Wie kann ich einfach nicht erkannte Zeichen verwenden?

Question

Dec 08, 2015, 07:52 PM

Wie kann ich einfach nicht erkannte Zeichen verwenden?

Ich habe es dank der Boost Spirit-Bibliothek geschafft, eine pgn-Datei zu analysieren, aber sie schlägt fehl, sobald es einige Zeichen gibt, die ich nicht "vorweggenommen" habe.

Hier ist meine Spirit Grammatik:

#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix.hpp>
#include <boost/fusion/include/adapt_struct.hpp>

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::pgn_tag,
    (std::string, key),
    (std::string, value)
)

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::game_move,
    (unsigned, move_number),
    (std::string, move_turn),
    (std::string, white_move),
    (std::string, black_move),
    (std::string, result)
)

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::pgn_game,
    (std::vector<loloof64::pgn_tag>, header),
    (std::vector<loloof64::game_move>, moves)
)

namespace loloof64 {
    namespace qi = boost::spirit::qi;
    namespace ascii = boost::spirit::ascii;
    namespace phoenix = boost::phoenix;

    template <typename Iterator>
    struct pgn_parser : qi::grammar<Iterator, std::vector<pgn_game>, qi::unused_type>
    {
        pgn_parser() : pgn_parser::base_type(games)
        {
            using qi::lexeme;
            using ascii::char_;
            using qi::uint_;
            using qi::alnum;
            using qi::space;
            using qi::omit;
            using qi::eol;
            using qi::lit;

            quoted_string %= lexeme[lit('"') >> *(char_ - '"') >> lit('"')];

            tag %=
                '['
                >> +alnum
                >> omit[+space]
                >> quoted_string
                >> ']'
                >> omit[+eol]
                ;

            header %= +tag;

            move_turn %= qi::string("...") | qi::string(".");

            regular_move %=
                +char_("a-hNBRQK")
                >> +char_("a-h1-8x=NBRQK")
                >> -qi::string("e.p.")
                ;
            castle_move %= qi::string("O-O-O") | qi::string("O-O");
            single_move %=
                (regular_move | castle_move) >> -(char_('+') | char_('#'))
                ;

            result %= qi::string("1-0") | qi::string("0-1") | qi::string("1/2-1/2") | qi::string("*");

            full_move %=
                uint_
                >> move_turn
                >> omit[*space]
                >> single_move
                >> -(omit[+space] >> single_move)
                >> -(omit[+space] >> result)
                ;

            game_description %= full_move
                >> *(omit[*space] >> full_move);

            single_game %=
                -header
                >> game_description
                ;

            games %=
                single_game
                >> *(omit[*(space|eol)] >> single_game)
                ;
        }

        qi::rule<Iterator, pgn_tag(), qi::unused_type> tag;
        qi::rule<Iterator, std::vector<pgn_tag>, qi::unused_type> header;
        qi::rule<Iterator, std::string(), qi::unused_type> quoted_string;

        qi::rule<Iterator, std::string(), qi::unused_type> result;
        qi::rule<Iterator, std::string(), qi::unused_type> regular_move;
        qi::rule<Iterator, std::string(), qi::unused_type> castle_move;
        qi::rule<Iterator, std::string(), qi::unused_type> single_move;
        qi::rule<Iterator, std::string(), qi::unused_type> move_turn;
        qi::rule<Iterator, game_move(), qi::unused_type> full_move;
        qi::rule<Iterator, std::vector<game_move>, qi::unused_type> game_description;

        qi::rule<Iterator, pgn_game, qi::unused_type> single_game;
        qi::rule<Iterator, std::vector<pgn_game>, qi::unused_type> games;
    };
}

Wie kann ich einfach einen Charakter konsumieren, den ich nicht "vorwegnehmen" kann? Ich meine, wie könnte ich ein Zeichen, das ich nicht will, in keiner meiner Grammatikregeln ignorieren?

Wie zu Testzwecken:

hier mein Parser-Header (pgn_games_extractor.hpp)

#ifndef PGNGAMESEXTRACTOR_HPP
#define PGNGAMESEXTRACTOR_HPP

#include <string>
#include <vector>
#include <fstream>
#include <stdexcept>

namespace loloof64 {

    struct pgn_tag {
        std::string key;
        std::string value;
    };

    struct game_move {
        unsigned move_number;
        std::string move_turn;
        std::string white_move;
        std::string black_move;
        std::string result;
    };

    struct pgn_game {
        std::vector<pgn_tag> header;
        std::vector<game_move> moves;
    };

    class PgnGamesExtractor
    {
        public:
            PgnGamesExtractor(std::string inputFilePath);
            PgnGamesExtractor(std::ifstream &inputFile);
            /*
            Both constructos may throw PgnParsingException (if bad pgn format) and InputFileException (if missing file)
            */
            std::vector<pgn_game> getGames() const { return games; }
            virtual ~PgnGamesExtractor();

        protected:

        private:
            std::vector<pgn_game> games;
            void parseInput(std::ifstream &inputFile);
    };

    class PgnParsingException : public std::runtime_error
    {
    public:
        PgnParsingException(std::string message):     std::runtime_error(message){}
    };

    class InputFileException : public std::runtime_error
    {
    public:
        InputFileException(std::string message) :     std::runtime_error(message){}
    };
}

#endif // PGNGAMESEXTRACTOR_HPP

Hier ist meine Parser-Quelle (pgn_games_extractor.cpp):

#include "pgn_games_extractor.hpp"

#include <boost/spirit/include/qi.hpp>
#include <boost/spirit/include/phoenix.hpp>
#include <boost/fusion/include/adapt_struct.hpp>

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::pgn_tag,
    (std::string, key),
    (std::string, value)
)

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::game_move,
    (unsigned, move_number),
    (std::string, move_turn),
    (std::string, white_move),
    (std::string, black_move),
    (std::string, result)
)

BOOST_FUSION_ADAPT_STRUCT(
    loloof64::pgn_game,
    (std::vector<loloof64::pgn_tag>, header),
    (std::vector<loloof64::game_move>, moves)
)

namespace loloof64 {
    namespace qi = boost::spirit::qi;
    namespace ascii = boost::spirit::ascii;
    namespace phoenix = boost::phoenix;

    template <typename Iterator>
    struct pgn_parser : qi::grammar<Iterator, std::vector<pgn_game>, qi::unused_type>
    {
        pgn_parser() : pgn_parser::base_type(games)
        {
            using qi::lexeme;
            using ascii::char_;
            using qi::uint_;
            using qi::alnum;
            using qi::space;
            using qi::omit;
            using qi::eol;
            using qi::lit;

            quoted_string %= lexeme[lit('"') >> *(char_ - '"') >> lit('"')];

            tag %=
                '['
                >> +alnum
                >> omit[+space]
                >> quoted_string
                >> ']'
                >> omit[+eol]
                ;

            header %= +tag;

            move_turn %= qi::string("...") | qi::string(".");

            regular_move %=
                +char_("a-hNBRQK")
                >> +char_("a-h1-8x=NBRQK")
                >> -qi::string("e.p.")
                ;
            castle_move %= qi::string("O-O-O") | qi::string("O-O");
            single_move %=
                (regular_move | castle_move) >> -(char_('+') | char_('#'))
                ;

            result %= qi::string("1-0") | qi::string("0-1") | qi::string("1/2-1/2") | qi::string("*");

            full_move %=
                uint_
                >> move_turn
                >> omit[*space]
                >> single_move
                >> -(omit[+space] >> single_move)
                >> -(omit[+space] >> result)
                ;

            game_description %= full_move
                >> *(omit[*space] >> full_move);

            single_game %=
                -header
                >> game_description
                ;

            games %=
                single_game
                >> *(omit[*(space|eol)] >> single_game)
                ;
        }

        qi::rule<Iterator, pgn_tag(), qi::unused_type> tag;
        qi::rule<Iterator, std::vector<pgn_tag>, qi::unused_type> header;
        qi::rule<Iterator, std::string(), qi::unused_type> quoted_string;

        qi::rule<Iterator, std::string(), qi::unused_type> result;
        qi::rule<Iterator, std::string(), qi::unused_type> regular_move;
        qi::rule<Iterator, std::string(), qi::unused_type> castle_move;
        qi::rule<Iterator, std::string(), qi::unused_type> single_move;
        qi::rule<Iterator, std::string(), qi::unused_type> move_turn;
        qi::rule<Iterator, game_move(), qi::unused_type> full_move;
        qi::rule<Iterator, std::vector<game_move>, qi::unused_type> game_description;

        qi::rule<Iterator, pgn_game, qi::unused_type> single_game;
        qi::rule<Iterator, std::vector<pgn_game>, qi::unused_type> games;
    };
}


loloof64::PgnGamesExtractor::PgnGamesExtractor(std::string inputFilePath)
{
    std::ifstream inputFile(inputFilePath);
    parseInput(inputFile);
}

loloof64::PgnGamesExtractor::PgnGamesExtractor(std::ifstream &inputFile)
{
    parseInput(inputFile);
}

loloof64::PgnGamesExtractor::~PgnGamesExtractor()
{
    //dtor
}

void loloof64::PgnGamesExtractor::parseInput(std::ifstream &inputFile)
{
    using namespace std;

    if (! inputFile) throw InputFileException("File does not exist !");

    string content("");
    getline(inputFile, content, (char) inputFile.eof());

    if (inputFile.fail() || inputFile.bad()) throw new     InputFileException("Could not read the input file !");

    loloof64::pgn_parser<string::const_iterator> parser;
    std::vector<loloof64::pgn_game> temp_games;

    string::const_iterator iter = content.begin();
    string::const_iterator end = content.end();

    bool success = boost::spirit::qi::phrase_parse(iter, end, parser, boost::spirit::qi::eol, temp_games);

    if (success && iter == end)
    {
        games = temp_games;
    }
    else
    {
        string error_fragment(iter, end);
        string error_message("");

        error_message = "Failed to parse the input at :'" + error_fragment + "' !";

        throw PgnParsingException(error_message);
    }
}

Ich stelle diese Frage, weil ich die folgende pgn nicht analysieren konnte: ScotchGambitPgn.zip. Ich denke, es liegt an einem Codierungsproblem mit dieser Datei.

Ich benutze Spirit 2 und C ++ 11 (Gnu)