7-前缀/字典树

重点知识

字典树也称作前缀树或者Trie树，是一种哈希树的变种。

典型应用：用于统计，排序和保存大量的字符串（但不仅限于字符串），所以经常被搜索引擎系统用于文本词频统计。

优点：利用字符串的公共前缀来减少查询时间，最大限度地减少无谓的字符串比较，查询效率比哈希树高。

字典树的特点
根节点不包含字符，除根节点外每一个节点都只包含一个字符；
从根节点到某一节点，路径上经过的字符连接起来，为该节点对应的字符串；
每个节点的所有子节点包含的字符都不相同。

字典树数据结构的实现方法
字典树要实现以下两种基本功能：

• 插入单词
• 搜索单词
• 搜索前缀

练习

leetCode 820. 单词的压缩编码

class MyPrefixTree:
    def __init__(self):
        self.trees = {}

    def search(self, word: str):
        my_tree = self.trees
        for s in word:
            if s not in my_tree:
                return False
            my_tree = my_tree[s]
        if my_tree.get("is_word"):
            return True
        return False

    def insert(self, word: str):
        my_tree = self.trees
        for s in word:
            if s in my_tree:
                my_tree = my_tree[s]
                my_tree["is_word"] = False
            else:
                my_tree[s] = {"is_word": False}
                my_tree = my_tree[s]
        if len(my_tree) == 1:
            my_tree["is_word"] = True

    def search_prefix(self, prefix: str):
        my_tree = self.trees
        step = 0
        for s in prefix:
            if s not in my_tree:
                return False, step
            my_tree = my_tree[s]
            step += 1
        if not my_tree.get("is_word"):
            step = 0
        return True, step


class Solution:
    def minimumLengthEncoding(self, words: list[str]) -> int:
        my_prefix_tree = MyPrefixTree()
        for word in set(words):
            reversed_w = word[-1:-len(word)-1:-1]
            my_prefix_tree.insert(reversed_w)
        ret = 0
        for word in set(words):
            reversed_w = word[-1:-len(word) - 1:-1]
            _, step = my_prefix_tree.search_prefix(reversed_w)
            if step:
                ret += step
                ret += 1
        return ret