实验一词法分析班级:计科12-3 学号:08123282 姓名:樊鑫一．实验目的1、学会针对DFA转换图实现相应的高级语言源程序。

2、深刻领会状态转换图的含义，逐步理解有限自动机。

3、掌握手工生成词法分析器的方法，了解词法分析器的内部工作原理。

二．实验内容TINY计算机语言的编译程序的词法分析部分实现。

从左到右扫描每行该语言源程序的符号，拼成单词，换成统一的内部表示（token）送给语法分析程序。

为了简化程序的编写，有具体的要求如下：（1）数仅仅是整数。

（2）空白符仅仅是空格、回车符、制表符。

（3）代码是自由格式。

（4）注释应放在花括号之内，并且不允许嵌套三．实验要求要求实现编译器的以下功能：（1）按规则拼单词,并转换成二元式形式（2）删除注释行（3）删除空白符 (空格、回车符、制表符)（4）列表打印源程序，按照源程序的行打印，在每行的前面加上行号，并且打印出每行包含的记号的二元形式（5）发现并定位错误，词法分析进行具体的要求：（1）记号的二元式形式中种类采用枚举方法定义；其中保留字和特殊字符是每个都一个种类，标示符自己是一类，数字是一类；单词的属性就是表示的字符串值。

（2）词法分析的具体功能实现是一个函数GetToken()，每次调用都对剩余的字符串分析得到一个单词或记号识别其种类，收集该记号的符号串属性，当识别一个单词完毕，采用返回值的形式返回符号的种类，同时采用程序变量的形式提供当前识别出记号的属性值。

这样配合语法分析程序的分析需要的记号及其属性，生成一个语法树。

（3）标示符和保留字的词法构成相同，为了更好的实现，把语言的保留字建立一个表格存储，这样可以把保留字的识别放在标示符之后，用识别出的标示符对比该表格，如果存在该表格中则是保留字，否则是一般标示符。

四．实验程序#include <iostream>#include <iomanip>#include <ctype.h>#include <fstream>#include <string>using namespace std;enum WordType {WRONG, NUMBER, BIAOSHI, IF, THEN, ELSE, END, REPEAT, UNTIL, //9 READ, WRITE, PLUS, SUBTRACT, MULTIPLY, DIVIDEY, EQUAL, LESSTHAN, COLON,//18ASSIGNMENT, LINE_FEED, SPACE, TAB, BRACKET_L, BRACKET_R}; //24enum CharType {ALPHABET, OTHER};enum WrongType {ZERO, ALP_NUM, NUM_ALP, UNLEAGL_S, NO_MA TCH, UNKNOW}; char *Words [] = {"wrong", "number", "biaoshi", "if", "then", "else", "end", "repeat", "until", //9 "read", "write", "+", "-", "*", "/", "=", "<", ":",//18":=", "\n", " ", " ", "{", "}"}; //24typedef struct{char *str;int wordtype;}Two;char ArrayChar[100], cbuffer;int i=-1, numline = 1, wordtype;string LineChar;Two T;ifstream fp("source.txt",ios::in);int main (){void HandleAlphabet();void HandleNumber ();void HandleOther ();if (!fp)cout<<"文件打开错误!"<<endl;else{cout << setiosflags (ios::left) << setw (6) << "行数" << "(";cout << setiosflags (ios::left) << setw (10)<< "类别编码" << ",";cout << setiosflags (ios::left) << setw (20) << "字符" << ")" << endl;fp.get (cbuffer);while (!fp.eof ()){if(cbuffer == '\n')//忽略换行符{cout << "第" << numline << " 行所有字符：" << LineChar << endl;numline++;LineChar = "";fp.get(cbuffer);}else if (cbuffer == ' ')//忽略空字符{LineChar += " ";fp.get(cbuffer);}else if (cbuffer == ' ')//忽略制表符{LineChar += " ";fp.get(cbuffer);}else if (isalpha(cbuffer))//判断是否是字母{HandleAlphabet ();}else if (isdigit(cbuffer))//判断是否是数字{HandleNumber ();}else //其他字符HandleOther ();i = -1;}}cout << "第" << numline << " 行所有字符：" << LineChar << endl;fp.close ();system ("pause");return 0;}int Match(char str[], int chartype)//查找匹配的字符{int i;switch (chartype){case ALPHABET:for (i = IF; i <= WRITE; i++){if (strcmp(Words[i],str) == 0)return i;}case OTHER:for (i = PLUS; i <= ASSIGNMENT; i++){if (strcmp(Words[i],str) == 0)return i;}}return WRONG;}void TypeWrong (int wrongtype, int line){switch (wrongtype){case ZERO:break;case ALP_NUM:cout << "字母后面不能紧跟数字！";break;case NUM_ALP:cout << "数字后面不能紧跟字母！";case UNLEAGL_S:break;cout << "非法特殊符号！";break;case NO_MA TCH:cout << "没有与第" << line << "行""匹配的""!";break;default:cout << "其它类型错误！";break;}}Two ConvertTwo (char str[], int wordtype, int wrongtype, int numline, int line)//进行二元转换{Two T;T.wordtype = wordtype;T.str = str;cout << setiosflags (ios::left) << setw (6) << numline << "(";cout << setiosflags (ios::left) << setw (10) << T.wordtype << ", ";cout << setiosflags (ios::left) << setw (20) << T.str << ")";if (T.wordtype == WRONG)TypeWrong (wrongtype, line);cout << endl;return T;}void HandleAlphabet ()//首字符为字母时的处理{bool mark = true;while(!fp.eof() && isalpha(cbuffer)){ArrayChar[++i]=cbuffer;fp.get(cbuffer);}if (isdigit (cbuffer)){mark = false;while(!fp.eof() && (isalpha(cbuffer)||isdigit(cbuffer))){ArrayChar[++i]=cbuffer;fp.get(cbuffer);}}ArrayChar[i+1]='\0';LineChar += ArrayChar;if (mark){wordtype = Match(ArrayChar, ALPHABET);T = ConvertTwo(ArrayChar,(BIAOSHI > wordtype ? BIAOSHI : wordtype), ZERO, numline, numline);}elseT = ConvertTwo(ArrayChar,WRONG, ALP_NUM, numline, numline);}void HandleNumber ()//首字符为数字时的处理{bool mark = true;while (!fp.eof() && isdigit(cbuffer)){ArrayChar[++i]=cbuffer;fp.get(cbuffer);}if (isalpha(cbuffer)){mark = false;while(!fp.eof() && (isalpha(cbuffer)||isdigit(cbuffer))){ArrayChar[++i]=cbuffer;fp.get(cbuffer);}}ArrayChar[i+1]='\0';LineChar += ArrayChar;if (mark)T = ConvertTwo(ArrayChar, NUMBER, ZERO, numline, numline);elseT = ConvertTwo(ArrayChar,WRONG, NUM_ALP, numline, numline);}void DeleteNote ()//删除注释{int record = numline;while (!fp.eof() && cbuffer != '}'){fp.get(cbuffer);while (!fp.eof() && cbuffer != '}'){if (cbuffer == '\n'){ArrayChar [i + 1] = '\0';LineChar += ArrayChar;cout << "第" << numline << " 行所有字符：" << LineChar << endl;LineChar = "";numline++;i = -1;fp.get(cbuffer);}ArrayChar[++i]=cbuffer;fp.get(cbuffer);}ArrayChar[i + 1]='\0';if (cbuffer == '}'){ArrayChar [++i] = '}';ArrayChar[i + 1]='\0';T = ConvertTwo("}", BRACKET_R, ZERO, numline, numline);}else{T = ConvertTwo("", WRONG, NO_MA TCH, numline, record);}}LineChar += ArrayChar;fp.get(cbuffer);}void HandleOther ()//字符为特殊字符时的处理{ArrayChar [++i] = cbuffer;if (ArrayChar [i] == '{')//删除注释{ConvertTwo("{", BRACKET_L, ZERO, numline, numline);DeleteNote ();}else//其他字符{fp.get(cbuffer);while (!fp.eof() && cbuffer != '\n' && cbuffer != ' ' && cbuffer != ' '&& !isalpha(cbuffer) && !isdigit(cbuffer)){ArrayChar [++i] = cbuffer;fp.get(cbuffer);}ArrayChar[i + 1]='\0';wordtype = Match(ArrayChar, OTHER);LineChar += ArrayChar;T = ConvertTwo(ArrayChar, wordtype, (wordtype > 0 ? 0: UNLEAGL_S), numline, numline);}}五．实验结果六.实验体会通过本次实验学会了针对DFA转换图实现相应的高级语言源程序，并且深刻领会了状态转换图的含义，逐步理解有限自动机，同时掌握手工生成词法分析器的方法，了解词法分析器的内部工作原理。