CInvertedFileChunk Class Reference

#include <CInvertedFileChunk.h>

Inheritance diagram for CInvertedFileChunk:

List of all members.

Public Member Functions
double	getCollectionFrequency (size_t inSize) const
void	addElement (TID inDocumentID, double inDocumentFrequency)
bool	writeBinary (ostream &, TID inFeatureID, size_t inSize) const

---

Detailed Description

A class which does all the necessary administrative work for generating inverted files. It contains the list of documents which contain a given feature

Definition at line 59 of file CInvertedFileChunk.h.

---

Member Function Documentation

double CInvertedFileChunk::getCollectionFrequency

(

size_t

inSize

)

const

returns the percentage of documents which contain this one feature

Definition at line 79 of file CInvertedFileChunk.cc.

Referenced by CAcIFFileSystem::newGenerateInvertedFile().

                                                                   {
  return double(size())/inSize;
}

void CInvertedFileChunk::addElement	(	TID	inDocumentID,
		double	inDocumentFrequency
	)

add one document of to the list for one feature

Definition at line 58 of file CInvertedFileChunk.cc.

Referenced by CAcIFFileSystem::newGenerateInvertedFile().

                                       {

  //cout << "a" << flush;

  insert(make_pair(inDocumentID,inDocumentFrequency));
};

bool CInvertedFileChunk::writeBinary	(	ostream &	outStream,
		TID	inFeatureID,
		size_t	inSize
	)			const

writes the binary part of the list for one feature

Definition at line 96 of file CInvertedFileChunk.cc.

References CDocumentFrequencyList::begin(), CDocumentFrequencyList::end(), and CDocumentFrequencyList::writeBinary().

Referenced by CAcIFFileSystem::newGenerateInvertedFile().

                                {
  
  bool lRetVal=true;

  {
    TID lFeatureID(inFeatureID);
    lRetVal = lRetVal && outStream.write((char*)&lFeatureID,
           sizeof(lFeatureID));
  }
  
  {
    float lCollectionFrequency=(float(size())
        /inSize);
    if((lCollectionFrequency >1)
       ||
       (lCollectionFrequency < 0)){
      cout << "The collection frequency is: "
     << lCollectionFrequency
     << endl;
      assert(!"collection frequency out of range");
    }
    lRetVal = lRetVal && outStream.write((char*)&lCollectionFrequency,
        sizeof(lCollectionFrequency));
  }
  
  int lSize(size());
  {
    // writing this twice means: there is
    // no space left in this list
    lRetVal = lRetVal && outStream.write((char*)&lSize,
           sizeof(lSize));
    // writing this twice means: there is
    // no space left in this list
    lRetVal = lRetVal && outStream.write((char*)&lSize,
           sizeof(lSize));
    cout << "writing chunk of size " << lSize <<endl;
  }
  
  
  CDocumentFrequencyList lList(lSize);
  
#ifndef _CDocumentFrequencyListIsList
  CDocumentFrequencyList::iterator j=lList.begin();
#endif
  
  for(const_iterator i=begin();
      i!=end();
      i++){
#ifndef _CDocumentFrequencyListIsList
    *(j++)=CDocumentFrequencyElement((*i).first,
             (*i).second);
#else
    lList.push_back(CDocumentFrequencyElement((*i).first,
                (*i).second)
        );
#endif
  }
#ifdef _CDocumentFrequencyListIsArray
  //lList.setEnd(j);// superfluous, because lList.begin()+lSize==j anyway
  assert(j==lList.begin()+lSize);
#endif
  
#ifndef _CDocumentFrequencyListIsList
  sort(lList.begin(),
       lList.end(),
       CSortByDocumentFrequency_DFE());
#else
  
  lList.sort(CSortByDocumentFrequency_DFE());
#endif
  
  for(CDocumentFrequencyList::const_iterator i=lList.begin();
      i!=lList.end();
      i++){
    lRetVal=lRetVal && i->writeBinary(outStream);
  }
  
  return lRetVal;
}

---

The documentation for this class was generated from the following files:

• gift/libGIFTAcInvertedFile/include/CInvertedFileChunk.h
• gift/libGIFTAcInvertedFile/cc/CInvertedFileChunk.cc