public class TreeNode {

    final String name;
    final List<TreeNode> children;

    public TreeNode(String name, List<TreeNode> children) {
        this.name = name;
        this.children = children;
    }

    public String toString() {
        StringBuilder buffer = new StringBuilder(50);
        print(buffer, "", "");
        return buffer.toString();
    }

    private void print(StringBuilder buffer, String prefix, String childrenPrefix) {
        buffer.append(prefix);
        buffer.append(name);
        buffer.append('\n');
        for (Iterator<TreeNode> it = children.iterator(); it.hasNext();) {
            TreeNode next = it.next();
            if (it.hasNext()) {
                next.print(buffer, childrenPrefix + "├── ", childrenPrefix + "│   ");
            } else {
                next.print(buffer, childrenPrefix + "└── ", childrenPrefix + "    ");
            }
        }
    }
}

附注:这个答案并不完全关注“二叉”树——相反，它打印了各种类型的树。解决方案的灵感来自linux中的“树”命令。