编译原理词法分析实验流程图传不上去，不传了

编制一个读单词过程，从输入的源程序中，识别出各个具有独立意义的单词，即基本保留字、标识符、常数、分界符四大类。并依次输出各个单词的内部编码及单词符号自身值。（遇到错误时可显示“Error”，然后跳过错误部分继续显示）

1.程序结构描述：函数调用格式、参数含义、返回值描述、函数功能；函数之间的调用关系图（请分项逐一描述）

#define keywordSum 34

保留字数量

string keyword[keywordSum]

保留字

char singleword[20]

单分界符

char doubleword[10]

双分界符首字符

char escapeend[14]

转义字符尾字符

FILE *fin,*fout;

输入输出文件指针

std::map <int,string> f;

std::map <int,string> s;

f是标识符表，s是字符串表。

int cffx()

词法分析函数，无输入。输出0，一切正常，输出1，输入文件错误，输出2，输出文件错误。

string temp="";

当前识别单词。

isalpha(ch)|isalnum(ch)|isdigit(ch)

Ctype库函数，判断ch，isalpha英文字母，isalnum字母加数字，isdigit数字。

2.程序总体执行流程图

传不上去，不传了。。。

在代码中是不同部分，这里是为了简化流程图

3.程序源代码

#include<stdio.h>#include<string>#include<ctype.h>#include<string.h>#include<iostream>#include<map>#define keywordSum  34using namespace std;//下面定义保留，为简化程序，使用字符指针数组保存所有保留字。//如果想增加保留字，可继续添加，并修改保留字数目string keyword[keywordSum]= {"include","if","else","goto","switch","case","do","while","for","contiue","break","return","default","typedef","int","long","short","float","double","char","unsigned","signed","const","void","volatile","enum","struct","union","auto","static","extern","register","read","write"};//下面定义纯单分界符，如需要可添加char singleword[20]= {'(',')','{','}',';',',',':','[',']','#','\'','&'};//下面定义双分界符的首字符char doubleword[10]= {'>','<','=','!','+','-','*','%'};//下面定义转义字符的尾字符char escapeend[14]= {'a','b','f','n','r','t','v','\\','\'','\"','\?','O'};FILE *fin,*fout; //用于指向输入输出文件的指针string Scanin,Scanout;std::map <int,string> f;std::map <int,string> s;int cffx(){char ch;printf("请输入源程序文件名（包括路径）：");cin >> Scanin;if((fin=fopen(Scanin.c_str(),"r"))==NULL) //判断输入文件名是否正确{printf("\n打开词法分析输入文件出错!\n");return(1);//输入文件出错返回错误代码1}printf("请输入词法分析输出文件名（包括路径）：");cin >> Scanout;if((fout=fopen(Scanout.c_str(),"w"))==NULL) //判断输出文件名是否正确{printf("\n创建词法分析输出文件出错!\n");return(2); //输出文件出错返回错误代码2}int f_i=1,s_i=100;ch=getc(fin);string temp="";int i = keywordSum;while(ch!=EOF){while (ch==' '||ch=='\n'||ch=='\t')ch=getc(fin);if (ch==EOF)break;if (ch=='_'||isalpha(ch)){temp += ch;ch=getc(fin);while(ch=='_'||isalnum(ch)){temp += ch;ch=getc(fin);}int bl = 0;while((bl<keywordSum) && strcmp(temp.c_str(),keyword[bl].c_str()))bl++;if (bl<keywordSum)   //不是保留字，输出标识符{fprintf(fout,"%15d\t%s\n",bl+1,temp.c_str());  //输出标识符符号temp = "";}//printf("%03d\t%s\n",bl+1,temp);else //是保留字，输出保留字{fprintf(fout,"%15d\t%s\n",200,temp.c_str());  //输出保留字符号map<int,string> ff(f);int flag=1;while (!ff.empty()&&flag){if(ff.begin()->second.c_str()==temp)flag=0;ff.erase(ff.begin());}if(flag){f[f_i]= temp.c_str();f_i++;}//printf("%\n",i,temp);temp = "";}}else if(isdigit(ch)){int flag = 1;temp += ch;ch=getc(fin);while(isdigit(ch)||(ch=='.'&&flag)){if(ch=='.'&&flag){flag = 0;temp += ch;ch=getc(fin);}else{temp += ch;ch=getc(fin);}}while (ch==' '||ch=='\n'||ch=='\t'){temp += ch;ch=getc(fin);}if(strchr(doubleword,ch)>0||strchr(singleword,ch)>0||ch=='/'){if(flag)fprintf(fout,"%15s\t%s\n","INT",temp.c_str());elsefprintf(fout,"%15s\t%s\n","FLOAT",temp.c_str());}else{while(!(strchr(doubleword,ch)>0||strchr(singleword,ch)>0||ch=='/')){temp += ch;ch=getc(fin);}fprintf(fout,"%15s\t%s\n","ERROR",temp.c_str());}temp = "";}else if(strchr(singleword,ch)>0){temp+=ch;char* ll = singleword;int l = (strchr(singleword,ch))-ll;ch=getc(fin);//读下一个符号以便识别下一个单词fprintf(fout,"%15d\t%s\n",l+keywordSum+2,temp.c_str());  //输出单分界符符号temp="";}else if(strchr(doubleword,ch)>0){temp+=ch;char* ll = doubleword;int l = (strchr(doubleword,ch))-ll;char ch1;ch1=getc(fin);  //读下一个字符判断是否为双分界符if (ch1=='=')   //如果是=，组合双分界符{temp+=ch1;fprintf(fout,"%15d\t%s\n",l+keywordSum+2+30,temp.c_str());ch=getc(fin);  //读下一个符号以便识别下一个单词}else{fprintf(fout,"%15d\t%s\n",l+keywordSum+2+20,temp.c_str());  //输出单或双分界符符号ch = ch1;}temp="";}else if (ch=='/')    //注释处理{ch=getc(fin);  //读下一个字符if (ch=='/'){do   //删除注释{ch=getc(fin);}while (ch!='\n'&&ch!=EOF);ch=getc(fin);//读下一个符号以便识别下一个单词}else if(ch=='*')   //如果是*，则开始处理注释{char ch1;ch1 = getc(fin);do   //删除注释{ch=ch1;ch1=getc(fin);}while ((ch!='*' || ch1!='/')&&ch1!=EOF); //直到遇到注释结束符*/或文件尾ch=getc(fin);//读下一个符号以便识别下一个单词}else //不是*则处理单分界符/{temp+='/';fprintf(fout,"%15d\t%s\n",102,temp.c_str());  //输出单分界符/temp="";}}else if(ch=='\\'){temp+=ch;ch=getc(fin);if(strchr(escapeend,ch)>0){temp+=ch;ch=getc(fin);char* ll = escapeend;int l = (strchr(escapeend,ch))-ll;fprintf(fout,"%15d\t%s\n",l+keywordSum+2+40,temp.c_str());}else{fprintf(fout,"%15d\t%s\n",101,temp.c_str());}temp="";}else if(ch=='"'){char ch1;ch1=getc(fin);int flag = 0;while(ch1!='"'&&!flag){temp += ch1;if(ch1=='\n'){flag = 1;}ch1=getc(fin);}if(flag){fprintf(fout,"%15d\t%c\n",102,ch);fprintf(fout,"%15s\t%s\n","ERROR",temp.c_str());ch = ch1;}else if(ch1=='"'){fprintf(fout,"%15d\t%c\n",102,ch);fprintf(fout,"%15s\t%s\n","STRING",temp.c_str());s[s_i] = temp.c_str();s_i++;fprintf(fout,"%15d\t%c\n",102,ch);ch = getc(fin);}temp="";}else{temp += ch;fprintf(fout,"%15s\t%s\n","ERROR",temp.c_str());temp = "";ch=getc(fin);}}fprintf(fout,"%15s\n","标识表");map<int,string>::iterator it;it = f.begin();for (it; it != f.end(); it++)fprintf(fout,"%15d\t%s\n",it->first,it->second.c_str());fprintf(fout,"%15s\n","字符串表");for(int i=100; i<s_i; i++)fprintf(fout,"%15d\t%s\n",i,s[i].c_str());fclose(fin);//关闭输入输出文件fclose(fout);return 0;  //返回主程序}int main(){int es=0;es=cffx();//调词法分析if (es>0)printf("词法分析有错,编译停止!");elseprintf("词法分析成功!\n");}

4.输出的单词表（符号表）

测试数据为：

#include "stdio.h"int main()23.32.5623,321.23,lkjk{int a;float b,sum=0,ave,max=0;for(a=1;a<=10;a++){scanf("%f",&b);printf("The student's achievement is:%f\n",b);//dsfdsfdsfsdfsum+=b;if(b>max) max=b;}ave=sum/10;printf("The average score of these students is:%f\nThe highest scores of these students are:%f\n",ave,max);return 0;"dadasdassda"}/*fsdfdsfsdfsdfsdfsdfdsffdsfsfssfsdfds*/

输出文件：

45   #
1   include
102   "
STRING   stdio.h
102   "
15   int
200   main
36   (
37   )
ERROR   23.32.5623
41   ,
FLOAT   321.23
41   ,
200   lkjk
38   {
15   int
200   a
40   ;
18   float
200   b
41   ,
200   sum
58   =
INT   0
41   ,
200   ave
41   ,
200   max
58   =
INT   0
40   ;
9   for
36   (
200   a
58   =
INT   1
40   ;
200   a
67   <=
INT   10
40   ;
200   a
60   +
60   +
37   )
38   {
200   scanf
36   (
102   "
STRING   %f
102   "
41   ,
47   &
200   b
37   )
40   ;
200   printf
36   (
102   "
STRING   The student's achievement is:%f\n
102   "
41   ,
200   b
37   )
40   ;
200   sum
70   +=
200   b
40   ;
2   if
36   (
200   b
56   >
200   max
37   )
200   max
58   =
200   b
40   ;
39   }
200   ave
58   =
200   sum
102   /
INT   10
40   ;
200   printf
36   (
102   "
STRING   The average score of these students is:%f\nThe highest scores of these students are:%f\n
102   "
41   ,
200   ave
41   ,
200   max
37   )
40   ;
12   return
INT   0
40   ;
102   "
ERROR   dadasdas

200 sda
102 "
ERROR

39   }
标识表
1   main
2   lkjk
3   a
4   b
5   sum
6   ave
7   max
8   scanf
9   printf
10   sda
字符串表
100   stdio.h
101   %f
102   The student's achievement is:%f\n
103   The average score of these students is:%f\nThe highest scores of these students are:%f\n

5.实验总结

通过本次实验，我掌握了c++中map的基本使用，对DFA的用处的理解更深了，设计了两个小时，编程时间三个小时，其中调试两个小时（主要是测试用例的对比）。

遇到的难题：词法分析的具体范畴（1.什么是词法分析做的，什么是语法分析做的，2.每个单词属于什么的范畴）。

比如stdio.h，它如果属于标识符就应该在标识符中加入点（.）的判断，但标识符不应该有点，解决：加一个string类型，将stdio.h、双引号内的字符统一识别为字符串，由语法进行处理。

比如23.32.2335，它不应该是一个float类型，它应该报错。解决思路：应读到再界符停止，使用flag，允许至多一个小数点，当有两个及以上小数点或有其他字符时输出错误。

分界符的定义，我将分界符分为了单双两种，单分界符中不包含双分界符的首符，双分界符没有尾符将直接输出首符并输出首符对应的code值。

字符串中转义字符是否需要进行转换，我当前只做了非字符串的转义字符处理。如果需要转化，转义字符处理是一个独立的模块，可以直接在字符串处理中使用。