东北大学编译原理实验1

——实验题目

班级：计算机1507班 姓名：罗艺博

一、实验目标：

词法分析扫描器的设计实现。在程序源文件中输入类C语言程序源文件，设计词法分析扫描器，并以TOKEN类别码序列的形式输出扫描结果。

二、实验内容：

1.概要设计：

将程序大致分为：主函数、识别器（有限自动机state_change）、词法分析器（state_to_code）三大部分。

其中，主函数所完成的功能为：打开、读取、关闭文件（即C语言源程序），重置token串等。

识别器的功能为：识别字符，完成token串的生成，判断词法错误等。

词法分析器的功能为：生成token类别码，判断token是否出错等。

学号：201377

主函数、识别器、词法分析器关系图

2.流程图

词法分析扫描器流程图

3.关键函数

a. 识别器（有限自动机state_change）

int state_change(state,ch) //识别器，有限自动机 { if((ch == ' ') || (ch == '\') || (ch == '\\n')) //略去空格、TAB、换行 return 0;

else if((IsAlpha(ch) == 1) || (ch == '_')) //判断是否为字母或'_'，从而判断是关键字、

标识符、CT字符还是ST字符串 {

if(state == 1) return 2; else if(state == 2) return 2; else if(state == 7) return 8; else if(state == 8) return 8;

else if(state == 10) return 11; else if(state == 11) return 11; else if(state == 6) //考虑与PT相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 2; } else

return -1; }

else if(IsNum(ch) == 1) //判断是否为数字，从而进一步判断是为IT,CT,KT {

if(state == 1) return 3; else if(state == 2) return 2; else if(state == 3) return 3; else if(state == 4) return 4; else if(state == 5) return 5; else if(state == 6) //考虑与PT相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 3; }

else

return -1; }

else if(ch == '.') //判断是否为CT中的小数点

{

if(state == 3) return 4; else

return -1; }

else if(ch == '\\'') {

if(state == 1) return 7;

else if(state == 6) {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 7; }

else if(state == 8) return 9; else

return -1; }

else if(ch == '\\\"') {

if(state == 1) return 10; else if(state == 6) {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 10; }

else if(state == 11) return 12; else

return -1; } else {

//判断是否为CT字符

//考虑与PT相连的情况

//判断是否为ST

//考虑与PT相连的情况

//判断是否为PT

if(state == 1) return 6; else if(state == 2) //考虑与IT,KT相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 3) //考虑与CT相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 9) //考虑与CT字符相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 12) //考虑与ST相连的情况 {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 6) {

if(((i == 1)&&(ch == '='))&&((token[0] == '>')||(token[0] == '<')||(token[0] == '=')))

{

return 6; } else

return 0; } else

return -1; } }

b. 词法分析器（state_to_code）

int state_to_code(state_before,token) {

int n; //循环 code1 = 0; code2 = 0; switch(state_before) { case {

if(strcmp(token, KT[n]) == 0) {

code1 = 1; code2 = n;

printf(\"<%d %d>\\n\code1,code2

break; } }

if(code1 == 0) {

printf(\"<2 0>\\n\"); }

break; case break; case 5: printf(\"<3 0>\\n\"); break; case 6: for(n=0; n<18; n++) {

if(strcmp(token, PT[n]) == 0) {

code1 = 6; code2 = n;

printf(\"<%d %d>\\n\//词法分析器 2://KT <1 X> for(n=0; n<6; n++) //why改变不了全局的//IT <2 0> 3://CT <3 0> printf(\"<3 0>\\n\"); //CT(小数) <3 0> //PT <6 X> //why改变不了全局的

code1,code2

break; } }

break; case 9: //CT字符 <4 0> printf(\"<4 0>\\n\"); break; case 12: printf(\"<5 0>\\n\"); break; default : printf(\"Sorry it's going wrong!\"); } return 0; }

c.判断函数

int IsAlpha(char c) {

if(((c>='a')&&(c<='z')) || ((c>='A')&&(c<='Z'))) return 1; else

return 0; }

int IsNum(char c) {

if(c>='0'&&c<='9') return 1; else

return 0; }

int IsKey(char *word) {

int m,n;

for(n=0; n<6; n++) {

if((m = strcmp(word, KT[n])) == 0) {

//ST <5 0>

//不被任何词识别

//判断是否为字母

//判断是否为数字

//判断是否为关键字

if(n == 0) return 2; return 1; } }

return 0; }

int IsDelimiter(char *token) {

int m,n;

for(n=0; n<18; n++) {

if((m = strcmp(token, PT[n])) == 0) {

if(n == 0) return 2; return 1; } }

return 0; }

//判断是否为界符

源程序代码：（加入注释）

#include #include #include FILE *fp; char input[200]; char token[10];

char word[10]; //存放识别出的单词流 char ch; int p;

int i=0; //token下标 int state;

int code2; int state_before;

//存放之前状态

//存放状态标记

//存放当前读入字符

//input下标

//存放输入字符串

//存放构成单词符号的符号串

int code1; //token类别码,1->KT,2->IT,3->CT,4->CTT,5->ST,6->PT

int num;

//存放整形值

char *KT[6] = {\"int\ //1

char *PT[18] = {\">=\ //18 //char *IT[20] = {NULL}; //2 //char *CT[20] = {NULL}; //3 //char *CTT[20] = {\"\\0\ //4 //char *ST[20] = {NULL}; //5 int state_change(state, ch);

int state_to_code(state_before, token); int IsAlpha(char c); int IsNum(char c); int IsKey(char *word); int IsDelimiter(char *token); int main() {

state = 1; //state=1设为初始态，0设为结束态

if((fp = fopen(\"E:\\\\Cwork\\\\Analysis\\\est.txt\ {

printf(\"Cannot open file.\\n\"); exit(1); } while((ch = fgetc(fp)) != '#')

{

{

printf(\" Err! Stupid man!\\n\"); }

token[i++] = ch;

{

state_to_code(state_before,token);

state_before = state;

state = state_change(state,ch);

if(state == -1)

else if (state != 0)

{

} else {

if(state_before != 1)

} }

int state_change(state,ch) //识别器，有限自动机 {

return 0;

else if((IsAlpha(ch) == 1) || (ch == '_')) CT字符还是ST字符串 {

if(state == 1) return 2; else if(state == 2) return 2; else if(state == 7) return 8; else if(state == 8) return 8; else if(state == 10) return 11; else if(state == 11) return 11; else if(state == 6) {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 2; } else

return -1; }

else if(IsNum(ch) == 1) {

//判断是否为数字，从而进一步判断是为IT,CT,KT

//考虑与PT相连的情况

//判断是否为字母或'_'，从而判断是关键字、标识符、

if((ch == ' ') || (ch == '\') || (ch == '\\n'))

//略去空格、TAB、换行

fclose(fp); return 0;

}

memset(token,0,10);

i = 0; state = 1;

}

if(state == 1) return 3; else if(state == 2) return 2; else if(state == 3) return 3; else if(state == 4) return 4; else if(state == 5) return 5; else if(state == 6) {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 3; } else

return -1; }

else if(ch == '.') {

if(state == 3) return 4; else

return -1; }

else if(ch == '\\'') {

if(state == 1) return 7; else if(state == 6) {

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 7; }

else if(state == 8) return 9; else

return -1; }

else if(ch == '\\\"')

//考虑与PT相连的情况

//判断是否为CT中的小数点

//判断是否为CT字符

//考虑与PT相连的情况

//判断是否为ST

{

if(state == 1) return 10; else if(state == 6)

{

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 10; }

else if(state == 11) return 12; else

return -1; } else

{

if(state == 1) return 6; else if(state == 2)

{

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 3)

{

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 9)

{

state_to_code(state_before,token); i = 0;

memset(token,0,10); return 6; }

else if(state == 12)

{

state_to_code(state_before,token); i = 0;

//考虑与PT相连的情况

//判断是否为PT

//考虑与IT,KT相连的情况

//考虑与CT相连的情况

//考虑与CT字符相连的情况

//考虑与ST相连的情况

memset(token,0,10); return 6; }

else if(state == 6) {

if(((i == 1)&&(ch == '='))&&((token[0] == '>')||(token[0] == '<')||(token[0] == '='))) {

return 6; } else return 0; } else

return -1; } }

int state_to_code(state_before,token) {

int n; //循环 code1 = 0; code2 = 0;

{

if(strcmp(token, KT[n]) == 0) {

code1 = 1; code2 = n;

printf(\"<%d %d>\\n\ //why改变不了全局的code1,code2 break; } }

if(code1 == 0) {

printf(\"<2 0>\\n\"); } break;

case

3:

//CT <3 0>

printf(\"<3 0>\\n\");

//IT <2 0>

switch(state_before)

{ case

2:

//KT <1 X>

for(n=0; n<6; n++)

//词法分析器

break; case 5:

printf(\"<3 0>\\n\"); break; case 6:

for(n=0; n<18; n++) {

if(strcmp(token, PT[n]) == 0) {

code1 = 6; code2 = n;

printf(\"<%d %d>\\n\ //why改变不了全局的code1,code2 break; } } break; case 9:

printf(\"<4 0>\\n\"); break; case 12:

printf(\"<5 0>\\n\"); break; default :

printf(\"Sorry it's going wrong!\"); }

int IsAlpha(char c) {

if(((c>='a')&&(c<='z')) || ((c>='A')&&(c<='Z'))) return 1; else return 0; }

int IsNum(char c) {

if(c>='0'&&c<='9') return 1; else return 0; }

//判断是否为数字

//判断是否为字母

} return 0;

//不被任何词识别

//ST <5 0>

//CT字符 <4 0>

//PT <6 X>

//CT(小数) <3 0>

int IsKey(char *word)

{

int m,n;

for(n=0; n<6; n++) {

if((m = strcmp(word, KT[n])) == 0) {

if(n == 0) return 2; return 1; } } return 0; }

int IsDelimiter(char *token)

{

int m,n;

for(n=0; n<18; n++) {

if((m = strcmp(token, PT[n])) == 0) {

if(n == 0) return 2; return 1; } } return 0; }

程序运行结果：（截屏）

输入：

//判断是否为关键字

//判断是否为界符

输出：