ByteString DataDocumentManager::getDocument(int ID)
{
ByteString recordString;
ByteString byteString;
byteString.insertLast(&ID,sizeof(int));
string idString = byteString.toString();
Key key(idString);
pair<Record*, BPlusTreeIterator*> searchResult = this->principalIndex->search(key);
if(searchResult.first != NULL)
{
int position = searchResult.first->getData()->readAsInt(0);
if(position > -1)
{
recordString = this->fileVariableRecord->readDocument(position);
}
//delete poddo
delete searchResult.first;
}
//delete poddo
if (searchResult.second != NULL)
delete searchResult.second;
return recordString;
}
bool DataDocumentManager::removeRecord(int ID)
{
bool deleted = false;
ByteString byteString;
byteString.insertLast(&ID,sizeof(int));
string idString = byteString.toString();
Key key(idString);
pair<Record*, BPlusTreeIterator*> searchResult = this->principalIndex->search(key);
if(searchResult.first != NULL)
{
int position = searchResult.first->getData()->readAsInt(0);
if(position > -1)
{
this->fileVariableRecord->deleteDocument(position);
this->principalIndex->removeKey(key);
ByteString bs(key.toString());
removeIDtoFileFlags(IndexWrapper::AUTOR,bs.readAsInt(0));
removeIDtoFileFlags(IndexWrapper::TITULO,bs.readAsInt(0));
removeIDtoFileFlags(IndexWrapper::PALABRAS,bs.readAsInt(0));
removeIDtoFileFlags(IndexWrapper::IDENTIFICADOR,bs.readAsInt(0));
removeIDtoFileFlags(IndexWrapper::FECHA,bs.readAsInt(0));
deleted = true;
}
}
return deleted;
}
list<int>* Mediator::getListOfPositions(int idBook, int idTerm)
{
//busca la lista de ocurrencias de un termino en un libro
ByteString* bs = new ByteString();
bs->insertLast(&idBook,sizeof(int));
bs->insertLast(&idTerm,sizeof(int));
string clave = bs->toString();
delete bs;
Key* k = new Key(clave);
list<int>* listOfPos = this->ocurrenceTree->searchAllIds(*k);
delete k;
return listOfPos;
}
int Mediator::getInfinityNorm(int idBook)
{
//busca la norma infinito
ByteString* bsId = new ByteString();
bsId->insertLast(&idBook,sizeof(int));
string clave = bsId->toString();
delete bsId;
Key* keyIdBook = new Key(clave);
pair<Record*, BPlusTreeIterator*> searchResult = this->infinityNormIndex->search(*keyIdBook);
int infinityNorm = searchResult.first->getData()->readAsInt(0);
delete searchResult.first;
delete searchResult.second;
delete keyIdBook;
return infinityNorm;
}
bool DataDocumentManager::addDocument(ByteString byteString)
{
bool saved = false;
int realID = this->autoIncInteger + 1;
int documentID;
string sid = Utility::intToString(realID);
string did;
Document document;
document.Hidratate(byteString);
document.setFakeId(realID);
if(document.getTitle()[0] == 'T')
documentID = this->twtautoIncInteger + 1;
else if(document.getTitle()[0] == 'N')
documentID = this->rssautoIncInteger + 1;
did = Utility::intToString(documentID);
document.setIdentificador(Utility::concat(document.getTitle(), did));
int position = this->fileVariableRecord->addDocument(document.Serialize());
if(position > -1)
{
this->autoIncInteger++;
if(document.getTitle()[0] == 'N')
rssautoIncInteger++;
else if(document.getTitle()[0] == 'T')
twtautoIncInteger++;
ByteString idDocument;
idDocument.insertLast(&realID, sizeof(int));
Key* key = new Key(idDocument.toString());
ByteString offset;
offset.insertLast(&position, sizeof(int));
Record* recPrincipalInd = new Record(key,&offset);
this->principalIndex->add(recPrincipalInd);
addIDtoFileFlags(IndexWrapper::AUTOR,idDocument.readAsInt(0));
addIDtoFileFlags(IndexWrapper::TITULO,idDocument.readAsInt(0));
addIDtoFileFlags(IndexWrapper::PALABRAS,idDocument.readAsInt(0));
addIDtoFileFlags(IndexWrapper::IDENTIFICADOR,idDocument.readAsInt(0));
addIDtoFileFlags(IndexWrapper::FECHA,idDocument.readAsInt(0));
saved = true;
}
return saved;
}
void BPlusTree::exportNode(ofstream& salida, Node* unNodo, int tabulacion,bool keyText,bool textContent) {
if (unNodo->isLeaf()) {
LeafNode *unNodoHoja = static_cast<LeafNode*> (unNodo);
salida << endl;
for(int i = 0 ; i < tabulacion ; i++)
salida << " ";
salida << "Numero: " << unNodoHoja->number << " Nivel: " << unNodoHoja->level << " Cant.Elem: " << unNodoHoja->keyMount
<< " Esp.Libre: " << TreeConstraits::getEfectiveSizeNode() - unNodoHoja->occupiedSpace << " Hoja.Sig: " << unNodoHoja->nextLeaf << endl;
for (int posicion = 0; posicion < unNodoHoja->keyMount; ++posicion) {
for(int i = 0 ; i < tabulacion + 4 ; i++)
salida << " ";
salida << "(";
Key unaClave = unNodoHoja->keys[posicion];
if(keyText)
{
salida << unaClave.toString();
}
else
{
ByteString clave(unaClave.toString());
salida << Utility::toString(clave.readAsInt(0));
}
salida << ";";
ByteString unDato = unNodoHoja->byteData[posicion];
if(textContent)
{
salida << unDato.toString();
}
else
{
salida << Utility::intToString(unDato.readAsInt(0));
}
salida << ")";
salida << (unaClave.getSize() + unDato.getSize() + TreeConstraits::getControlSizeRecord()) << endl;
}
salida << endl;
} else {
InnerNode *unNodoInterior = static_cast<InnerNode*> (unNodo);
salida << endl << endl;
for(int i=0; i<tabulacion ; i++)
salida << " ";
salida << "Numero: " << unNodoInterior->number << " Nivel: " << unNodoInterior->level << " Cant.Elem: " << unNodoInterior->keyMount
<< " Esp.Libre: " << TreeConstraits::getEfectiveSizeNode() - unNodoInterior->occupiedSpace << endl;
for (int posicion = 0; posicion <= unNodoInterior->keyMount; ++posicion) {
if (posicion < unNodoInterior->keyMount) {
Key unaClave = unNodoInterior->keys[posicion];
for(int i=0; i<(tabulacion+1) ; i++)
salida << " ";
salida << "(";
if(keyText)
{
salida << unaClave.toString();
}
else
{
ByteString clave(unaClave.toString());
salida << Utility::toString(clave.readAsInt(0));
}
salida << ")";
salida << unaClave.getSize();
salida << " ";
salida << unNodoInterior->getSons()[posicion] << " ";
}
}
for (int posicion = 0; posicion <= unNodoInterior->keyMount; ++posicion) {
Node *hijo = hidratateNode(unNodoInterior->sons[posicion]);
exportNode(salida, hijo, tabulacion + 2,keyText,textContent);
if (hijo)
freeNodeMemory(hijo);
}
for(int i=0; i<tabulacion ; i++)
salida << " ";
salida << endl;
}
}
bool Mediator::indexingListOnIndex(IndexWrapper::indexItem item)
{
bool done = false;
List<int>* flagsLs = this->dataBookManager->getListToFileFlag(item);
if(!flagsLs->isEmpty())
{
cout << "Process Started: " << Utility::getDate() << endl;
ListIterator<int> it = flagsLs->getIterator();
int ID;
while(it.hasNext())
{
ID = it.next();
ByteString bs = this->dataBookManager->getBook(ID);
Book book;
book.Hidratate(bs);
string clave;
Record* record;
ByteString bsIdBook;
ByteString* bsId;
bsIdBook.insertLast(&ID,sizeof(int));
Key* key;
switch(item){
case IndexWrapper::AUTOR :
clave = book.getAuthor();
cout << "Book begin process: " << book.getTitle() << " - Author:" << clave << " - Start:" << Utility::getDate();
key = new Key(clave);
bsId = new ByteString(bsIdBook);
record = new Record(key,bsId);
done = this->indexWrapper->add(record,item);
break;
case IndexWrapper::EDITORIAL :
clave = book.getEditorial();
cout << "Book begin process: " << book.getTitle() << " - Editorial:" << clave << " - Start:" << Utility::getDate();
key = new Key(clave);
bsId = new ByteString(bsIdBook);
record = new Record(key,bsId);
done = this->indexWrapper->add(record,item);
break;
case IndexWrapper::TITULO :
clave = book.getTitle();
cout << "Book begin process: " << book.getTitle() << " - Start:" << Utility::getDate();
key = new Key(clave);
bsId = new ByteString(bsIdBook);
record = new Record(key,bsId);
done = this->indexWrapper->add(record,item);
break;
case IndexWrapper::PALABRAS :
this->fileParser->setWords(book.getText());
// ----------------------------------------------------------------------------------------------- //
// Se incorpora la norma infinito del documento
// ----------------------------------------------------------------------------------------------- //
Key* keyIdBook = new Key(bsIdBook.toString());
ByteString* bsInfinityNorm = new ByteString();
unsigned int infinityNorm = this->fileParser->getInfinityNorm();
bsInfinityNorm->insertLast(&infinityNorm, sizeof(unsigned int));
Record* recInfinityNorm = new Record(keyIdBook, bsInfinityNorm);
this->infinityNormIndex->add(recInfinityNorm);
delete recInfinityNorm;
// ----------------------------------------------------------------------------------------------- //
map<string,Term*> terms = this->fileParser->getTerms();
cout << "Book begin process: " << book.getTitle() << " - Words Count:" << Utility::intToString(terms.size()) << " - Start:" << Utility::getDate();
int i = 0;
for(map<string,Term*>::iterator it = terms.begin(); it != terms.end(); ++it)
{
i++;
// ------------------------ //
key = new Key(it->first);
string word = it->first;
list<int>* listDocuments = this->indexWrapper->searchAllIds(key, IndexWrapper::PALABRAS);
int idTerm;
// Si no tiene datos --> la palabra no existe en el hash
if (listDocuments->size() == 0)
{
//delete poddo
delete listDocuments;
// Se agrega la palabra al vocabulario y se obtiene su id
int possibleID = this->autoIncInteger + 1;
// La clave es el id de termino y el dato la palabra
ByteString bsIdTerm;
bsIdTerm.insertLast(&possibleID, sizeof(int));
Key* keyIdTerm = new Key(bsIdTerm.toString());
ByteString* bsTerm = new ByteString();
bsTerm->insertLast(word);
Record* recVocabulary = new Record(keyIdTerm, bsTerm);
this->vocabularyIndex->add(recVocabulary);
delete recVocabulary;
idTerm = possibleID;
this->autoIncInteger++;
}
else
{
// Devuelve en la ultima posicion el id del termino y en las restantes los id de documentos
Word* newWord = new Word(word, listDocuments);
idTerm = newWord->getIdWord();
delete newWord; // se elimina la lisdDocuments tambien
}
//Aca se agrega las ocurrencias del termino en el libro
list<unsigned int> listOfPositions = it->second->getPositions();
ByteString bs;
bs.insertLast(&ID,sizeof(int));
bs.insertLast(&idTerm,sizeof(int));
Record* r = new Record();
r->setKey(new Key(bs.toString()));
this->ocurrenceTree->addList(r,listOfPositions);
listOfPositions.clear();
delete r;
// ------------------------ //
bsId = new ByteString(bsIdBook);
bsId->insertLast(&idTerm, sizeof(int));
record = new Record(key,bsId);
done = this->indexWrapper->add(record,item);
delete record;
}
break;
}
cout << "Book processed: " << book.getTitle() << " - Finish:" << Utility::getDate() << endl;
}
cout << "Process Ended: " << Utility::getDate();
}
//delete poddo
if (flagsLs != NULL)
delete flagsLs;
return done;
}
void ClassifBPlusTree::exportNode(ofstream& salida, Node* unNodo, int tabulacion,bool keytext, bool textContent) {
if (unNodo->isLeaf()) {
LeafNode *unNodoHoja = static_cast<LeafNode*> (unNodo);
salida << endl;
for(int i = 0 ; i < tabulacion ; i++)
salida << " ";
salida << "Numero: " << unNodoHoja->number << " Nivel: " << unNodoHoja->level << " Cant.Elem: " << unNodoHoja->keyMount << " Esp.Libre: " << TreeConstraits::getEfectiveSizeNode() - unNodoHoja->occupiedSpace << " Hoja.Sig: " << unNodoHoja->nextLeaf << endl;
for (int posicion = 0; posicion < unNodoHoja->keyMount; ++posicion) {
for(int i = 0 ; i < tabulacion + 4 ; i++)
salida << " ";
salida << "(";
Key unaClave = unNodoHoja->keys[posicion];
if(keytext)
{
salida << unaClave.toString();
}
else
{
ByteString clave(unaClave.toString());
if(clave.getSize()==(2*sizeof(int)))
{
salida << Utility::toString(clave.readAsInt(0));
salida << "-";
salida << Utility::toString(clave.readAsInt(sizeof(int)));
}
else
salida << Utility::toString(clave.readAsInt(0));
}
salida << ";";
ByteString unDato = unNodoHoja->byteData[posicion];
int idBlock = unDato.readAsInt(0);
string datoString = Utility::intToString(idBlock);
salida << datoString;
salida << ")";
salida << (unaClave.getSize() + unDato.getSize() + TreeConstraits::getControlSizeRecord()) << endl;
for(int i = 0 ; i < tabulacion + 4 ; i++)
salida << " ";
ListofID listOfID(this->fileBlockManager,idBlock);
list<int> listOfInt = listOfID.getListID();
list<int>::iterator it;
int id;
//int count = 0;
ByteString bs;
unsigned int i = 0;
for(it=listOfInt.begin();it!=listOfInt.end();++it)
{
i++;
id = *it;
bs.insertLast(Utility::intToString(id));
bs.insertLast(" ");
if((i == listOfInt.size()) || (bs.toString().size() >80))
{
salida << bs.toString();
salida << endl;
if(i != listOfInt.size())
{
for(int i = 0 ; i < tabulacion + 4 ; i++)
salida << " ";
}
bs.clean();
}
}
}
} else {
InnerNode *unNodoInterior = static_cast<InnerNode*> (unNodo);
salida << endl << endl;
for(int i=0; i<tabulacion ; i++)
salida << " ";
salida << "Numero: " << unNodoInterior->number << " Nivel: " << unNodoInterior->level << " Cant.Elem: " << unNodoInterior->keyMount
<< " Esp.Libre: " << TreeConstraits::getEfectiveSizeNode() - unNodoInterior->occupiedSpace << endl;
for (int posicion = 0; posicion <= unNodoInterior->keyMount; ++posicion) {
if (posicion < unNodoInterior->keyMount) {
Key unaClave = unNodoInterior->keys[posicion];
for(int i=0; i<(tabulacion+1) ; i++)
salida << " ";
salida << "(";
if(keytext)
{
salida << unaClave.toString();
}
else
{
ByteString clave(unaClave.toString());
salida << Utility::toString(clave.readAsInt(0));
}
salida << ")";
salida << unaClave.getSize();
salida << " ";
salida << unNodoInterior->getSons()[posicion] << " ";
}
}
for (int posicion = 0; posicion <= unNodoInterior->keyMount; ++posicion) {
Node *hijo = hidratateNode(unNodoInterior->sons[posicion]);
exportNode(salida, hijo, tabulacion + 2,keytext,textContent);
if (hijo)
freeNodeMemory(hijo);
}
for(int i=0; i<tabulacion ; i++)
salida << " ";
salida << endl;
}
}
