编译原理实验二语法分析程序设计与实现

一、实验目的

任选一种有代表性的语法分析方法，如算符优先法、递归下降法、LL(1)、SLR(1)、LR(1)等，通过设计、编制、调试实现一个典型的语法分析程序，对实验一所得扫描器提供的单词序列进行语法检查和结构分析，实现并进一步掌握常用的语法分析方法。

二、基本实验内容与要求

选择对各种常见高级程序设计语言都较为通用的语法结构——算术表达式的一个简化子集——作为分析对象，根据如下描述其语法结构的 BNF 定义G2[<算术表达式>]，任选一种学过的语法分析方法，针对运算对象为无符号常数和变量的四则运算，设计并实现一个语法分析程序。

G2[<算术表达式>]： <算术表达式> → <项> | <算术表达式>+<项> | <算术表达式>-<项> <项> → <因式> |<项><因式> | <项>/<因式> <因式> → <运算对象> | (<算术表达式>)
若将语法范畴<算术表达式>、<项>、<因式>和<运算对象>分别用 E、T、F 和 i 代表，则 G2 可写成：
G2[E]：E → T | E+T | E-T T → F | TF | T/F F → i | (E)

输入：由实验一输出的单词串，例如：UCON，PL，UCON，MU，ID ······
输出：若输入源程序中的符号串是给定文法的句子，则输出“RIGHT”，并且给出每一步分析过程；若不是句子，即输入串有错误，则输出“ERROR”，并且显示分析至此所得的中间结果，如分析栈、符号栈中的信息等，以及必要的出错说明信息。
要求:
1、确定语法分析程序的流程图，同时考虑相应的数据结构，编写一个语法分析源程序。
2、将词法、语法分析合在一起构成一个完整的程序，并调试成功。3、供测试的例子应包括符合语法规则的语句，及分析程序能判别的若干错例。对于所输入的字符串，不论对错，都应有明确的信息输出。

采用SLR1分析方法，首先将给定文法的SLR1分析表手工画出，存入到程序中，然后进行语法分析程序的实现。

四、源程序

#include <iostream>
#include <string>
#include <iomanip>
#include <cstring>
#include <fstream>
#include <stdio.h>
#include <cassert>
#include <vector>
//#define _CRT_SECURE_NO_WARNINGS
using namespace std;
/*
begin   BEGIN   1
end     END     2
if      IF      3
then    THEN    4
else    ELSE    5
while   WHILE   6
do      DO      7
标识符 ID      8
浮点常数UCON    9
<        LT      10
<=      LE      11
==        EQ      12
<>        NE      13
>        GT      14
>=      GE      15
=      IS      16
+      PL      17
-       MI      18
*       MU      19
/       DI      20
(       LP      21  新增加的左括号
)       RP      22  新增加的右括号
*/const char* table1[] = { " ","begin","end","if","then","else","while","do" };
const char* table2[] = { " ","BEGIN","END","IF","THEN","ELSE","WHILE","DO","ID","UCON","LT","LE","EQ","NE","GT","GE","IS","PL","MI","MU","DI","LP","RP" };
char TOKEN[20];//用来依次存放一个单词词文中的各个字符。
char str[50];//存放词法分析后的表达式，将输入的表达式转变成只含有i和运算符的表达式
int len = 0;//词法分析后表达式的长度//action表   char vt[8] = { '(',')','+','-','*','/','i','#' };
string action[16][8] = { {"s4","null","null" ,"null" ,"null","null","s5" , "null" },{"null","null","s6","s7","null","null","null","acc"},{"null","r3","r3","r3","s8","s9","null","r3"} ,{"null","r6","r6","r6","r6","r6","null","r6"},{"s4","null","null","null","null","null","s5","null"},{"null","r8","r8","r8","r8","r8","null","r8"},{"s4","null","null","null","null","null","s5","null"},{"s4","null","null","null","null","null","s5","null"},{"s4","null","null","null","null","null","s5","null"},{"s4","null","null","null","null","null","s5","null"},{"null","s15","s6","s7","null","null","null","null"} ,{"null","r1","r1","r1","s8","s9","null","r1"},{"null","r2","r2","r2","s8","s9","null","r2"} ,{"null","r4","r4","r4","r4","r4","null","r4"} ,{"null","r5","r5","r5","r5","r5","null","r5"} ,{"null","r7","r7","r7","r7","r7","null","r7"} };
//go表 char vn[3] = { 'E','T','F' };
int go[16][3] = { {1 , 2 , 3 },{ -1 , -1 , -1 },{ -1 , -1 , -1 } ,{ -1 , -1 , -1 },{ 10 , 2 , 3 },{ -1 , -1 , -1 },{ -1 , 11 , 3 },{ -1 , 12 , 3 },{ -1 , -1 , 13 },{ -1 , -1 , 14 },{ -1 , -1 , -1 } ,{ -1 , -1 , -1 },{ -1 , -1 , -1 } ,{ -1 , -1 , -1 } ,{ -1 , -1 , -1 } ,{ -1 , -1 , -1 } };
string grammar[9] = { "S->E","E->E+T","E->E-T","E->T","T->T*F","T->T/F","T->F","F->(E)","F->i" };//存放文法表达式
int grammarlength[9] = { 4,6,6,4,6,6,4,6,4 };//每个文法表达式的长度
char grammarleft[9] = { 'S','E','E','E','T','T','T','F','F' };//文法表达式的左部
string grammarright[9] = { "E","E+T","E-T","T","T*F","T/F","F","(E)","i" };//文法表达式的右部
int grammarrightlength[9] = { 1,3,3,1,3,3,1,3,1 };//文法表达式右部长度int lookup(char a[])/*每调用一次，就以 TOKEN 中的字符串查保留字表，若查到，就将相应关键字的类别码赋给整型变量 c；否则将 c 置为零。*/
{int i;for (i = 1; i <= 7; i++){if (strcmp(a, table1[i]) == 0)return i;}if (i == 8){return 0;}}
void out(int a, const char b[])
/*一般仅在进入终态时调用此函数，调用的形式为 OUT(c，VAL)。
其中，实参 c 为相应单词的类别码助记符；实参 VAL 为 TOKEN（即词文）或为空串。
函数 OUT 的功能是，在送出一个单词的内部表示之后，返回到调用该词法分析程序的那个程序。*/
{ofstream fout("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\result.txt", fstream::app);char temp = ' ';if (a == 8 || a == 9){fout << "（" << table2[a] << " ，" << b << "）" << endl;}else{fout << "（" << table2[a] << " ，" << temp << "）" << endl;}fout.close();
}void report_error(void)//返回错误
{ofstream fout("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\result.txt", fstream::app);fout << "输入不合规范！";fout.close();exit(0);
}
void scanner_example(FILE* fp)
{char ch;int i, c, flag = 0;;//ch = fgetc(fp);while ((ch = fgetc(fp)) != EOF){if (ch == ' ')ch = fgetc(fp);if (ch == '\n')ch = fgetc(fp);if (ch == '#')break;if (isalpha(ch)) //判断字符是否为字母，识别标识符和关键字{TOKEN[0] = ch; ch = fgetc(fp); i = 1;while (isalnum(ch))//检查所传的字符是否是字母和数字{TOKEN[i] = ch; i++;ch = fgetc(fp);}TOKEN[i] = '\0';//结束符fseek(fp, -1, 1); //从文件当前位置后退一个字符c = lookup(TOKEN);//查找在第二个表中的位置，如果是关键字返回下标，标识符返回0if (c == 0){out(8, TOKEN);//识别标识符str[len] = 'i';len++;}else out(c, TOKEN);//识别关键字}else if (isdigit(ch))//判断字符是否为数字字符{TOKEN[0] = ch; ch = fgetc(fp); i = 1;while (isdigit(ch) || ch == '.'){if (ch == '.'){if (flag == 0){flag = 1;TOKEN[i] = ch; i++;}else report_error();}else{TOKEN[i] = ch; i++;}ch = fgetc(fp);}TOKEN[i] = '\0';fseek(fp, -1, 1);out(9, TOKEN);//识别数字str[len] = 'i';len++;}elseswitch (ch){case '<': ch = fgetc(fp);if (ch == '=')out(11, TOKEN);//<=else if (ch == '>') out(13, TOKEN);//<>else{fseek(fp, -1, 1);out(10, TOKEN);//<}break;case '=': ch = fgetc(fp);if (ch == '=')out(12, TOKEN);//==else{fseek(fp, -1, 1);out(16, TOKEN);//=}break;case '>': ch = fgetc(fp);if (ch == '=')out(15, TOKEN);//>=else{fseek(fp, -1, 1);out(14, TOKEN);//>}break;case '+':{out(17, TOKEN);str[len] = '+';len++;break;}case '-':{out(18, TOKEN);str[len] = '-';len++;break;}case '*':{out(19, TOKEN);str[len] = '*';len++;break;}case '/':{out(20, TOKEN);str[len] = '/';len++;break;}case '('://新增识别左括号{out(21, TOKEN);str[len] = '(';len++;break;}case ')'://新增识别右括号{out(22, TOKEN);str[len] = ')';len++;break;}default: report_error(); break;}}return;
}int main()
{cout << "请输入要分析的表达式（以#结束）：" << endl;char strr[50];int i = 0;//输入表达式的长度ofstream out1("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\test.txt");out1.close();ofstream out("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\test.txt", ios::out);while (true)//将表达式存入文件test{cin >> strr[i];out << strr[i];if (strr[i] == '#')break;i++;}out.close();FILE* fp1;FILE* fp2;fp2 = fopen("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\result.txt", "w");fclose(fp2);fp1 = fopen("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\test.txt", "r");if (!fp1){printf("打开文件失败！");}scanner_example(fp1);//对test文件中的内容进行词法分析，存入result文件string temp;cout << endl;cout << "词法分析结果为：" << endl;ifstream myfile2("C:\\Users\\Administrator\\Desktop\\编译原理\\实验代码\\result.txt", ios::in);if (!myfile2.is_open()){cout << "未成功打开文件" << endl;}while (getline(myfile2, temp))//输出rusult文件中的内容{cout << temp << endl;}myfile2.close();str[len] = '#';cout << endl;cout << "词法分析后表达式为：" << endl;for (int i = 0; i <= len; i++){cout << str[i];}cout << endl;cout << endl;cout << "语法分析结果为：" << endl;int count;//步骤int state[100], stateIndex;//状态串及下标state[0] = 0; stateIndex = 0;//初始化状态串char inSymbol[100], outSymbol[100];//栈中符号串  余留符号串inSymbol[0] = '#';//初始化int inSymbolIndex = 0, outSymbolIndex = 0;string analysis;//分析动作//string nextState;//下一状态for (int i = 0; i <= len; i++){outSymbol[i] = str[i];}cout << "字符串长度为：" << len << endl;cout << "步骤" << "\t" << "状态" << "\t\t";cout << "栈中符号串" << "\t\t" << "余留符号串" << "\t\t";cout << "分析动作" << "\t\n";for (count = 1;;){cout << count << fixed << setprecision(3) << "\t";for (int i = 0; i <= stateIndex; i++)//输出状态串{cout << state[i] << " " << fixed << setprecision(4);}cout << "\t\t";for (int i = 0; i <= inSymbolIndex; i++)//输出栈中符号串{cout << inSymbol[i] << " " << fixed << setprecision(4);}cout << "\t\t";for (int i = outSymbolIndex; i <= len; i++)//输出余留符号串{cout << outSymbol[i] << " " << fixed << setprecision(4);}cout << "\t\t";int row = state[stateIndex], column;//根据状态串和余留符号串查action表if (outSymbol[outSymbolIndex] == '(')column = 0;else if (outSymbol[outSymbolIndex] == ')')column = 1;else if (outSymbol[outSymbolIndex] == '+')column = 2;else if (outSymbol[outSymbolIndex] == '-')column = 3;else if (outSymbol[outSymbolIndex] == '*')column = 4;else if (outSymbol[outSymbolIndex] == '/')column = 5;else if (outSymbol[outSymbolIndex] == 'i')column = 6;else if (outSymbol[outSymbolIndex] == '#')column = 7;if (action[row][column] == "null"){cout << "error!" << endl;break;}else//查表不为空{analysis = action[row][column];cout << analysis << "\t";//输出分析动作if (analysis[0] == 's')//查到以s开头的分析动作{cout << endl;stateIndex++;if (analysis.length() == 3){state[stateIndex] = ((int)(analysis[1]) - 48) * 10 + ((int)(analysis[2]) - 48);}else state[stateIndex] = (int)analysis[1] - 48;inSymbolIndex++;inSymbol[inSymbolIndex] = outSymbol[outSymbolIndex];outSymbolIndex++;count++;}else if (analysis[0] == 'r')//查到以r开头的分析动作{int temp = (int)analysis[1] - 48;cout << grammar[temp] << endl;stateIndex = stateIndex - grammarrightlength[temp];inSymbolIndex = inSymbolIndex + 1 - grammarrightlength[temp];inSymbol[inSymbolIndex] = grammarleft[temp];cout << " " << fixed << setprecision(3) << "\t";for (int i = 0; i <= stateIndex; i++)//输出状态串{cout << state[i] << " " << fixed << setprecision(4);}cout << "\t\t";for (int i = 0; i <= inSymbolIndex; i++)//栈中符号串{cout << inSymbol[i] << " " << fixed << setprecision(4);}cout << "\t\t";for (int i = outSymbolIndex; i <= len; i++)//余留符号串{cout << outSymbol[i] << " " << fixed << setprecision(4);}cout << "\t\t";int row2 = state[stateIndex], column2;//根据状态串和栈中符号串查go表if (inSymbol[inSymbolIndex] == 'E')column2 = 0;else if (inSymbol[inSymbolIndex] == 'T')column2 = 1;else if (inSymbol[inSymbolIndex] == 'F')column2 = 2;if (go[row2][column2] == -1){cout << "error!" << endl;break;}else//不为空，输出{cout << "go[" << row2 << "," << column2 << "]=" << go[row2][column2] << endl;stateIndex++;state[stateIndex] = go[row2][column2];count++;}}else if (analysis == "acc"){cout << endl << "分析成功！";break;}}}fclose(fp1);fclose(fp2);return 0;
}

五、实验结果及分析

1、输入正确的表达式

2、输入错误表达式