C++ MCollect::push_back方法代码示例

// Генерация слов, фонетически близких к заданному word.
// Возвращается список вариантов, включая исходное слово, и список их достоверностей.
void LexicalAutomat::ProducePhonInv(
    const lem::UCString &word,
    int id_language,
    lem::MCollect<lem::UCString> &res,
    lem::MCollect<lem::Real1> &rels,
    LA_RecognitionTrace *trace
)
{
    MCollect<LA_AA_list*> packs;
    LA_AA_list *list = new LA_AA_list;
    list->reserve(16);
    list->push_back(LA_AA_item(word, Real1(100)));

    // Теперь мутированные варианты.
    LA_Pack *pack = AlephAuto(id_language, word, 1, trace);

    for (Container::size_type j = 0; j < pack->size(); j++)
    {
        const Solarix::Lexem &ph_lex = *(pack->get(j));
        if (res.find(ph_lex) == UNKNOWN)
        {
            Real1 r = pack->get(j)->get_Val();
            rels.push_back(r);
            res.push_back(ph_lex);
        }
    }

    return;
}

void LemmatizatorStorage_SQLITE::Lemmatize(
                                           const lem::UCString &word,
                                           lem::MCollect<lem::UCString> &lemmas
                                          )
{
 lemmas.clear();

 lem::MemFormatter mem;
 mem.printf( "SELECT L.lemma"
             " FROM lexemes_n X, lemmas L"
             " WHERE X.lexeme='%us' AND L.id=X.id_lemma", to_upper(word).c_str() );

 lem::Ptr<LS_ResultSet> rs(cnx->Select(lem::to_utf8(mem.string())));
 while( rs->Fetch() )
  {
   lemmas.push_back( rs->GetUCString(0) );
  }

 if( lemmas.empty() )
  {
   lemmas.push_back(word);
  }

 return;
}

bool LA_PreprocessorRules::Crop(
    const lem::UCString &word,
    lem::MCollect<lem::UCString> &results,
    lem::MCollect<lem::Real1> &rels,
    LA_RecognitionTrace *trace
) const
{
    bool applied = false;

    if (!crop_rules.empty())
    {
        // сначала применяем префиксные правила
        typedef CROP_RULES::const_iterator IT;

        LA_CropRule::HashType prefix_hash = LA_CropRule::CalcHash(word.c_str(), true, false);
        std::pair<IT, IT> pp = prefix_crop_rules.equal_range(prefix_hash);

        lem::UCString result;

        for (auto it = pp.first; it != pp.second; ++it)
        {
            const LA_CropRule *r = it->second;
            if (r->Apply(word, result))
            {
                applied = true;
                results.push_back(result);
                rels.push_back(r->GetRel());
                if (trace != nullptr)
                {
                    trace->CropRuleApplied(word, result, r);
                }
            }
        }

        // теперь отсекаем аффикс

        LA_CropRule::HashType affix_hash = LA_CropRule::CalcHash(word.c_str(), false, true);
        pp = affix_crop_rules.equal_range(affix_hash);

        for (auto it = pp.first; it != pp.second; ++it)
        {
            const LA_CropRule *r = it->second;
            if (r->Apply(word, result))
            {
                applied = true;
                results.push_back(result);
                rels.push_back(r->GetRel());
                if (trace != nullptr)
                {
                    trace->CropRuleApplied(word, result, r);
                }
            }
        }
    }

    return applied;
}

void LEMM_Compiler::LoadNGram( lem::Iridium::Macro_Parser & txtfile, Dictionary & dict, lem::MCollect<int> & terms, int order ) const
{
 lem::Iridium::BSourceState beg = txtfile.tellp();

 while( !txtfile.eof() )
 {
  lem::Iridium::BethToken t = txtfile.read();
  if( lem::is_int(t.string()) )
   terms.push_back( lem::to_int(t.string()) );
  else
   {
    txtfile.seekp(t);
    break;
   }
 }

 if( terms.size() != order+1 )
  {
   dict.GetIO().merr().printf( "%vfDInvalid ngram%vn\n" );
   lem::Iridium::Print_Error( beg, txtfile );
   throw lem::E_ParserError();
  }

 return;
}

void Lemmatizator::Lemmatize( const lem::MCollect<lem::UCString> & words, lem::MCollect<lem::UCString> &lemmas )
{
 #if defined LEM_THREADS
 lem::Process::CritSecLocker lock(&cs);
 #endif

 if( !model_loaded )
  {
   bin->seekp( model_pos );
   model_loaded = true;
   model_available = bin->read_bool();
   if( model_available )
    {
     LoadModel();
    }
  }


 if( model_available )
  {
   LemmatizeViaModel( words, lemmas );
  }
 else
  {
   for( lem::Container::size_type i=0; i<words.size(); ++i )
    {
     lem::UCString lemma;
     Lemmatize( words[i], lemma );
     lemmas.push_back( lemma );
    }
  }

 return;
}

void LexerTextPos::Collect_Right2Left(int count, lem::MCollect<const LexerTextPos*> & inverted_path) const
{
    inverted_path.push_back(this);
    if (count > 0 && !IsBegin() && previous != nullptr)
        GetPrev()->Collect_Right2Left(count - 1, inverted_path);

    return;
}

void SynPatternResult::GetExportCoordPairs(lem::MCollect< std::pair<int, int> > & pairs) const
{
    for (auto it = exported_coords.begin(); it != exported_coords.end(); ++it)
    {
        pairs.push_back(*it);
    }

    return;
}

void LexerTextPos::CollectPathToLeft(int count, lem::MCollect<const Word_Form*> & org) const
{
    LEM_CHECKIT_Z(count >= 0);

    org.push_back(wordform);

    if (count > 0 && previous != NULL)
        previous->CollectPathToLeft(count - 1, org);

    return;
}

void LexerTextPos::Collect_Right2Left(const LexerTextPos *left_boundary, lem::MCollect<const LexerTextPos*> & inverted_path) const
{
    LEM_CHECKIT_Z(left_boundary != nullptr);

    inverted_path.push_back(this);

    if (this != left_boundary && previous != nullptr)
        previous->Collect_Right2Left(left_boundary, inverted_path);

    return;
}

void SynPatterns::GetUnresolvedForwardDeclarations( lem::MCollect<lem::UCString> & unresolved_names ) const
{
 typedef std::map< lem::UCString, int >::const_iterator IT;
 for( IT it=name2id.begin(); it!=name2id.end(); ++it )
  {
   if( id2count.find( it->second )==id2count.end() )
    unresolved_names.push_back( it->first );
  }

 return;
}

void SyllabContext::GetResultSyllabs(lem::MCollect<lem::UCString> & result_syllabs, bool Normalized) const
{
    for (auto point : points)
    {
        if (point->IsLeftBoundary() || point->IsRightBoundary())
            continue;

        result_syllabs.push_back(point->BuildSyllab(Normalized));
    }

    return;
}

void SyllabContext::GetResultSyllabs( lem::MCollect<lem::UCString> & result_syllabs, bool Normalized ) const
{
 for( lem::Container::size_type i=0; i<points.size(); ++i )
  {
   const SyllabContextPoint * p = points[i];
   if( p->IsLeftBoundary() || p->IsRightBoundary() )
    continue;
   
   result_syllabs.push_back( p->BuildSyllab(Normalized) );
  }

 return;
}

// *************************************************************************************
// Ищем парадигмы, чьи условия подходят для указанной базовой формы, возвращает
// список id таких парадигм.
// *************************************************************************************
void ParadigmaFinder::Find( int PartOfSpeech, const lem::UCString &entry_name, lem::MCollect<int> &found_ids )
{
 #if defined LEM_THREADS
 lem::Process::RWU_ReaderGuard rlock(cs);
 #endif

 if( !loaded )
  {
   #if defined LEM_THREADS
   lem::Process::RWU_WriterGuard wlock(rlock);
   #endif
   LoadFromDB();
  }

 if( PartOfSpeech==UNKNOWN || PartOfSpeech==ANY_STATE )
  {
   for( lem::Container::size_type i=0; i<matchers.size(); ++i )
    if( matchers[i]->Match(entry_name) )
     {
      found_ids.push_back(ids[i]);
     }
  }
 else
  {
   CLASS2DECL::const_iterator it=class2decl.find(PartOfSpeech);
   if( it!=class2decl.end() )
    {
     for( lem::Container::size_type i=0; i<it->second->size(); ++i )
      if( it->second->get(i).second->Match(entry_name) )
       {
        found_ids.push_back( it->second->get(i).first );
       }
    }
  }
 
 return;
}

void NGramsStorage_SQLITE::SelectInts( const lem::FString &Select, lem::MCollect<int> &list )
{
 LEM_CHECKIT_Z( !Select.empty() );

 sqlite3_stmt *stmt=NULL;
 int res = sqlite3_prepare_v2( hdb, Select.c_str(), -1, &stmt, NULL );
 if( res==SQLITE_OK )
  {
   while( sqlite3_step( stmt ) == SQLITE_ROW )
    {
     int i = sqlite3_column_int(stmt,0);
     list.push_back(i);
    } 

   sqlite3_finalize(stmt);
  }

 return;
}

void SG_DeclensionTable::GenerateForms(
                                       const Lexem &entry_name,
                                       lem::MCollect<Lexem> & res,
                                       lem::PtrCollect<CP_Array> & form_dims, 
                                       const SynGram &sg,
                                       const SG_DeclensionAutomat &dsa 
                                      ) const
{
 res.reserve(form.size());

 for( lem::Container::size_type i=0; i<form.size(); i++ )
  {
   UCString frm( dsa.ProduceForm( entry_name, GetClass(), *form[i], sg ) );

   res.push_back( frm);
   form_dims.push_back( new CP_Array( form[i]->GetDim() ) );
  }

 return;
}