C语言词法分析和语法分析

使用C语言编写程序对C语言程序进行语法和词法分析。

注注注注注：Dev需先配置C++11的环境，lex.h与语法分析程序同一目录下

目录

1.程序要求

1.1词法分析

1.2语法分析

2.原理图

2.1词法

2.2种别码表

2.3语法

 3.运行截图

3.1词法分析

3.2语法分析

4.源码

4.1词法分析(lex.h)

4.2语法分析

---

1.程序要求

1.1词法分析

已知待分析的Ｃ语言子集的词法：

１. 关键字： main  if  else  int  while  char 均为小写。

２.专用符号： = + - * / < <= > >=  ==  !=   ; , {  }  (  )

3、其他标记ID和NUM通过以下正则式定义： ID: letter(letter|digit)* NUM: digitdigit* letter→a|b|c|d…|z|A|B|C…|Z digit →0|1|2|3|4|5|6|7|8|9

4、空格由空白、制表符、换行符组成，用来分隔ID、NUM、专用符号与关键字，词法分析阶段常被忽略。

要求： 设计、编制并调试一个词法分析程序

输入：源程序文件

输出：二元组（syn,token或sum)构成的序列（文件），其中：syn为单词种别码，token为存放的单词自身字符串，sum为整型常量。

1.2语法分析

在上机(一)词法分析的基础上,采用递归子程序法或其他适合的语法分析方法,实现其语法分析程序。要求编译后能检查出语法错误。

2.原理图

2.1词法

2.2种别码表

单词符号	种别码	单词符号	种别码	单词符号	种别码
main	1	+	22	;	31
int	2	-	23	>	32
char	3	*	24	<	33
if	4	/	25	>=	34
else	5	(	26	<=	35
while	6	)	27	==	36
ID	10	{	28	!=	37
NUM	20	}	29
=	21	,	30

2.3语法

 3.运行截图

3.1词法分析

3.2语法分析

 

 

 

 

4.源码

4.1词法分析(lex.h)

#ifndef LEX_H
#define LEX_H
#include"lex.h" 

#define E 65535
#define EOI -1
#include<iostream>
#include<cstdio>
#include<cstring>
#include<string>
#include<cctype>
#include<vector>
#include<map>
using namespace std;
int line;
int column,cc;
typedef struct CI{
	int line;
	int column;
	int type;
}Ci;
map<string,int> mp{{"ID",10},{"INT",20},{"<",33},{"<=",35},{"<>",36},{"!=",37},
				{">",32},{">=",34},{"=",21},{"+",22},{"-",23},
				{"*",24},{"/",25},{"(",26},{")",27},{"{",28},
				{"}",29},{",",30},{";",31},{"[",32},{"]",33}, 
				{"int",2},{"char",3},{"float",7},{"main",1},{"switch",12},
				{"double",8},{"case",10},{"for",9},{"if",4},{"auto",9},
				{"else",5},{"do",11},{"while",6},{"void",13},{"static",14},
				{"return",15},{"break",16},{"struct",17},{"const",18},
				{"union",19},{"typedef",20},{"enum",21}};						
int lookup(string TOKEN){
	if(mp.find(TOKEN)==mp.end())return -1;
	else return mp[TOKEN];
}
void out(FILE* fout,string str,int id,string explain,vector<Ci> &v){
	fprintf(fout,"line_%-5d%-9s%-5d%-9s\n",(line+1),str.c_str(),id,explain.c_str());
	v.push_back({line+1,column,id});
}
void print_error(FILE* fout,char ch,vector<Ci> &v){
	fprintf(fout,"line_%-5d%-9c%-5c%-9s\n",(line+1),ch,'E',"异常字符");
	v.push_back({line+1,column,E});
}
vector<Ci> analysis(FILE* fin,FILE* fout){
	char ch;
	int i,c;
	string TOKEN;
	vector<Ci> v;
	column=1;
	fprintf(fout,"%-10s%-9s%-5s%-9s\n","line_num","string","num","explain");
	while((ch=fgetc(fin))!=EOF){
		if(ch==' '){
			column++;
			continue;
		}
		if(ch=='\t'){
			column+=4;
			continue;
		}
		if(ch=='\n'){
			line++;
			column=1; 
			continue;
		}
		if(isalpha(ch)){//标识符或关键字 
			cc=column;
			TOKEN=string(1,ch);
			ch=fgetc(fin); 
			cc++;
			while(isalnum(ch)){
				cc++;
				TOKEN+=ch;
				ch=fgetc(fin);
			}
			fseek(fin,-1,1);//回退
			c=lookup(TOKEN);
			if(c==-1)out(fout,TOKEN,mp["ID"],"标识符",v);//是标识符(1) 
			else out(fout,TOKEN,c,"关键字",v);//是关键字
			column=cc;
		}else if(isdigit(ch)){
			cc=column;
			TOKEN=string(1,ch);
			ch=fgetc(fin);
			cc++;
			while(isdigit(ch)){
				cc++;
				TOKEN+=ch;
				ch=fgetc(fin);
			}
			fseek(fin,-1,1);//回退
			out(fout,TOKEN,mp["INT"],"无符号整数",v);
			column=cc;
		}else{
			switch(ch){
				//运算符 
				case '<':{
					ch=fgetc(fin);
					cc=column+2;
					if(ch=='=')out(fout,"<=",mp["<="],"运算符",v);
					else if(ch=='>')out(fout,"<>",mp["<>"],"运算符",v);
					else{
						cc--;
						fseek(fin,-1,1);//回退
						out(fout,"<",mp["<"],"运算符",v);	
					}
					column=cc;
					break;
				}
				case '>':{
					ch=fgetc(fin);
					cc=column+2;
					if(ch=='=')out(fout,">=",mp[">="],"运算符",v);
					else{
						cc--;
						fseek(fin,-1,1);//回退
						out(fout,">",mp[">"],"运算符",v);	
					}
					column=cc;
					break;
				}
				case '=':out(fout,string(1,ch),mp["="],"运算符",v);column++;break;
				case '+':out(fout,string(1,ch),mp["+"],"运算符",v);column++;break;
				case '-':out(fout,string(1,ch),mp["-"],"运算符",v);column++;break;
				case '*':out(fout,string(1,ch),mp["*"],"运算符",v);column++;break;
				case '/':out(fout,string(1,ch),mp["/"],"运算符",v);column++;break;
				//分界符 
				case '(':out(fout,string(1,ch),mp["("],"分界符",v);column++;break;
				case ')':out(fout,string(1,ch),mp[")"],"分界符",v);column++;break;
				case '[':out(fout,string(1,ch),mp["["],"分界符",v);column++;break;
				case ']':out(fout,string(1,ch),mp["]"],"分界符",v);column++;break;
				case '{':out(fout,string(1,ch),mp["{"],"分界符",v);column++;break;
				case '}':out(fout,string(1,ch),mp["}"],"分界符",v);column++;break;
				case ',':out(fout,string(1,ch),mp[","],"分界符",v);column++;break;
				case ';':out(fout,string(1,ch),mp[";"],"分界符",v);column++;break;
				default:print_error(fout,ch,v);column++;break;
			}
		}
		
	}
	v.push_back({line+1,column,EOI});
	return v;
}
#endif

4.2语法分析

#include<iostream>
#include"lex.h"
using namespace std;
int index;
FILE *fin,*fout,*flog; 
vector<Ci> v;
void advance(){
	index++;
}
bool match(string s){
	if(index>=v.size())return 0;
	if(v[index].type==E){
		printf("line:%-3d,column:%-3d 语法错误\n",v[index].line,v[index].column); 
		advance();
		return 0;
	} 
	else return v[index].type==mp[s]; 
}
bool match(int ch){
	if(index>v.size())return 0;
	else return v[index].type==ch;
}
void ERROR(string e){
	printf("line:%-3d,column:%-3d %s\n",v[index].line,v[index].column,e.c_str());
	fprintf(flog,"line:%-3d,column:%-3d %s\n",v[index].line,v[index].column,e.c_str());
	fclose(flog);
	exit(111111);
}
void expression();
void block();
void cacu_operator(){
	if(match("<")||match("<=")||match(">")||match(">=")||match("<>")||match("!=")){
		advance(); 
	}else{
		ERROR("错误的运算符");
	}
}
void condition(){
	expression();
	cacu_operator();
	expression();
}
void Y(){ 
	if(match("ID")||match("INT")){
		advance();
	}else if(match("(")){
		advance(); 
		expression();
		if(!match(")"))ERROR("括号不匹配 缺少')'");
		else advance();
	}else{
		ERROR("缺数字或变量");
	}
}
void X(){
	while(1){
		Y();
		if(match("*")||match("/")){
			advance();
			continue;
		}else{
			break;
		}
	}
}

void giveValue(){
	advance();
	if(match("=")){
		advance(); 
		expression(); 
	}else{
		ERROR("没有'='");
	}
}
void check(){
	advance();
	if(match("(")){
		advance();
		condition();
		if(match(")")){
			advance();
			block();
		}else{
			ERROR("无')'");
		}
	}else{
		ERROR("无'('");
	}
}
void circulation(){
	advance();
	if(match("(")){
		advance();
		condition();
		if(match(")")){
			advance();
			block();
		}else{
			ERROR("无')'");
		}
	}else{
		ERROR("无'('");
	}
}
void expression(){
	while(1){
		X();
		if(match("+")||match("-")){
			advance();
			continue;
		}else{
			break;
		}
	}
}

void sentence(){
	if(match("ID")){
		giveValue();
	}else if(match("if")){
		check();
	}else if(match("while")){
		circulation();
	}else if(match("INT")){
		ERROR("不能是数字");
	}else if(match(E)){
		ERROR("语法错误");
		advance();
		sentence(); 
	}else{
		ERROR("标识符不合法");
	} 
} 
void sentences(){
	while(1){
		sentence();
		if(match(";")){
			advance();
			if(match("}")){
				break; 
			}
			else continue;
		}else{
			ERROR("前面没有';'");
		}
	}
}
void block(){
	if(match("{")){
		advance();
		if(match("}")){
			advance(); 
			return;
		}
		sentences();
		if(!match("}"))ERROR("无'}'");
		else advance();
	}else ERROR("无'{'");
}
void process(){
	if(match("main")){
		advance();
		if(match("(")){
			advance();
			if(match(")")){
				advance();
				block();
			}else ERROR("无')'");
		}else ERROR("无'('");
	}else ERROR("无 main方法");
}
char finname[64],foutname[64],flogname[64];
int main(){
	cout<<"输入文件的路径:";
	cin>>finname;
	cout<<"输出的词法分析文件的路径:";
	cin>>foutname;
	cout<<"输出的语法分析文件的路径:";
	cin>>flogname;
	fin=fopen(finname,"r");
	fout=fopen(foutname,"w");
	v=analysis(fin,fout);
	fclose(fin);
	fclose(fout); 
	flog=fopen(flogname,"w");
	process();
	fclose(flog);
	cout<<"accept!"<<endl;
	return 0;
}

5.总结

        语法分析程序不是很完善，可以考虑在排查到错误后继续往后识别。