【Lintcode】997. Print Organization Chart

题目地址：

https://www.lintcode.com/problem/print-organization-chart/description

给定一个列表，每个元素是个字符串列表，每个列表代表一个员工，按照该员工姓名，上级姓名，职位和年份排列。上级为”NULL“代表该员工是最高层的员工，没有上级。题目保证公司员工按照上下级关系的结构是个树结构。要求返回一个列表，列表里含的是字符串，格式是其姓名、空格、左括号、职位、右括号、空格、年份，字符串之间的顺序是先序遍历整棵树。并且该树的同一层的员工按照姓名的字典序排序。

思路是DFS。先建立树结构，再进行DFS。这里建立树结构的时候，最好将每个员工包装成一个类，将其姓名、下级、职位和年份包装在类里。代码如下：

import java.util.*;

public class Solution {
    
    class Person implements Comparable<Person> {
        String name, pos, year;
        Set<Person> sub;
        
        public Person(String name) {
            this.name = name;
            // sub用TreeSet，保证下级按照字典序排序
            sub = new TreeSet<>();
        }
    
        @Override
        public int compareTo(Person p) {
            return name.compareTo(p.name);
        }
    }
    
    /**
     * @param relationship: the relationship
     * @return: the organization chart
     */
    public List<String> getOrganization(List<List<String>> relationship) {
        // Write your code here
        List<String> res = new ArrayList<>();
        Person root = buildTree(relationship);
        dfs(root, 0, res);
        return res;
    }
    
    private void dfs(Person cur, int level, List<String> res) {
        StringBuilder sb = new StringBuilder();
    
        sb.append("-".repeat(level)).append(cur.name).append(" (").append(cur.pos).append(") ").append(cur.year);
        res.add(sb.toString());
    
        for (Person sub : cur.sub) {
            dfs(sub, level + 1, res);
        }
    }
    
    private Person buildTree(List<List<String>> relationship) {
    	// 建立员工名和员工Person对象之间的映射关系
        Map<String, Person> map = new HashMap<>();
        // 存树根对应的Person
        Person root = null;
        for (List<String> relation : relationship) {
            String nameSub = relation.get(0), nameSup = relation.get(1);
            
            if ("NULL".equals(nameSup)) {
                root = map.getOrDefault(nameSub, new Person(relation.get(0)));
                root.pos = relation.get(2);
                root.year = relation.get(3);
                map.put(root.name, root);
                continue;
            }
            
            Person personSub = map.getOrDefault(nameSub, new Person(relation.get(0)));
            map.putIfAbsent(nameSub, personSub);
            personSub.pos = relation.get(2);
            personSub.year = relation.get(3);
            
            Person personSup = map.getOrDefault(nameSup, new Person(relation.get(1)));
            map.putIfAbsent(nameSup, personSup);
            
            personSup.sub.add(personSub);
        }
        
        return root;
    }
}

时空复杂度$O(nl)$，$l$代表每个Person对应的字符串长度。