NCWU慎用

1．实验要求
（1）从源程序文件中读取有效字符流并将其分析识别单词符号，转换成二元组内部表示形式输出。
（2）可视化方式展示词法分析识别过程或者词法分析器工作原理（选做）。
（3）实验时间4学时。
（4）实验完成后，要提交实验报告（包括源程序清单）。
2．实验内容
2.1主程序设计考虑：
主程序的说明部分为各种表格和变量安排空间（关键字和特殊符号表）。
id 和ci 数组分别存放标识符和常数；还有一些为造表填表设置的变量。
主程序的工作部分建议设计成便于调试的循环结构。每个循环处理一个单词；调用词法分析过程；输出每个单词的内部码（种别编码，属性值）。建议从文件中读取要分析的符号串。
2.2词法分析过程考虑
该过程根据输入单词的第一个有效字符（有时还需读第二个字符），判断单词种别，产生种别编码。对于标识符和常数，需分别与标识符表和常数表中已登记的元素相比较，如表中已有该元素，则记录其在表中的位置，如未出现过，将标识符按顺序填入数组 id 中，将常数存入数组中 ci 中，并记录其在表中的位置。
注：所有识别出的单词都用二元组表示。第一个表示单词的种别编码。例如：关键字的 t=1；标识符的 t=2；常数 t=3；运算符 t=4；界符 t=5。第二个为该单词在各自表中的指针或内部码值（常数表和标识符表是在编译过程中建立起来的。其 i 值是根据它们在源程序中出现的顺序确定的）。
关键字和特殊符号如下（表中数字只是标记，不代表种别编码，种别编码自己定义）：

将词法分析程序设计成独立一遍扫描源程序的结构。参考设计流程图如下：

图1 词法分析程序流程图

实验代码：

// Exper1.cpp : 此文件包含 "main" 函数。程序执行将在此处开始并结束。
#include <iostream>
#include <stdlib.h>
#include <fstream>
#include <string>
using namespace std;static string KeyWord[15] = {//关键字  t=1"int","char","float","void","if","else","do","const","while","scanf","printf","return","main","read","void",
};
static string Operator[23] = {//运算符  t=2"+", "-", "*", "/","%","=","==",">","<","!=",">=","<=","&&","||","!","<>","++","--","/=","*=","+=","-=","%=",
};
static string Boundary[9] = {//界符  t=3"(", ")" ,"{" ,"}" ,";" ,"," , "\"" ,"[","]",
};
static string Id[1000] = { "" };//标识符表  t=4
static string Const[1000] = { "" };//常数  t=4
static int Id_num = 0;
static int Const_num = 0;void Pretreatment(string str[], int Len, string& Str) {   //预处理int num = 0;int flag = 0;for (int i = 0; i < Len; i++) {int len = str[i].length();for (int j = 0; j < len; j++) {if (flag != 0) {if (str[i][j] == '/' && str[i][j + 1] == '*') {flag++;}//可能在/* */里面还有/* */if (str[i][j] == '*' && str[i][j + 1] == '/') {flag--;j++;}}else {if (str[i][j] == '/') {//处理单行注释if (str[i][j + 1] == '/')break;//处理多行注释else if (str[i][j + 1] == '*') {flag++;}else {//是/运算符Str += str[i][j];}}else if (str[i][j] == ' ') {//将一连串空格归结为一个空格if (str[i][j - 1] == ' ') {continue;}else {Str += str[i][j];}}else if (str[i][j] != '\n' && str[i][j] != '\t' && str[i][j] != '\r' && str[i][j] != '\v') {Str += str[i][j];}}//if_flag}//for_j}//for_i
}bool IsLetter(char c) {//判断是不是字母或下划线if (c >= 'a' && c <= 'z' || c >= 'A' && c <= 'Z' || c == '_') {return true;}return false;
}
bool IsNumber(char c) {//判断是否为数字if (c >= '0' && c <= '9') {return true;}return false;
}
/*复杂写法
int IsKeyword(string s) {//判断是否为关键字 for (int i = 0; i < 15; i++) {if (s==KeyWord[i]) {return i+1;break;}}return 0;
}
int IsId(string s) {//判断标识符中是否已经存在for (int i = 0; i < Id_num; i++) {if (s == Id[i]) {return i + 1;}}return 0;
}
int IsConst(string s) {//判断该常数是否已在常数表中for (int i = 0; i < Const_num; i++) {if (s == Const[i]) {return i + 1;break;}}return 0;
}
int IsBoundary(string s) {//判断是否为界符for (int i = 0; i < 9; i++) {if (s == Boundary[i]) {return  i + 1;}}return 0;
}
int IsOperator(string s) {//判断是否为界符for (int i = 0; i < 23; i++) {if (s == Operator[i]) {return  i + 1;}}return 0;
}
*/
int Judge(string s, string str[], int len) {//判断for (int i = 0; i < len; i++) {if (s == str[i]) {return  i + 1;}}return -1;
}void Analysis(string str) {//进行子程序分析ofstream outfile("Out.txt");int Len = str.length();string temp;//用来暂时存放当前单词int npc = -1;for (int i = 0; i < Len; i++) {if (IsLetter(str[i])) {    //第一个字符为单词npc = i;int f;for (int j = i + 1; j < Len; j++) {if (!IsLetter(str[j]) && !IsNumber(str[j])) {temp = str.substr(npc, j - npc);//相当于回退f = j;break;}}int flag = Judge(temp, KeyWord, 15);//查关键字表，flag记录内部编码if (flag != -1) {//是关键字outfile << temp << "\t关键字" << "\tt=1\t" << "内部编码" << flag << endl;cout << temp << "\t关键字" << "\tt=1\t" << "内部编码" << flag << endl;}else {//不是关键字int pc = Judge(temp, Id, Id_num);if (pc != -1) {   //标识符中已经有了outfile << temp << "\t标识符" << "\tt=4\t" << "内部编码" << pc << endl;cout << temp << "\t标识符" << "\tt=4\t" << "内部编码" << pc << endl;}else {   //标识符中还没有Id[Id_num++] = temp;outfile << temp << "\t标识符" << "\tt=4\t" << "内部编码" << Id_num << endl;cout << temp << "\t标识符" << "\tt=4\t" << "内部编码" << Id_num << endl;}}i = f - 1;}else if (IsNumber(str[i])) {//第一个字符为数字npc = i;string temp;int f;for (int j = i + 1; j < Len; j++) {if (!IsNumber(str[j])) {temp = str.substr(npc, j - npc);f = j;break;}}int flag = Judge(temp, Const, Const_num);if (flag != -1) {outfile << temp << "\t常数" << "\tt=5\t" << "内部编码" << flag << endl;cout << temp << "\t常数" << "\tt=5\t" << "内部编码" << flag << endl;}else {Const[Const_num++] = temp;outfile << temp << "\t常数" << "\tt=5\t" << "内部编码" << Const_num << endl;cout << temp << "\t常数" << "\tt=5\t" << "内部编码" << Const_num << endl;}i = f - 1;}else {//第一个字符不是字母或数字if (str[i] == ' ') {continue;}else {npc = i;int f;string s;s += str[i];if (Judge(s, Boundary, 9) != -1) {//判断是否为界符outfile << s << "\t界符" << "\tt=3\t" << "内部编码" << Judge(s, Boundary, 9) << endl;cout << s << "\t界符" << "\tt=3\t" << "内部编码" << Judge(s, Boundary, 9) << endl;continue;}for (int j = i + 1; j < Len; j++) {string pt;pt += str[j];if (Judge(pt, Boundary, 9) != -1) {  //第一个不是界符，当前是界符  类似 c++;}f = j;temp = str.substr(npc, j - npc);break;}else if (IsLetter(str[j]) || IsNumber(str[j]) || str[j] == ' ') {  //类似   +atemp = str.substr(npc, j - npc);f = j;break;}}int flag = Judge(temp, Boundary, 9);if (flag != -1) {   //是界符outfile << temp << "\t界符" << "\tt=3\t" << "内部编码" << flag << endl;cout << temp << "\t界符" << "\tt=3\t" << "内部编码" << flag << endl;}else {flag = Judge(temp, Operator, 23);if (flag != -1) {//是运算符outfile << temp << "\t运算符" << "\tt=2\t" << "内部编码" << flag << endl;cout << temp << "\t运算符" << "\tt=2\t" << "内部编码" << flag << endl;}else {//既不是界符也不是运算符outfile << temp << "\t非法符号" << endl;cout << temp << "\t非法符号" << endl;}}i = f - 1;}}}cout << endl << endl << "新加入的标识符：";outfile << endl << endl << "新加入的标识符：";for (int i = 0; i < Id_num; i++) {cout << Id[i] << " ";outfile << Id[i] << " ";}cout << endl << endl << "新加入的常数：";outfile << endl << endl << "新加入的常数：";for (int i = 0; i < Const_num; i++) {cout << Const[i] << " ";outfile << Const[i] << " ";}outfile.close();
}int main()
{string BefSource[20];string AftSource;ifstream infile("Source.txt");if (!infile) cout << "error" << endl;int Len = 0;while (infile.good()) {getline(infile, BefSource[Len++]);}infile.close();cout << "从文件中读取到的程序：" << endl;for (int i = 0; i < Len; i++) {cout << BefSource[i] << endl;}cout << endl << endl;Pretreatment(BefSource, Len, AftSource);cout << endl << "预处理后的程序：" << endl;cout << AftSource << endl << endl << endl;Analysis(AftSource);
}

实验中会用到的文件

这里的Out.txt文件可以为空文件，因为它是输出结果的文件。Source.txt需要存入进行分析的源代码。

运行结果