首页 > 代码库 > Trie树

Trie树

Trie树,即字典树或单词查找树,主要用于大量字符串的检索、去重、排序等操作。

主要原理就是利用字符串的公共前缀建立一棵多叉树,牺牲空间换取时间。

 1 //Trie树
 2 #include <iostream>
 3 #include <string>
 4 using std::cin;
 5 using std::cout;
 6 using std::endl;
 7 using std::string;
 8 
 9 const int SIZE = 26;
10 const char BASE = a;
11 
12 class TrieNode {
13 public:
14     //is_terminal means that this char is the last one of this string
15     //we can get this string by going from the root to this terminal
16     bool is_terminal;
17     TrieNode **childs;
18     TrieNode() {
19         is_terminal = false;
20         childs = new TrieNode*[SIZE];
21         for (int i = 0; i < SIZE; ++i) {
22             childs[i] = nullptr;
23         }
24     }
25     ~TrieNode() {
26         for (int i = 0; i < SIZE; ++i) {
27             delete childs[i];
28         }
29         delete[] childs;
30     }
31 };
32 
33 class StringPatterns {
34 private:
35     TrieNode *root;
36 public:
37     StringPatterns() {
38         root = new TrieNode();
39     }
40     ~StringPatterns() {
41         delete root;
42     }
43     void add_pattern(const string &pattern) {
44         TrieNode *now=root;
45         for(int i=0;i<pattern.length();i++){
46             //this char is new in current level, means that this char is from a new different string
47             if(now->childs[pattern[i]-BASE]==nullptr){
48                 //add this char
49                 now->childs[pattern[i]-BASE]=new TrieNode();
50             }
51             //go to next char
52             now=now->childs[pattern[i]-BASE];
53         }
54         //now points to the last char of this added string
55         now->is_terminal=true;
56     }
57     bool match(const string &buffer) const {
58         TrieNode *now=root;
59         for(int i=0;i<buffer.length();i++){
60             if(now->childs[buffer[i]-BASE]==nullptr){
61                 return false;
62             }
63             now=now->childs[buffer[i]-BASE];
64         }
65         return now->is_terminal;
66     }
67 };
68 
69 int main() {
70     StringPatterns trie;
71     int n;
72     cin >> n;
73     for (int i = 0; i < n; ++i) {
74         string pattern;
75         cin >> pattern;
76         trie.add_pattern(pattern);
77     }
78     string str_buffer;
79     cin >> str_buffer;
80     if (trie.match(str_buffer)) {
81         cout << "match success" << endl;
82     } else {
83         cout << "match failed" << endl;
84     }
85     return 0;
86 }

 

 用Trie树求一个string的所有不相同的子串:

 1 //find all different substrings of a string
 2 #include <iostream>
 3 #include <string>
 4 using std::cin;
 5 using std::cout;
 6 using std::endl;
 7 using std::string;
 8 
 9 const int SIZE = 26;
10 const char BASE = a;
11 
12 class TrieNode {
13 public:
14     bool is_terminal;
15     TrieNode **childs;
16     TrieNode() {
17         is_terminal=false;
18         childs=new TrieNode*[SIZE];
19         for(int i=0;i<SIZE;i++){
20             childs[i]=nullptr;
21         }
22     }
23     ~TrieNode() {
24         for(int i=0;i<SIZE;i++){
25             delete childs[i];
26         }
27         delete[] childs;
28     }
29 };
30 
31 class StringPatterns {
32 private:
33     TrieNode *root;
34     //number of substrings
35     int subs;
36 public:
37     StringPatterns() {
38         subs=0;
39         root=new TrieNode();
40     }
41     ~StringPatterns() {
42         delete root;
43     }
44     void add_pattern(const string &pattern) {
45         TrieNode *now=root;
46         for(int i=0;i<pattern.length();i++){
47             now=root;
48             //add in all substrings starting with pattern[j]
49             for(int j=i;j<pattern.length();j++){
50                 if(now->childs[pattern[j]-BASE]==nullptr){
51                     //means this is a new different substring
52                     now->childs[pattern[j]-BASE]=new TrieNode();
53                     subs++;
54                 }
55                 now=now->childs[pattern[j]-BASE];
56             }
57         }
58         cout<<subs<<endl;
59     }
60     
61     
62 };
63 
64 int main() {
65     StringPatterns strp;
66     string it;
67     cin>>it;
68     strp.add_pattern(it);
69     return 0;
70 }

 

Trie树