Lexesis/include/Lexesis/re.h

#pragma once
#ifndef RE_H
#define RE_H

#include "Lexesis/automata.h"

#include <memory>
#include <stdexcept>

namespace lxs {
    /**
     * An abstract regular expression
     */
    class RE
    {
        public:
            virtual ~RE() {}
            /**
             * Convert this regex to an ENFA
             * This extends the given enfa, and attaches itself to the given `attach` state
             */
            virtual State toENFA(ENFA& enfa, State attach) = 0;

            /**
             * Convert this regex to a text regex (useful to check if parsing was correct)
             */
            virtual std::string toRe() = 0;
    };

    /**
     * A regex for the empty language
     */
    class EmptyRE : public RE
    {
        public:
            EmptyRE() {}
            ~EmptyRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();
    };

    /**
     * A regex for the language containing only the empty string
     */
    class EpsilonRE : public RE
    {
        public:
            EpsilonRE() {}
            ~EpsilonRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();
    };

    /**
     * A regex for the language containing a single character
     */
    class SingleRE : public RE
    {
        public:
            /**
             * @param c The character of the language
             */
            SingleRE(char c) : c(c) {}
            ~SingleRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();

            char c;
    };

    /**
     * A regex for the language containing multiple single-symbol strings
     */
    class MultiRE : public RE
    {
        public:
            /**
             * @param chars The list of symbols contained in the language
             */
            MultiRE(std::vector<char> chars) : chars(std::move(chars)) {}
            ~MultiRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();

            std::vector<char> chars;
    };

    /**
     * A regex for the concatenation of two languages
     */
    class ConcatRE : public RE
    {
        public:
            /**
             * @param e The first language
             * @param f The second language
             */
            ConcatRE(std::shared_ptr<RE> e, std::shared_ptr<RE> f) : e(e), f(f) {}
            ~ConcatRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();

            std::shared_ptr<RE> e, f;
    };

    /**
     * The regex for the kleene star of a language
     */
    class StarRE : public RE
    {
        public:
            /**
             * @param e The language to apply the star to
             */
            StarRE(std::shared_ptr<RE> e) : e(e) {}
            ~StarRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();

            std::shared_ptr<RE> e;
    };

    /**
     * A regex for the sum/disjunction of two languages
     */
    class PlusRE : public RE
    {
        public:
            /**
             * @param e The first language
             * @param f The second language
             */
            PlusRE(std::shared_ptr<RE> e, std::shared_ptr<RE> f) : e(e), f(f) {}
            ~PlusRE() {}
            virtual State toENFA(ENFA& enfa, State attach);
            virtual std::string toRe();

            std::shared_ptr<RE> e, f;
    };


} //namespace lxs

#endif //RE_H