Analizador Lexico

AnalizadorLexico.cpp

#include <iostream>
#include <cstring>
#include <algorithm>
#include <vector>
#include <stack>
#include <map>
#include <fstream>

using namespace std;

const map<int, string> TOKENS = {
    {1, "IDENTIFICADOR"},
    {2, "ENTERO"},
    {4, "ASIGNACION"},
    {5, "SUMA"},
    {6, "MULTIPLICACION"},
    {7, "PARENTESIS_IZQ"},
    {8, "PARENTESIS_DER"},
    {9, "PUNTO_COMA"},
    {10, "ERROR"},
    {11, "FIN"}
};

// Count final states
const unsigned int NUM_FS = 10;

const unsigned int MATRIX[12][11] = {
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		1,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {10,    10,		10,		4,		10,		10,		10,		10,		10,		10,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    {1,		2,		3,		10,		5,		6,		7,		8,		9,		11,		10},
    /*{,		,		,		,		,		,		,		,		,		,		}*/
};

const unsigned int INITIAL_STATE = 0;

const unsigned int FINAL_STATES[NUM_FS] = {
    1,  2,  4,  5,  6,  7,  8,  9,  10, 11
};

template <typename T>
void print(T const& v)
{
    typename T::const_iterator it;

    for (it = v.begin(); it != v.end(); ++it)
    {
        std::cout << *it << ' ';
    }

    std::cout << std::endl;
}

// Delete spaces, tabs, EOF
string parseSource(string str)
{
    str.erase(std::remove(str.begin(), str.end(), ' '), str.end());
    str.erase(std::remove(str.begin(), str.end(), '\t'), str.end());
    str.erase(std::remove(str.begin(), str.end(), '\n'), str.end());
    str.insert(str.end(), '$');

    return str;
}

string readFile(string path)
{
    string line;
    string sourceCode = "";
    ifstream myfile (path.c_str());

    if (myfile.is_open()) {

        cout << "Leyendo archivo..." << endl << endl;

        while ( getline (myfile,line) ) {
            cout << line << '\n';
            sourceCode += line;
        }
        myfile.close();
    }
    else {
        cout << "Archivo no encontrado." << endl;
        exit(1);
    }

    return parseSource(sourceCode);
}

bool isFinalState(unsigned int state)
{
    for (unsigned int i = 0; i < NUM_FS; i++) {
        if (FINAL_STATES[i] == state) {
            return true;
        }
    }

    return false;
}

vector<unsigned int> automata(vector<unsigned int> w)
{
    unsigned int         q, s;
    vector<unsigned int> tokens;

    q = INITIAL_STATE;

    for(int i = 0; i < w.size(); i++){
        s = w.at(i);
        q = MATRIX[q][s];

        if (isFinalState(q)) {
            if (tokens.empty()) {
                tokens.push_back(q);
            } else if (q != tokens.back()) {
                tokens.push_back(q);
            }
        }

    }

    return tokens;
}

vector<unsigned int> charsToIndexes(string word)
{
    vector<unsigned int> wordIndexes;

    for (unsigned int i = 0; i < word.size(); i++) {
        char c = word.at(i);

        if (isalpha(c)) {
            wordIndexes.push_back(0);
        } else if (isdigit(c)) {
            wordIndexes.push_back(1);
        } else if (c == ':') {
            wordIndexes.push_back(2);
        } else if (c == '=') {
            wordIndexes.push_back(3);
        } else if (c == '+') {
            wordIndexes.push_back(4);
        } else if (c == '*') {
            wordIndexes.push_back(5);
        } else if (c == '(') {
            wordIndexes.push_back(6);
        } else if (c == ')') {
            wordIndexes.push_back(7);
        } else if (c == ';') {
            wordIndexes.push_back(8);
        } else if (c == '$') {
            wordIndexes.push_back(9);
        } else {
            wordIndexes.push_back(10);
        }
    }

    return wordIndexes;
}

int main()
{
    string code = readFile("source.txt");
    vector<unsigned int> tokens;

    tokens = automata(charsToIndexes(code));

    cout << endl  << "Tokens generados..." << endl << endl;

    for(unsigned int i = 0; i < tokens.size(); i++) {
        cout << TOKENS.at(tokens.at(i)) << endl;
    }

    return 0;
}

source.txt

AA := (BB + 5) * CC;
DD =: 458##;