需求
- 如何判断一堆不重复的字符串是否以某个前缀开头?
- 用Set\Map存储字符串
- 遍历所有字符串进行判断
- 时间复杂度: O(n)
- 有没有更优的数据结构实现前缀搜索
Trie
简介
- Trie 也叫做字典树、前缀树(Prefix Tree)、单词查找树
- Trie搜索字符串的效率主要跟字符串的长度有关
- 假设使用Trie存储cat、dog、doggy、does、cast、add六个单词
接口设计
或者
实现类
public class Trie<V> {
private int size;
private Node<V> root;
public int size() {
return size;
}
public boolean isEmpty() {
return size == 0;
}
/**
* 清空数据,根节点置空即可
*/
public void clear() {
size = 0;
root = null;
}
public V get(String key) {
Node<V> node = node(key);
return node != null && node.isEnd ? node.value : null;
}
public boolean contains(String key) {
Node<V> node = node(key);
return node != null && node.isEnd;
}
public V add(String key, V value) {
keyCheck(key);
// 创建根节点
if (root == null) {
root = new Node<>(null);
}
Node<V> node = root;
int len = key.length();
for (int i = 0; i < len; i++) {
char c = key.charAt(i);
boolean emptyChildren = node.children == null;
Node<V> childNode = emptyChildren ? null : node.children.get(c);
if (childNode == null) {
childNode = new Node<>(node);
childNode.character = c;
node.children = emptyChildren ? new HashMap<>() : node.children;
node.children.put(c, childNode);
}
node = childNode;
}
// 已经存在这个单词
if (node.isEnd) {
V oldValue = node.value;
node.value = value;
return oldValue;
}
// 新增一个单词
node.isEnd = true;
node.value = value;
size++;
return null;
}
public V remove(String key) {
// 找到最后一个节点
Node<V> node = node(key);
// 如果不是单词结尾,不用作任何处理
if (node == null || !node.isEnd) {
return null;
}
size--;
V oldValue = node.value;
// 如果还有子节点
if (node.children != null && !node.children.isEmpty()) {
node.isEnd = false;
node.value = null;
return oldValue;
}
// 如果没有子节点
Node<V> parent = null;
while ((parent = node.parent) != null) {
parent.children.remove(node.character);
//parent.isEnd: parent是否为一个单词的尾结点
//parent.children.isEmpty(): parent是否还有子节点
if (parent.isEnd || !parent.children.isEmpty()) {
break;
}
node = parent;
}
return oldValue;
}
public boolean startsWith(String prefix) {
return node(prefix) != null;
}
private Node<V> node(String key) {
keyCheck(key);
Node<V> node = root;
int len = key.length();
for (int i = 0; i < len; i++) {
if (node == null || node.children == null || node.children.isEmpty()) {
return null;
}
char c = key.charAt(i);
node = node.children.get(c);
}
return node;
}
//判断key是否为空
private void keyCheck(String key) {
if (key == null || key.length() == 0) {
throw new IllegalArgumentException("key must not be empty");
}
}
private static class Node<V> {
Node<V> parent;
HashMap<Character, Node<V>> children;
Character character;
V value;
// 是否为单词的结尾(是否为一个完整的单词)
boolean isEnd;
public Node(Node<V> parent) {
this.parent = parent;
}
}
}
测试类
Main.java
public class Main {
static void test1() {
Trie<Integer> trie = new Trie<>();
trie.add("cat", 1);
trie.add("dog", 2);
trie.add("catalog", 3);
trie.add("cast", 4);
trie.add("hmx", 5);
Asserts.test(trie.size() == 5);
Asserts.test(trie.startsWith("do"));
Asserts.test(trie.startsWith("c"));
Asserts.test(trie.startsWith("ca"));
Asserts.test(trie.startsWith("cat"));
Asserts.test(trie.startsWith("cata"));
Asserts.test(!trie.startsWith("hehe"));
Asserts.test(trie.get("hmx") == 5);
Asserts.test(trie.remove("cat") == 1);
Asserts.test(trie.remove("catalog") == 3);
Asserts.test(trie.remove("cast") == 4);
Asserts.test(trie.size() == 2);
Asserts.test(trie.startsWith("do"));
Asserts.test(!trie.startsWith("c"));
}
public static void main(String[] args) {
test1();
}
}
Asserts.java
public class Asserts {
public static void test(boolean value) {
try {
if (!value) {
throw new Exception("测试未通过");
}
} catch (Exception e) {
e.printStackTrace();
}
}
}
总结
- 优点:
- 搜索前缀的效率主要跟前缀的长度有关
- 缺点:
- 需要耗费大量的内存,因此还有待改进
- 与Trie相关的数据结构和算法
- Double-array Trie、Suffix Tree、Patricia Tree、Crit-bit Tree、AC自动机