SyntacticAnalyzer-nochains.h

Go to the documentation of this file.

// Copyright 2002-2013 CEA LIST
// SPDX-FileCopyrightText: 2022 CEA LIST <gael.de-chalendar@cea.fr>
//
// SPDX-License-Identifier: MIT
 
#ifndef SYNTACTICANALYZERNOCHAINS_H
#define SYNTACTICANALYZERNOCHAINS_H
 
#include "SyntacticAnalysisExport.h"
#include "SyntacticData.h"
#include "common/MediaProcessors/MediaProcessUnit.h"
 
#include <string>
#include <stack>
 
namespace Lima
{
namespace LinguisticProcessing
{
namespace SyntacticAnalysis
{
 
#define SACLOGINIT LOGINIT("LP::SyntacticAnalysis::Chains")
#ifdef WIN32
#undef min
#undef max
#endif
 
#define SYNTACTICANALYZERCHAINS_FIRSTID 1
#define SYNTACTICANALYZERNOCHAINS_CLASSID "SyntacticAnalyzerNoChains"
 
class LIMA_SYNTACTICANALYSIS_EXPORT SyntacticAnalyzerNoChains : public MediaProcessUnit
{
public:
  
  SyntacticAnalyzerNoChains();
 
  //  SyntacticAnalyzer(SyntacticAnalyzer&);
 
  virtual ~SyntacticAnalyzerNoChains() {}
 
  void init(
    Common::XMLConfigurationFiles::GroupConfigurationStructure& unitConfiguration,
    Manager* manager) override;
 
  LimaStatusCode process(AnalysisContent& analysis) const override;
 
  void identifyChains(SyntacticData* data,
                      const LinguisticGraphVertex& s,
                      const LinguisticGraphVertex& t,
                      uint64_t& startChainId) const;
 
 
  MediaId language() const;
 
private:
 
  static const uint64_t m_firstChainId = SYNTACTICANALYZERCHAINS_FIRSTID;
  
  class NoChainEndInStackException : public std::exception {};
  
  typedef boost::tuple< LinguisticGraphVertex, bool, std::vector< LinguisticGraphVertex > > ChainStackTuple;
 
  void reportChainInGraph(
                          SyntacticData* data,
                          const std::vector< ChainStackTuple >& pile,
                          Common::MediaticData::ChainsType type,
                          std::set< LinguisticGraphVertex >& alreadyFinished,
                          uint64_t& chainId,
                          const LinguisticGraphVertex& stop) const;
 
  bool parentsFinished(
                        const SyntacticData* data,
                        const LinguisticGraphVertex& v,
                        const std::set< LinguisticGraphVertex >& alreadyFinished) const;
  
  std::string stringChain(
                          const SyntacticData* data,
                          const std::vector< ChainStackTuple >& pile,
                          Common::MediaticData::ChainsType type,
                          std::set< LinguisticGraphVertex >& alreadyFinished,
                          uint64_t chainId,
                          const LinguisticGraphVertex& stop = std::numeric_limits<LinguisticGraphVertex>::max()) const;
 
  LinguisticGraphVertex unstackUptoChainEnd(
                                          const SyntacticData* data,
                                          std::vector< ChainStackTuple >& pile,
                                          Common::MediaticData::ChainsType type
                                        ) const;
 
protected:
 
  MediaId m_language;
 
  std::shared_ptr<SyntagmDefStruct> m_chainMatrix;
 
  uint64_t m_maxChainsNbByVertex;
  
  uint64_t m_maxChainLength;
 
  const Common::PropertyCode::PropertyAccessor* m_macroAccessor;
  const Common::PropertyCode::PropertyAccessor* m_microAccessor;
  LinguisticCode m_ponctuCategory;
};
 
inline MediaId SyntacticAnalyzerNoChains::language() const
{
  return m_language;
}
 
} // closing namespace SyntacticAnalysis
} // closing namespace LinguisticProcessing
} // closing namespace Lima
 
#endif