CompressedBitParallelIndex.h

Go to the documentation of this file.

 
/*
 * This file is part of Algorithms library toolkit.
 * Copyright (C) 2017 Jan Travnicek (jan.travnicek@fit.cvut.cz)
 
 * Algorithms library toolkit is free software: you can redistribute it and/or modify
 * it under the terms of the GNU General Public License as published by
 * the Free Software Foundation, either version 3 of the License, or
 * (at your option) any later version.
 
 * Algorithms library toolkit is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 
 * You should have received a copy of the GNU General Public License
 * along with Algorithms library toolkit.  If not, see <http://www.gnu.org/licenses/>.
 */
 
#pragma once
 
#include <ext/iostream>
 
#include <alib/set>
#include <alib/string>
 
#include <common/DefaultSymbolType.h>
 
#include <core/components.hpp>
#include <exception/CommonException.h>
 
#include <common/SparseBoolVector.hpp>
 
#include <core/normalize.hpp>
#include <alphabet/common/SymbolNormalize.h>
 
namespace indexes {
 
namespace stringology {
 
class GeneralAlphabet;
 
template < class SymbolType = DefaultSymbolType >
class CompressedBitParallelIndex final : public core::Components < CompressedBitParallelIndex < SymbolType >, ext::set < SymbolType >, component::Set, GeneralAlphabet > {
    ext::map < SymbolType, common::SparseBoolVector > m_vectors;
 
public:
    explicit CompressedBitParallelIndex ( ext::set < SymbolType > alphabet, ext::map < SymbolType, common::SparseBoolVector > vectors );
 
    const ext::map < SymbolType, common::SparseBoolVector > & getData ( ) const &;
 
    ext::map < SymbolType, common::SparseBoolVector > && getData ( ) &&;
 
    ext::vector < SymbolType > getString ( ) const;
 
    const ext::set < SymbolType > & getAlphabet ( ) const & {
        return this->template accessComponent < GeneralAlphabet > ( ).get ( );
    }
 
    ext::set < SymbolType > && getAlphabet ( ) && {
        return std::move ( this->template accessComponent < GeneralAlphabet > ( ).get ( ) );
    }
 
    void setCompressedBitVectorForSymbol ( SymbolType symbol, common::SparseBoolVector data );
 
    bool removeSymbolFromAlphabet ( const SymbolType & symbol ) {
        return this->template accessComponent < GeneralAlphabet > ( ).remove ( symbol );
    }
 
    auto operator <=> ( const CompressedBitParallelIndex & other ) const {
        return std::tie ( getData ( ), getAlphabet ( ) ) <=> std::tie ( other.getData ( ), other.getAlphabet ( ) );
    }
 
    auto operator == ( const CompressedBitParallelIndex & other ) const {
        return std::tie ( getData ( ), getAlphabet ( ) ) == std::tie ( other.getData ( ), other.getAlphabet ( ) );
    }
 
    friend ext::ostream & operator << ( ext::ostream & out, const CompressedBitParallelIndex & instance ) {
        return out << "(CompressedBitParallelIndex " << instance.m_vectors << ")";
    }
};
 
} /* namespace stringology */
 
} /* namespace indexes */
 
namespace indexes {
 
namespace stringology {
 
template < class SymbolType >
CompressedBitParallelIndex < SymbolType >::CompressedBitParallelIndex ( ext::set < SymbolType > alphabet, ext::map < SymbolType, common::SparseBoolVector > vectors ) : core::Components < CompressedBitParallelIndex, ext::set < SymbolType >, component::Set, GeneralAlphabet > ( std::move ( alphabet ) ), m_vectors ( std::move ( vectors ) ) {
}
 
template < class SymbolType >
const ext::map < SymbolType, common::SparseBoolVector > & CompressedBitParallelIndex < SymbolType >::getData ( ) const & {
    return m_vectors;
}
 
template < class SymbolType >
ext::map < SymbolType, common::SparseBoolVector > && CompressedBitParallelIndex < SymbolType >::getData ( ) && {
    return std::move ( m_vectors );
}
 
template < class SymbolType >
ext::vector < SymbolType > CompressedBitParallelIndex < SymbolType >::getString ( ) const {
    ext::vector < SymbolType > res;
 
    unsigned index = 0;
    do {
        for ( const std::pair < const SymbolType, common::SparseBoolVector > & compressedBitVector : m_vectors )
            if ( compressedBitVector.second.size ( ) > index && compressedBitVector.second [ index ] ) {
                res.push_back ( compressedBitVector.first );
                continue;
            }
 
    } while ( res.size ( ) == index ++ + 1 );
 
    return res;
}
 
template < class SymbolType >
void CompressedBitParallelIndex < SymbolType >::setCompressedBitVectorForSymbol ( SymbolType symbol, common::SparseBoolVector data ) {
    this->m_vectors [ symbol ] = std::move ( data );
}
 
} /* namespace stringology */
 
} /* namespace indexes */
 
namespace core {
 
template < class SymbolType >
class SetConstraint < indexes::stringology::CompressedBitParallelIndex < SymbolType >, SymbolType, indexes::stringology::GeneralAlphabet > {
public:
    static bool used ( const indexes::stringology::CompressedBitParallelIndex < SymbolType > & index, const SymbolType & symbol ) {
        const ext::map < SymbolType, common::SparseBoolVector > & content = index.getData ( );
        return content.find( symbol ) != content.end();
    }
 
    static bool available ( const indexes::stringology::CompressedBitParallelIndex < SymbolType > &, const SymbolType & ) {
        return true;
    }
 
    static void valid ( const indexes::stringology::CompressedBitParallelIndex < SymbolType > &, const SymbolType & ) {
    }
};
 
template < class SymbolType >
struct normalize < indexes::stringology::CompressedBitParallelIndex < SymbolType > > {
    static indexes::stringology::CompressedBitParallelIndex < > eval ( indexes::stringology::CompressedBitParallelIndex < SymbolType > && value ) {
        ext::set < DefaultSymbolType > alphabet = alphabet::SymbolNormalize::normalizeAlphabet ( std::move ( value ).getAlphabet ( ) );
 
        ext::map < DefaultSymbolType, common::SparseBoolVector > vectors;
        for ( std::pair < SymbolType, common::SparseBoolVector > && vector : ext::make_mover ( std::move ( value ).getData ( ) ) )
            vectors.insert ( std::make_pair ( alphabet::SymbolNormalize::normalizeSymbol ( std::move ( vector.first ) ), std::move ( vector.second ) ) );
 
        return indexes::stringology::CompressedBitParallelIndex < > ( std::move ( alphabet ), std::move ( vectors ) );
    }
};
 
} /* namespace core */