Beispiel #1
0
 /**
  * Adds a document to this segment.
  *
  * @param \Zend\Search\Lucene\Document $document
  * @throws \Zend\Search\Lucene\Exception
  */
 public function addDocument(Document $document)
 {
     $storedFields = array();
     $docNorms = array();
     $similarity = Similarity::getDefault();
     foreach ($document->getFieldNames() as $fieldName) {
         $field = $document->getField($fieldName);
         if ($field->storeTermVector) {
             /**
              * @todo term vector storing support
              */
             throw new Lucene\Exception('Store term vector functionality is not supported yet.');
         }
         if ($field->isIndexed) {
             if ($field->isTokenized) {
                 $analyzer = Analyzer\Analyzer::getDefault();
                 $analyzer->setInput($field->value, $field->encoding);
                 $position = 0;
                 $tokenCounter = 0;
                 while (($token = $analyzer->nextToken()) !== null) {
                     $tokenCounter++;
                     $term = new Index\Term($token->getTermText(), $field->name);
                     $termKey = $term->key();
                     if (!isset($this->_termDictionary[$termKey])) {
                         // New term
                         $this->_termDictionary[$termKey] = $term;
                         $this->_termDocs[$termKey] = array();
                         $this->_termDocs[$termKey][$this->_docCount] = array();
                     } else {
                         if (!isset($this->_termDocs[$termKey][$this->_docCount])) {
                             // Existing term, but new term entry
                             $this->_termDocs[$termKey][$this->_docCount] = array();
                         }
                     }
                     $position += $token->getPositionIncrement();
                     $this->_termDocs[$termKey][$this->_docCount][] = $position;
                 }
                 if ($tokenCounter == 0) {
                     // Field contains empty value. Treat it as non-indexed and non-tokenized
                     $field = clone $field;
                     $field->isIndexed = $field->isTokenized = false;
                 } else {
                     $docNorms[$field->name] = chr($similarity->encodeNorm($similarity->lengthNorm($field->name, $tokenCounter) * $document->boost * $field->boost));
                 }
             } else {
                 if (($fieldUtf8Value = $field->getUtf8Value()) == '') {
                     // Field contains empty value. Treat it as non-indexed and non-tokenized
                     $field = clone $field;
                     $field->isIndexed = $field->isTokenized = false;
                 } else {
                     $term = new Index\Term($fieldUtf8Value, $field->name);
                     $termKey = $term->key();
                     if (!isset($this->_termDictionary[$termKey])) {
                         // New term
                         $this->_termDictionary[$termKey] = $term;
                         $this->_termDocs[$termKey] = array();
                         $this->_termDocs[$termKey][$this->_docCount] = array();
                     } else {
                         if (!isset($this->_termDocs[$termKey][$this->_docCount])) {
                             // Existing term, but new term entry
                             $this->_termDocs[$termKey][$this->_docCount] = array();
                         }
                     }
                     $this->_termDocs[$termKey][$this->_docCount][] = 0;
                     // position
                     $docNorms[$field->name] = chr($similarity->encodeNorm($similarity->lengthNorm($field->name, 1) * $document->boost * $field->boost));
                 }
             }
         }
         if ($field->isStored) {
             $storedFields[] = $field;
         }
         $this->addField($field);
     }
     foreach ($this->_fields as $fieldName => $field) {
         if (!$field->isIndexed) {
             continue;
         }
         if (!isset($this->_norms[$fieldName])) {
             $this->_norms[$fieldName] = str_repeat(chr($similarity->encodeNorm($similarity->lengthNorm($fieldName, 0))), $this->_docCount);
         }
         if (isset($docNorms[$fieldName])) {
             $this->_norms[$fieldName] .= $docNorms[$fieldName];
         } else {
             $this->_norms[$fieldName] .= chr($similarity->encodeNorm($similarity->lengthNorm($fieldName, 0)));
         }
     }
     $this->addStoredFields($storedFields);
 }
Beispiel #2
0
 /**
  * Returns norm vector, encoded in a byte string
  *
  * @param string $fieldName
  * @return string
  */
 public function normVector($fieldName)
 {
     $fieldNum = $this->getFieldNum($fieldName);
     if ($fieldNum == -1 || !$this->_fields[$fieldNum]->isIndexed) {
         $similarity = Similarity::getDefault();
         return str_repeat(chr($similarity->encodeNorm($similarity->lengthNorm($fieldName, 0))), $this->_docCount);
     }
     if (!isset($this->_norms[$fieldNum])) {
         $this->_loadNorm($fieldNum);
     }
     return $this->_norms[$fieldNum];
 }
Beispiel #3
0
 /**
  * Retrive similarity used by index reader
  *
  * @return \Zend\Search\Lucene\Search\Similarity
  */
 public function getSimilarity()
 {
     return Similarity::getDefault();
 }