http://blog.csdn.net/violet_xrym/article/details/9120239
一、实验概述
1.1、实验要求
选择计算机高级语言之一-----C语言,运用恰当的此法分析技术线路,设计和实现其对应的词法分析器。
建议:编程语言,选择《计算机程序设计》课程所采用的语言。
提示:技术线路选择如下两种之一:
正则式→NFA→DFA→minDFA→程序设计
或正则文法→NFA→DFA→minDFA→程序设计。
要求:分析器输出结果存入到磁盘文件中,具有出错处理功能。
1.2、实验目的
1)加深对编译原理及其构造词法分析器的原理和技术理解与应用,进一步提高学生编程能力;
2)培养、提高学生分析问题、解决问题的综合能力;
3)整理资料,撰写规范的实验报告;
二、系统分析
2.1、系统需求
根据C语言语法,待分析的词可以分为如下几类:
(1) 关键字
如if, else, whlile, int 等。
(2) 标示符
开头只能为字母,后面可以接数字或者字母,用来表示各种名字,如变量名、常量名和过程名等
(3) 常数
各种类型的常数,如整型(1, 30),浮点型(2.16),字符串型(”AHD”),字符型(’A’)
(4) 运算符与界符
如+, *, <= , 逗号等。
2.2、系统功能
读入一个C语言源程序(经过预处理的),对每一个单词输出一些三元组的集合。
2.3、系统实现步骤
按照如下顺序构造词法分析器:
(1) 设计出各类单词的正规式,画出有限状态自动机。
(2) 将各类单词的正规式转换成相应的NFA M,并将其合并成一个NFA M`
(3) 将NFA M`转换成对应的DFA M``
(4) 将DFA M``最小化为DFA M```
(5) 根据DFA M```用C语言设计出相应的词法分析器。
三、系统设计
3.1、有限状态自动机设计
状态机说明:由于单词的构成较为复杂,所以再设计时,边的变迁不再是一个字符,而是一个函数。若当前输入串满足该函数,则当前状态可以变迁到该边连接的下一状态。
根据终态可以看出自动机可以分离的状态有:
INT | 整数 |
FLOAT | 浮点数 |
CHAR | 字符型 |
CHARS | 字符串型 |
IDENT | 标识符(包括关键字) |
SYMBOL | 符号 |
其中,关键字的分离在辅助程序中进行。
3.2、单词符号对应的种别码
种别码 | 单词符号及说明 | 种别码 | 单词符号及说明 |
0 | INT(整数) | 30 | >= |
1 | FLOAT(浮点数) | 31 | <= |
2 | CHAR(字符型) | 32 | > |
3 | CHAR(字符串型) | 33 | < |
4 | IDENT(标识符) | 34 | == |
5 | if | 35 | = |
6 | else | 36 | != |
7 | int | 37 | ++ |
8 | char | 38 | += |
9 | float | 39 | + |
10 | double | 40 | / |
11 | long | 41 | - |
12 | short | 42 | \ |
13 | return | 43 | ; |
14 | while | 44 | ( |
15 | break | 45 | ) |
16 |
| 46 | { |
17 |
| 47 | } |
18 |
| 48 | [ |
19 |
| 49 | ] |
20 |
| 50 | : |
21 |
| 51 | -> |
22 |
| 52 | ? |
23 |
| 53 | , |
24 |
| 54 | . |
25 |
| 55 | * |
26 |
| 56 |
|
27 |
| 57 |
|
28 |
| 58 |
|
29 |
| 59 |
|
3.3、基本数据结构及代码设计
代码是用C++完成的。为状态机定义了两个基本的结构体,分别为STATE和LIST。其中STATE是LIST的友元,STATE表示的是状态机中的一个状态,包括error和start等状态。LIST的实例是依附于一个STATE的实例存在的,他表示一条边,边的值是一个函数指针,该边指向一个满足该函数的另一状态。
另外用到了STL库中的MAP模板,定义为map<string,string>type用以存放关键字及其对应的种别码。
程序的输出为三元组的集合,其中三元组定义为<单词名, 单词含义 ,种别码>。如一个标识符abc的三元组为<abc, IDENT , 4> 。若某个单词错误,则会输出 error: name。 输出的最后会显示共识别了多少单词,并发现多少错误。
四、系统实现
4.1 系统运行
l 在命令行里直接输入待翻译的文件和输出的文件名。如果没有给参数,缺省为输入”input.txt”,输出”output.txt”。
l 输出的结果。
4.2 系统结果
l input.txt
其中,第六行为错误行。
int main()
{
freopen("input.txt","r",stdin);
char input[255],*s = input;
int t = 1;
floatp = 12.4;
int 0a = 2;
init();
while(gets(s))
{
curn = 0;
printf("Line %d:\n",t++);
while((*s)!= 0)
{
while((*s) == ' ') s++;
curn = 0;
curtype = 0;
print(s,start->start(s));
s += curn;
}
}
return 0;
}
l output.txt
********Line 1*********:
<int, int, 7>
<main, IDENT, 3>
<(, (, 34>
<), ), 35>
********Line 2*********:
<{, {, 36>
********Line 3*********:
<freopen, IDENT, 3>
<(, (, 34>
<"input.txt",CHARS, 2>
<,, ,, 43>
<"r", CHARS, 2>
<,, ,, 43>
<stdin, IDENT, 3>
<), ), 35>
<;, ;, 33>
********Line 4*********:
<char, char, 8>
<input, IDENT, 3>
<[, [, 38>
<255, INT, 0>
<], ], 39>
<,, ,, 43>
<*, *, 45>
<s, IDENT, 3>
<=, =, 25>
<input, IDENT, 3>
<;, ;, 33>
********Line 5*********:
<int, int, 7>
<t, IDENT, 3>
<=, =, 25>
<1, INT, 0>
<;, ;, 33>
********Line 6*********:
<float, float, 9>
<p, IDENT, 3>
<=, =, 25>
<12.4, FLOAT, 1>
<;, ;, 33>
********Line 7*********:
<int, int, 7>
error: 0a
<=, =, 25>
<2, INT, 0>
<;, ;, 33>
********Line 8*********:
********Line 9*********:
<init, IDENT, 3>
<(, (, 34>
<), ), 35>
<;, ;, 33>
********Line 10*********:
********Line 11*********:
<while, while, 14>
<(, (, 34>
<gets, IDENT, 3>
<(, (, 34>
<s, IDENT, 3>
<), ), 35>
<), ), 35>
********Line 12*********:
<{, {, 36>
********Line 13*********:
<curn, IDENT, 3>
<=, =, 25>
<0, INT, 0>
<;, ;, 33>
********Line 14*********:
<printf, IDENT, 3>
<(, (, 34>
<"Line %d:\n",CHARS, 2>
<,, ,, 43>
<t, IDENT, 3>
<++, ++, 27>
<), ), 35>
<;, ;, 33>
********Line 15*********:
<while, while, 14>
<(, (, 34>
<(, (, 34>
<*, *, 45>
<s, IDENT, 3>
<), ), 35>
<!=, !=, 26>
<0, INT, 0>
<), ), 35>
********Line 16*********:
<{, {, 36>
********Line 17*********:
<while, while, 14>
<(, (, 34>
<(, (, 34>
<*, *, 45>
<s, IDENT, 3>
<), ), 35>
<==, ==, 24>
<' ', CHAR, 2>
<), ), 35>
<s, IDENT, 3>
<++, ++, 27>
<;, ;, 33>
********Line 18*********:
<curn, IDENT, 3>
<=, =, 25>
<0, INT, 0>
<;, ;, 33>
********Line 19*********:
<curtype, IDENT, 3>
<=, =, 25>
<0, INT, 0>
<;, ;, 33>
********Line 20*********:
<print, IDENT, 3>
<(, (, 34>
<s, IDENT, 3>
<,, ,, 43>
<start, IDENT, 3>
<-, -, 31>
<>, >, 23>
<start, IDENT, 3>
<(, (, 34>
<s, IDENT, 3>
<), ), 35>
<), ), 35>
<;, ;, 33>
********Line 21*********:
<s, IDENT, 3>
<+=, +=, 28>
<curn, IDENT, 3>
<;, ;, 33>
********Line 22*********:
<}, }, 37>
********Line 23*********:
<}, }, 37>
********Line 24*********:
********Line 25*********:
<return, return, 13>
<0, INT, 0>
<;, ;, 33>
********Line 26*********:
<}, }, 37>
*******************************
1 error!
116 Word Have Been Found Out!
源代码:
- #include <iostream>
- #include <string.h>
- #include <map>
- #include <stdio.h>
- #define num_before_symbol 20
- using namespace std;
- bool isNum(char *a);
- bool isWord(char *a);
- bool isSymbol(char *a);
- bool isNULL(char *a);
- map<string,string>type;
- char symbol[][10] = {">=","<=","<",">","==","=","!=","++","+=","+","/","-","\\",";","(",")","{","}",
- "[","]",":","->","?",",",".","*","\0"};
- int curn = 0;
- int curtype = 0;
- int nerror = 0;
- class STATE;
- class LIST;
- class STATE
- {
- LIST *list;
- static STATE *error;
- public:
- static int count;
- int type;
- char *name;
- void enlist(bool (*fun)(char *),STATE *out);
- const STATE *next(char *in)const;
- const STATE *start(char *)const;
- STATE(char *name);
- ~STATE();
- };
- class LIST{
- LIST *next;
- bool (*fun)(char *);
- STATE *output;
- LIST(bool (*fun)(char *),STATE *out);
- ~LIST();
- friend class STATE;
- };
- STATE *STATE::error = 0;
- int STATE::count = 0;
- LIST::LIST(bool (*fun)(char *),STATE *out)
- {
- this->next = NULL;
- this->fun = fun;
- this->output = out;
- }
- LIST::~LIST() //怎么delete????
- {
- if(this->next!=NULL)
- delete this->next;
- }
- const STATE *STATE::next(char *in)const
- {
- LIST *p = list;
- //if(this == error) return error;
- while(p!=NULL)
- {
- if(p->fun(in))
- return p->output;
- else
- p = p->next;
- }
- return error;
- }
- const STATE *STATE::start(char *s)const
- {
- const STATE *p;
- if(list == NULL)
- {
- if(this != error)
- count++;
- else
- {
- while(isWord(s))
- curn++;
- }
- return this;
- }
- p = this->next(s);
- if(p == error)
- return error; //error是否要加前缀
- return p->start(s+1);
- }
- STATE::STATE(char *name)
- {
- if(name == 0)
- {
- error = this;
- this->type = 1;
- return;
- }
- if(strcmp(name,"SYMBOL"))
- this->type = 0;
- else
- this->type = 1;
- this->name = new char[strlen(name)]; //strlen+1
- strcpy(this->name,name);
- this->list = NULL;
- }
- STATE::~STATE()
- {
- if(list)
- {
- delete list;
- list = 0;
- }
- if(name)
- {
- delete name;
- name = 0;
- }
- }
- void STATE::enlist(bool (*fun)(char *),STATE *out)
- {
- LIST *p = new LIST(fun,out);
- LIST *cur = this->list;
- if(cur == NULL)
- this->list = p;
- else
- {
- while(cur->next!=NULL)
- cur = cur->next;
- cur->next = p;
- }
- }
- bool mystrcmp(char *a,char *s)
- {
- int i = 0;
- while(s[i]!='\0')
- {
- if(a[i]!=s[i])
- return false;
- i++;
- }
- return true;
- }
- bool isNum(char *a)
- {
- if(a[0]<='9' && a[0]>='0')
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isDot(char *a)
- {
- if(a[0] == '.')
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isWord(char *a)
- {
- if((a[0]<='Z' && a[0]>='A') || (a[0]>='a' && a[0]<='z'))
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isNotNumOrWord(char *a)
- {
- if((!(a[0]<='9' && a[0]>='0')) && !isWord(a))
- {
- return true;
- }
- curn--;
- return false;
- }
- bool isSymbol(char *a)
- {
- int i = 0;
- while(strcmp(symbol[i],"\0"))
- {
- if(mystrcmp(a,symbol[i]))
- {
- curtype = i;
- curn = strlen(symbol[i]);
- break;
- }
- i++;
- }
- }
- bool isDQuotation(char *a)
- {
- if(a[0] == '"')
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isNotDQuotation(char *a)
- {
- if(a[0] != '"')
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isNotSQuotation(char *a)
- {
- if(a[0]!='\'')
- {
- curn++;
- return true;
- }
- return false;
- }
- bool isSQuotation(char *a)
- {
- if(*(a-1)!='\\' && (*a)== '\'')
- {
- curn++;
- return true;
- }
- return false;
- }
- STATE *start = new STATE("start");
- STATE *s1 = new STATE("s1");
- STATE *s2 = new STATE("s2");
- STATE *s3 = new STATE("s3");
- STATE *s4 = new STATE("s4");
- STATE *s5 = new STATE("s5");
- STATE *s6 = new STATE("s6");
- STATE *INT = new STATE("INT");
- STATE *FLOAT = new STATE("FLOAT");
- STATE *IDENT = new STATE("IDENT");
- STATE *SYMBOL = new STATE("SYMBOL");
- STATE *CHAR = new STATE("CHAR");
- STATE *CHARS = new STATE("CHARS");
- STATE error(0);
- void init()
- {
- start->enlist(isNum,s1);
- start->enlist(isWord,s3);
- start->enlist(isSymbol,SYMBOL);
- start->enlist(isDQuotation,s4);
- start->enlist(isSQuotation,s5);
- s1->enlist(isNum,s1);
- s1->enlist(isDot,s2);
- s1->enlist(isNotNumOrWord,INT);
- s1->enlist(isWord,&error); //需要将错误部分剩下的跳过,对error类型进行标识。
- s2->enlist(isNum,s2);
- s2->enlist(isNotNumOrWord,FLOAT);
- s2->enlist(isWord,&error);
- s3->enlist(isWord,s3);
- s3->enlist(isNum,s3);
- s3->enlist(isNotNumOrWord,IDENT);
- s4->enlist(isNotDQuotation,s4); //if is """ will get wrong answer
- s4->enlist(isDQuotation,CHARS);
- s5->enlist(isNotSQuotation,s6);
- s5->enlist(isSQuotation,&error);
- s6->enlist(isSQuotation,CHAR);
- type["INT"] = "0";
- type["FLOAT"] = "1";
- type["CHAR"] = "2";
- type["CHARS"] = "2";
- type["IDENT"] = "3";
- type["if"] = "5";
- type["else"] = "6";
- type["int"] = "7";
- type["char"] = "8";
- type["float"] = "9";
- type["double"] = "10";
- type["long"] = "11";
- type["short"] = "12";
- type["return"] = "13";
- type["while"] = "14";
- type["break"] = "15";
- //引号算什么??
- }
- void print(char *s,const STATE *p)
- {
- int i = 0;
- char temp[255];
- for(i = 0;i<curn;i++)
- temp[i] = s[i];
- temp[i] = '\0';
- if(p->name == 0)
- {
- printf("error: %s\n",temp);
- nerror++;
- return;
- }
- printf("<");
- for(i = 0;i<curn;i++)
- printf("%c",*(s+i));
- if(curtype == 0)
- {
- if(type.find(temp) == type.end())
- cout<<", "<<p->name<<", "<<type[p->name]<<">"<<endl;
- else
- cout<<", "<<temp<<", "<<type[temp]<<">"<<endl;
- }
- else
- printf(", %s, %d>\n",symbol[curtype],num_before_symbol+curtype);
- }
- int main(int argv,char *argc[])
- {
- char temp[2][255];
- if(argv<2)
- strcpy(temp[0],"input.txt");
- else
- strcpy(temp[0],argc[1]);
- if(argv<3)
- strcpy(temp[1],"output.txt");
- else
- strcpy(temp[1],argc[2]);
- freopen((const char*)temp[0],"r",stdin);
- freopen((const char*)temp[1],"w",stdout);
- char input[255],*s = input;
- int t = 1;
- init();
- while(gets(s))
- {
- curn = 0;
- printf("********Line %d*********:\n",t++);
- if(t == 18)
- t = 18;
- while((*s)!='\0') //n is last edit
- {
- while((*s) == ' ') s++; //滤掉空格
- curn = 0;
- curtype = 0;
- print(s,start->start(s));
- s += curn;
- while((*s) == ' ') s++; //滤掉空格
- }
- s = input;
- printf("\n");
- }
- printf("*******************************\n%d error!\n%d Word Have Been Found Out!\n",nerror,STATE::count);
- return 0;
- }