參考資料:https://blog.csdn.net/gongsai20141004277/article/details/52949995
代碼:
tompson。cpp
#pragma once
#include "Thompson.h"
#include<string.h>
#include<iostream>
using namespace std;
int STATE_NUM = 0;
/**表達式轉NFA處理函數,返回最終的NFA結合
*/
cell express_2_NFA(string expression)
{
int length = expression.size();
char element;
cell Cell, Left, Right;
stack<cell> STACK;
for (int i = 0; i<length; i++)
{
element = expression.at(i);
switch (element)
{
case '|':
Right = STACK.top();
STACK.pop();
Left = STACK.top();
STACK.pop();
Cell = do_Unite(Left, Right);
STACK.push(Cell);
break;
case '*':
Left = STACK.top();
STACK.pop();
Cell = do_Star(Left);
STACK.push(Cell);
break;
case '+':
Right = STACK.top();
STACK.pop();
Left = STACK.top();
STACK.pop();
Cell = do_Join(Left, Right);
STACK.push(Cell);
break;
default:
Cell = do_Cell(element);
STACK.push(Cell);
}
}
cout << "處理完畢!" << endl;
Cell = STACK.top();
STACK.pop();
return Cell;
}
//處理 a|b
cell do_Unite(cell Left, cell Right)
{
cell NewCell;
NewCell.EdgeCount = 0;
edge Edge1, Edge2, Edge3, Edge4;
//獲得新的新狀態節點
state StartState = new_StateNode();
state EndState = new_StateNode();
//構建邊
Edge1.StartState = StartState;
Edge1.EndState = Left.EdgeSet[0].StartState;
Edge1.TransSymbol = '#'; //代表空串
Edge2.StartState = StartState;
Edge2.EndState = Right.EdgeSet[0].StartState;
Edge2.TransSymbol = '#';
Edge3.StartState = Left.EdgeSet[Left.EdgeCount - 1].EndState;
Edge3.EndState = EndState;
Edge3.TransSymbol = '#';
Edge4.StartState = Right.EdgeSet[Right.EdgeCount - 1].EndState;
Edge4.EndState = EndState;
Edge4.TransSymbol = '#';
//構建單元
//先將Left和Right的EdgeSet復制到NewCell
cell_EdgeSet_Copy(NewCell, Left);
cell_EdgeSet_Copy(NewCell, Right);
//將新構建的四條邊加入EdgeSet
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge1;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge2;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge3;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge4;
//構建NewCell的啟示狀態和結束狀態
NewCell.StartState = StartState;
NewCell.EndState = EndState;
return NewCell;
}
//處理 ab
cell do_Join(cell Left, cell Right)
{
//將Left的結束狀態和Right的開始狀態合並,將Right的邊復制給Left,將Left返回
//將Right中所有以StartState開頭的邊全部修改
for (int i = 0; i<Right.EdgeCount; i++)
{
if (Right.EdgeSet[i].StartState.StateName.compare(Right.StartState.StateName) == 0)
{
Right.EdgeSet[i].StartState = Left.EndState;
STATE_NUM--;
}
else if (Right.EdgeSet[i].EndState.StateName.compare(Right.StartState.StateName) == 0)
{
Right.EdgeSet[i].EndState = Left.EndState;
STATE_NUM--;
}
}
Right.StartState = Left.EndState;
cell_EdgeSet_Copy(Left, Right);
//將Left的結束狀態更新為Right的結束狀態
Left.EndState = Right.EndState;
return Left;
}
//處理 a*
cell do_Star(cell Cell)
{
cell NewCell;
NewCell.EdgeCount = 0;
edge Edge1, Edge2, Edge3, Edge4;
//獲得新的新狀態節點
state StartState = new_StateNode();
state EndState = new_StateNode();
//構建邊
Edge1.StartState = StartState;
Edge1.EndState = EndState;
Edge1.TransSymbol = '#'; //閉包取空串
Edge2.StartState = Cell.EndState;
Edge2.EndState = Cell.StartState;
Edge2.TransSymbol = '#'; //取字符,自連接
Edge3.StartState = StartState;
Edge3.EndState = Cell.StartState;
Edge3.TransSymbol = '#';
Edge4.StartState = Cell.EndState;
Edge4.EndState = EndState;
Edge4.TransSymbol = '#';
//構建單元
//先將Cell的EdgeSet復制到NewCell
cell_EdgeSet_Copy(NewCell, Cell);
//將新構建的四條邊加入EdgeSet
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge1;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge2;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge3;
NewCell.EdgeSet[NewCell.EdgeCount++] = Edge4;
//構建NewCell的啟示狀態和結束狀態
NewCell.StartState = StartState;
NewCell.EndState = EndState;
return NewCell;
}
//處理 a
cell do_Cell(char element)
{
cell NewCell;
NewCell.EdgeCount = 0;
edge NewEdge;
//獲得新的新狀態節點
state StartState = new_StateNode();
state EndState = new_StateNode();
//構建邊
NewEdge.StartState = StartState;
NewEdge.EndState = EndState;
NewEdge.TransSymbol = element;
//構建單元
NewCell.EdgeSet[NewCell.EdgeCount++] = NewEdge;
NewCell.StartState = NewCell.EdgeSet[0].StartState;
NewCell.EndState = NewCell.EdgeSet[0].EndState;//EdgeCount此時為1
return NewCell;
}
void cell_EdgeSet_Copy(cell& Destination, cell Source)
{
int D_count = Destination.EdgeCount;
int S_count = Source.EdgeCount;
for (int i = 0; i<S_count; i++)
{
Destination.EdgeSet[D_count + i] = Source.EdgeSet[i];
}
Destination.EdgeCount = D_count + S_count;
}
/*
獲得新的狀態節點,統一產生,便於管理,不能產生重復的狀態
並添加到state_set[]數組中
*/
state new_StateNode()
{
state newState;
newState.StateName = STATE_NUM + 65;//轉換成大寫字母
STATE_NUM++;
return newState;
}
//接收輸入正規表達式,RegularExpression作為回傳函數
void input(string &RegularExpression)
{
cout << "請輸入正則表達式: (操作符:() * |;字符集:a~z A~Z)" << endl;
// do
// {
cin >> RegularExpression;
// } while (!check_legal(RegularExpression));
}
/**檢測輸入的正則表達式是否合法
*/
/*int check_legal(string check_string)
{
if (check_character(check_string) && check_parenthesis(check_string))
{
return true;
}
return false;
}
/**
檢查輸入的字符是否合適 () * | a~z A~Z
合法返回true,非法返回false
*/
/*/*int check_character(string check_string)
{
int length = check_string.size();
for (int i = 0; i<length; i++)
{
char check = check_string.at(i);
if (is_letter(check))//小寫和大寫之間有5個字符,故不能連續判斷
{
//cout<<"字母 合法";
}
else if (check == '(' || check == ')' || check == '*' || check == '|')
{
//cout<<"操作符 合法";
}
else
{
cout << "含有不合法的字符!" << endl;
cout << "請重新輸入:" << endl;
return false;
}
}
return true;
}
/**先檢查括號是否匹配
*合法返回true,非法返回false
*/
/*int check_parenthesis(string check_string)
{
int length = check_string.size();
char * check = new char[length+1];
strcpy_s(check, length+1, check_string.c_str());
stack<int> STACK;
for (int i = 0; i<length; i++)
{
if (check[i] == '(')
STACK.push(i);
else if (check[i] == ')')
{
if (STACK.empty())
{
cerr << "有多余的右括號" << endl;//暫時不記錄匹配位置location
cout << "請重新輸入:" << endl;
return false;
}
else
STACK.pop();
}
}
if (!STACK.empty())
{
//暫時不記錄匹配位置location
cerr << "有多余的左括號" << endl;
cout << "請重新輸入:" << endl;
return false;
}
return true;
}*/
/**檢測是否是字母
是返回true,否則false
*/
int is_letter(char check)
{
if (check >= 'a'&&check <= 'z' || check >= 'A'&&check <= 'Z')
return true;
return false;
}
/**添加交操作符“+”,便於中綴轉后綴表達式
例如 abb->a+b+b
*/
string add_join_symbol(string add_string)
{
/* 測試終止符\0
string check_string = "abcdefg\0aaa";
cout<<check_string<<endl;
int length = check_string.size();
char * check = new char[2*length];
strcpy(check,check_string.c_str());
cout<<check<<endl;
char *s = "ssss\0 aa";
cout<<s<<endl;
string a(s);
cout<<a<<endl;
*/
int length = add_string.size();
int return_string_length = 0;
char *return_string = new char[2 * length+2];//最多是兩倍
char first, second;
for (int i = 0; i<length - 1; i++)
{
first = add_string.at(i);
second = add_string.at(i + 1);
return_string[return_string_length++] = first;
//要加的可能性如ab 、 *b 、 a( 、 )b 等情況
//若第二個是字母、第一個不是'('、'|'都要添加
if (first != '('&&first != '|'&&is_letter(second))
{
return_string[return_string_length++] = '+';
}
//若第二個是'(',第一個不是'|'、'(',也要加
else if (second == '('&&first != '|'&&first != '(')
{
return_string[return_string_length++] = '+';
}
}
//將最后一個字符寫入
return_string[return_string_length++] = second;
return_string[return_string_length] = '\0';
string STRING(return_string);
cout << "加'+'后的表達式:" << STRING << endl;
return STRING;
}
/*
構造優先級表規則:(1)先括號內,再括號外;(2)優先級由高到低:閉包、|、+;(3)同級別,先左后右。
優先級表:
# ( * | + )
isp 0 1 7 5 3 8
icp 0 8 6 4 2 1
*/
/*********************運算符優先級關系表********************/
/*
c1\c2 ( * | + ) #
( < < < < = >
* < > > > > >
| < < > > > >
+ < < < > > >
# < < < < < =
*/
/***********************************************************/
//in stack priority 棧內優先級,棧頂的字符的優先級
int isp(char c)
{
switch (c)
{
case '#': return 0;
case '(': return 1;
case '*': return 7;
case '|': return 5;
case '+': return 3;
case ')': return 8;
}
//若走到這一步,說明出錯了
cerr << "ERROR!" << endl;
return false;
}
//in coming priority 棧外優先級,當前掃描到的字符的優先級
int icp(char c)
{
switch (c)
{
case '#': return 0;
case '(': return 8;
case '*': return 6;
case '|': return 4;
case '+': return 2;
case ')': return 1;
}
//若走到這一步,說明出錯了
cerr << "ERROR!" << endl;
return false;
}
/**中綴表達式轉后綴表達式
*/
string postfix(string e)
{
//設定e的最后一個符號式“#”,而其“#”一開始先放在棧s的棧底
e = e + "#";
stack<char> s;
char ch = '#', ch1, op;
s.push(ch);
//讀一個字符
string out_string = "";
int read_location = 0;
ch = e.at(read_location++);
while (!s.empty())
{
if (is_letter(ch))
{
out_string = out_string + ch;
//cout<<ch;
ch = e.at(read_location++);
}
else
{
//cout<<"輸出操作符:"<<ch<<endl;
ch1 = s.top();
if (isp(ch1)<icp(ch))
{
s.push(ch);
//cout<<"壓棧"<<ch<<" 讀取下一個"<<endl;
ch = e.at(read_location++);
}
else if (isp(ch1)>icp(ch))
{
op = s.top();
s.pop();
//cout<<"退棧"<<op<<" 添加到輸出字符串"<<endl;
out_string = out_string + op;
//cout<<op;
}
else //考慮優先級相等的情況
{
op = s.top();
s.pop();
//cout<<"退棧"<<op<<" 但不添加到輸入字符串"<<endl;
if (op == '(')
ch = e.at(read_location++);
}
}
}
cout << "后綴表達式:" << out_string << endl;
return out_string;
}
/*
顯示NFA
*/
void Display(cell Cell)
{
cout << "NFA 的邊數:" << Cell.EdgeCount << endl;
cout << "NFA 的起始狀態:" << Cell.StartState.StateName << endl;
cout << "NFA 的結束狀態:" << Cell.EndState.StateName << endl;
for (int i = 0; i<Cell.EdgeCount; i++)
{
cout << "第" << i + 1 << "條邊的起始狀態:" << Cell.EdgeSet[i].StartState.StateName
<< " 結束狀態:" << Cell.EdgeSet[i].EndState.StateName
<< " 轉換符:" << Cell.EdgeSet[i].TransSymbol << endl;
}
cout << "結束" << endl;
}
//主函數
int main()
{
string Regular_Expression = "(a|b)*abb";
cell NFA_Cell;
Regular_Expression = "(a|b)*abb";
//接受輸入
input(Regular_Expression);//調試需要先屏蔽
//添加“+”,便於轉后綴表達式
Regular_Expression = add_join_symbol(Regular_Expression);
//中綴轉后綴
Regular_Expression = postfix(Regular_Expression);
//表達式轉NFA
NFA_Cell = express_2_NFA(Regular_Expression);
//顯示
Display(NFA_Cell);
}
tompson.h
#ifndef THOMPSON_H
#define THOMPSON_H
#include <iostream>
#include <stdio.h>
#include <cctype>
#include <stack>
#include <string>
using namespace std;
#define MAX 100
//節點,定義成結構體,便於以后擴展
struct state
{
string StateName;
};
//邊,空轉換符用'#'表示
struct edge
{
state StartState;
state EndState;
char TransSymbol;
};
//NFA單元
struct cell
{
edge EdgeSet[MAX];
int EdgeCount;
state StartState;
state EndState;
};
/***************NFA的矩陣結構****************/
/*struct node
{
edge* In_edges;
edge* Out_edges;
state
};*/
/********************************************/
//函數聲明
void input(string&);
int check_legal(string);
int check_character(string);
int check_parenthesis(string);
int is_letter(char);
//添加“+”,便於轉后綴表達式
string add_join_symbol(string);
//中綴轉后綴
string postfix(string);
//優先級 in stack priority
int isp(char);
//優先級 in coming priority
int scp(char);
//表達式轉NFA
cell express_2_NFA(string);
//處理 a|b
cell do_Unite(cell, cell);
//處理 ab
cell do_Join(cell, cell);
//處理 a*
cell do_Star(cell);
//處理 a
cell do_Cell(char);
//將一個單元的邊的集合復制到另一個單元
void cell_EdgeSet_Copy(cell&, cell);
//產生一個新的狀態節點,便於管理
state new_StateNode();
//顯示
void Display(cell);
#endif //THOMPSON.H