【编译原理】- 手工构造词法分析器

【问题描述】通过设计c语言常见单词的正规文法或正规式，而后得到NFA，再确定化得到DFA，根据DFA的转换矩阵或转换图，用c++语言实现词法分析器。
【输入形式】输入一段完整的c语言程序
【输出形式】各类单词的token字
【样例输入】

int main(){
    int a = 10;
    double b = -20.9;
    if(a<=b)
    	a+=b;
    return a;
}

【样例输出】

line1:(type, int)

line1:(keyword, main)

line1:(bracket, ()

line1:(bracket, ))

line1:(bracket, {)

line2:(type, int)

line2:(identify, a)

line2:(OPT, =)

line2:(integer, 10)

line2:(bracket, ;)

line3:(type, double)

line3:(identify, b)

line3:(OPT, =)

line3:(decimal, -20.9)

line3:(bracket, ;)

line4:(keyword, if)

line4:(bracket, ()

line4:(identify, a)

line4:(OPT, <=)

line4:(identify, b)

line4:(bracket, ))

line5:(identify, a)

line5:(OPT, +=)

line5:(identify, b)

line5:(bracket, ;)

line6:(identify, a)

line6:(OPT, =)

line6:(integer, 0)

line6:(bracket, ;)

line7:(keyword, return)

line7:(identify, a)

line7:(bracket, ;)

line8:(bracket, })

【样例说明】需要识别的关键字包括main, return, if, else, do, while, for, scanf, printf, sqrt, abs；type类型包括void, int, double, float, char；运算符(算术、关系、逻辑、位)；需要识别的其他单词有标识符, 整数（十进制形式、指数形式），实数（十进制形式、指数形式），字符串（输出类型名为string）；过滤注释及空格。
【评分标准】根据设计文档的质量、lex文件的正确性，代码的正确性、代码的时间空间复杂度、识别单词的种类等综合评分

实现代码：

#include <algorithm>
#include <iostream>
#include <vector>
#include <cstring>

using namespace std;

vector<string> keyword = {"scanf", "printf", "if", "else", "for", "while", "return", "do", "main", "abs", "sqrt"};
vector<string> type = {"int", "void", "char", "double", "short", "float"};
vector<char> bracket = {',', '\\', ';', ':', '(', ')', '[', ']', '{', '}', '"', '\''};

struct Node
{
    int line = 0;
    string type;
    string word;
};

vector<Node> stack;
int line = 1;
char text[int(1e5)] = "";
char ch = ' ';
int len = 0;
int i = 0;
string word;
Node temp;

void makeword(string s)
{
    temp.line = line;
    temp.type = s;
    temp.word = word;
    stack.push_back(temp);
    word.clear();
}

void JudgeE()
{
    word += ch;
    ch = text[++i];
    if (ch == '+' || ch == '-')
    {
        word += ch;
        ch = text[++i];
    }
    if (ch >= '1' && ch <= '9')
    {
        word += ch;
        while ((ch = text[++i]) && (ch >= '0' && ch <= '9'))
            word += ch;
        makeword("float");
    }
    else
    {
        cout << "Error at Line " << line << ": Illegal floating point number \"" << word << "\".\n";
        exit(-1);
    }
}

void jump()
{
    while (ch == ' ' || ch == '\n')
    {
        if (ch == '\n')
            line++;
        ch = text[++i];
    }
}

void makenumber()
{
    while (ch >= '0' && ch <= '9')
    {
        word += ch;
        ch = text[++i];
    }
    if (ch == '.')
    {
        word += ch;
        ch = text[++i];
        while (ch >= '0' && ch <= '9')
        {
            word += ch;
            ch = text[++i];
        }
        if (ch == 'e')
            JudgeE();
        else
            makeword("decimal");
    }
    else if (ch == 'e')
        JudgeE();
    else
        makeword("integer");
}

void fun()
{
    ch = text[i];
    jump();
    while (ch != '\0' && ch != EOF)
    {
        jump();
        if (ch == '+' || ch == '-')
        {
            word += ch;
            ch = text[++i];
            if (ch == '=')
            {
                word += ch;
                makeword("OPT");
                ch = text[++i];
            }
            else
                makenumber();
        }
        if (ch >= '0' && ch <= '9')
        {
            word += ch;
            ch = text[++i];
            makenumber();
        }
        if (isalnum(ch) || ch == '_')
        {
            while (isalnum(ch) || ch == '_')
            {
                word += ch;
                ch = text[++i];
            }
            if (find(keyword.begin(), keyword.end(), word) != keyword.end())
                makeword("keyword");
            else if (find(type.begin(), type.end(), word) != type.end())
                makeword("type");
            else
                makeword("identify");
        }
        if (ch == '/')
        {
            char temp = text[++i];
            if (temp == '/')
                do
                {
                    ch = text[++i];
                } while (ch != '\n');
            else if (temp == '*')
            {
                do
                {
                    ch = text[++i];
                } while (ch != '/');
                ch = text[++i];
            }
            else if (temp == '=')
            {
                word += "/=";
                makeword("OPT");
                ch = text[++i];
            }
            else
            {
                word += '/';
                makeword("OPT");
                ch = text[++i];
            }
        }
        if (ch == '0')
        {
            word += ch;
            ch = text[++i];
            makeword("integer");
        }
        if (ch == '%' || ch == '&')
        {
            if (isalnum(text[i+1]))
            {
                word += ch;
                word += text[++i];
                ch = text[++i];
                makeword("typeidentify");
            }
        }
        if (find(bracket.begin(), bracket.end(), ch) != bracket.end())
        {
            word += ch;
            ch = text[++i];
            makeword("bracket");
        }
        if (ch == '*' || ch == '=' || ch == '<' || ch == '>' || ch == '!')
        {
            word += ch;
            if (text[i + 1] == '=')
                word += text[++i];
            ch = text[++i];
            makeword("OPT");
        }
    }
}

int main()
{
    text[len] = getchar();
    while (text[len] != '\0' && text[len] != EOF)
    {
        text[++len] = getchar();
    }
    fun();
    for (vector<Node>::iterator iter = stack.begin(); iter != stack.end(); iter++)
        cout << "line" << (*iter).line << ":(" << (*iter).type << ", " << (*iter).word << ')' << endl;
}