Zend Framework之Search_Lucene实例

//创建具有不同特点的字段
   $doc = new Zend_Search_Lucene_Document();// Field is not tokenized, but is indexed and stored within the index.
   // Stored fields can be retrived from the index.
   $doc->addField(Zend_Search_Lucene_Field::Keyword('doctype', 'autogenerated'));
   // Field is not tokenized nor indexed, but is stored in the index.
   $doc->addField(Zend_Search_Lucene_Field::UnIndexed('created', time()));
   // Binary String valued Field that is not tokenized nor indexed,
   // but is stored in the index.
   $doc->addField(Zend_Search_Lucene_Field::Binary('icon', $iconData));
   // Field is tokenized and indexed, and is stored in the index.
   $doc->addField(Zend_Search_Lucene_Field::Text('annotation', 'Document annotation text'));
   // Field is tokenized and indexed, but that is not stored in the index.
   $doc->addField(Zend_Search_Lucene_Field::UnStored('contents', 'My document content'));

Keyword 关键词字段是被保存和被索引的，意思是它们既可以被搜索，也可以在搜索结果中被显示。它们并没有以记号化的方式拆分为多个不同的词。在 Zend_Search_Lucene 中，枚举形的数据库字段通常可以很好的转化为关键词字段。

• 
  UnIndexed 不索引字段是不可搜索的，但是它们会在搜索结果中返回用于生成点击信息。数据库的时间戳、主键、文件系统的路径、以及其它标识是不索引字段的好的候选人。
  
• 
  Binary 二进制字段是不记号化和不被索引的，但是被保存以供生成点击信息。它们可以用于保存任何以二进制方式编码的信息，例如图标等等。
  
• 
  Text 文本字段是被保存的、被索引的和记号化的。文本字段适合用于保存像是主题、标题这样既能被搜索又能作为搜索结果返回的信息。
  
• 
  UnStored 不保存字段是记号化和被索引的，但并不保存在索引中。大量的文本信息最好使用这种字段类型。保存的数据在硬盘上创建了大量的索引，如果你需要搜索而不需要在搜索结果中显示这些数据，就使用不保存字段。当结合使用 Zend_Search_Lucene 索引和关系数据库时最适合使用不保存字段。你通过不保存字段索引大量数据用于搜索，并通过作为标识的特定字段从你的关系数据库中获取它们。
  表 27.1. Zend_Search_Lucene_Field 类型
  字段类型
  保存
  索引
  记号化
  二进制
  Keyword
  是
  是
  否
  否
  UnIndexed
  是
  否
  否
  否
  Binary
  是
  否
  否
  是
  Text
  是
  是
  是
  否
  UnStored
  否
  是
  是
  否
     
     //创建索引
     // Setting the second argument to TRUE creates a new index
     $index = new Zend_Search_Lucene('/data/my-index', true);
     $doc = new Zend_Search_Lucene_Document();
     // Store document URL to identify it in search result.
     $doc->addField(Zend_Search_Lucene_Field::Text('url', $docUrl));
     // Index document content
     $doc->addField(Zend_Search_Lucene_Field::UnStored('contents', $docContent));
     // Add document to the index.
     $index->addDocument($doc);
     // Write changes to the index.
     $index->commit();
     
     //更新索引
     // Open existing index
     $index = new Zend_Search_Lucene('/data/my-index');
     $doc = new Zend_Search_Lucene_Document();
     // Store document URL to identify it in search result.
     $doc->addField(Zend_Search_Lucene_Field::Text('url', $docUrl));
     // Index document content
     $doc->addField(Zend_Search_Lucene_Field::UnStored('contents', $docContent));
     // Add document to the index.
     $index->addDocument($doc);
     // Write changes to
     $index->commit();
     
     //搜索索引
     $index = new Zend_Search_Lucene('/data/my_index');
     $index->find($query);
     
     //搜索结果是一个 Zend_Search_Lucene_Search_QueryHit 对象数组
     $index = new Zend_Search_Lucene('/data/my_index');
     $hits = $index->find($query);
     foreach ($hits as $hit) {   
       echo $hit->id;   
       echo $hit->score;   
       echo $hit->title;   
       echo $hit->author;
     }
     
     //原始的 Zend_Search_Lucene_Document 对象可以从 Zend_Search_Lucene_Search_QueryHit 获得
     $index = new Zend_Search_Lucene('/data/my_index');
     $hits = $index->find($query);
     foreach ($hits as $hit) {   
       // return Zend_Search_Lucene_Document object for this hit   
       echo $document = $hit->getDocument();   
       // return a Zend_Search_Lucene_Field object   
       // from the Zend_Search_Lucene_Document   
       echo $document->getField('title');   
       // return the string value of the Zend_Search_Lucene_Field object   
       echo $document->getFieldValue('title');   
       // same as getFieldValue()   
       echo $document->title;
     }
     
     //Zend_Search_Lucene 使用和 Java Lucene 一样的评分算法。搜索结果是按照分值进行排序的。分值越大，相应的搜索结果点击排在排位越靠前。
     $hits = $index->find($query);
     foreach ($hits as $hit) {   
      echo $hit->id;   
      echo $hit->score;
     }
     
     //单项查询
     $hits = $index->find('word1');//查询字串
     //通过 API 创建查询：
     $term  = new Zend_Search_Lucene_Index_Term('word1');   
     $query = new Zend_Search_Lucene_Search_Query_Term($term);   
     $hits  = $index->find($query);
     
     //多项查询
     $hits = $index->find('+word1 author:word2 -word3');//查询字串
     //通过 API 创建查询：
     $query = new Zend_Search_Lucene_Search_Query_MultiTerm();   
     $query->addTerm(new Zend_Search_Lucene_Index_Term('word1'), true);   
     $query->addTerm(new Zend_Search_Lucene_Index_Term('word2'), null);   
     $query->addTerm(new Zend_Search_Lucene_Index_Term('word3'), false);   
     $hits  = $index->find($query);
     
     //短语查询
     $query1 = new Zend_Search_Lucene_Search_Query_Phrase();
     // Add 'word1' at 0 relative position.
     $query1->addTerm(new Zend_Search_Lucene_Index_Term('word1'));
     // Add 'word2' at 1 relative position.
     $query1->addTerm(new Zend_Search_Lucene_Index_Term('word2'));
     // Add 'word3' at 3 relative position.
     $query1->addTerm(new Zend_Search_Lucene_Index_Term('word3'), 3);
     
     Zend_Search_Lucene_Search_Query_Phrase([array $terms[, array $offsets[, string $field]]]);
     //查询zend framework
     $query = new Zend_Search_Lucene_Search_Query_Phrase(array('zend', 'framework'));
     //将会搜索短语“zend ????? download”并匹配'zend platform download'、 'zend studio download'
     $query = new Zend_Search_Lucene_Search_Query_Phrase(array('zend', 'download'), array(0, 2));
     //在title字段中搜索zend framework
     $query = new Zend_Search_Lucene_Search_Query_Phrase(array('zend', 'framework'), null, 'title');
     
     Zend_Search_Lucene_Search_Query_Phrase::addTerm(Zend_Search_Lucene_Index_Term $term[, integer $position]);
     //查询zend framework
     $query = new Zend_Search_Lucene_Search_Query_Phrase();
     $query->addTerm(new Zend_Search_Lucene_Index_Term('zend'));
     $query->addTerm(new Zend_Search_Lucene_Index_Term('framework'));
     ////将会搜索短语“zend ????? download”并匹配'zend platform download'、 'zend studio download'
     $query = new Zend_Search_Lucene_Search_Query_Phrase();
     $query->addTerm(new Zend_Search_Lucene_Index_Term('zend'), 0);
     $query->addTerm(new Zend_Search_Lucene_Index_Term('framework'), 2);
     //在title字段中搜索zend framework
     $query = new Zend_Search_Lucene_Search_Query_Phrase();
     $query->addTerm(new Zend_Search_Lucene_Index_Term('zend', 'title'));
     $query->addTerm(new Zend_Search_Lucene_Index_Term('framework', 'title'));
     //查询模糊因子
     // Query without a gap.
     $query = new Zend_Search_Lucene_Search_Query_Phrase(array('word1', 'word2'));
     // Search for 'word1 word2', 'word1 ... word2'
     $query->setSlop(1);
     $hits1 = $index->find($query);
     // Search for 'word1 word2', 'word1 ... word2','word1 ... ... word2', 'word2 word1'
     $query->setSlop(2);
     $hits2 = $index->find($query);
     
     //字符集
     $doc = new Zend_Search_Lucene_Document();
     $docText = iconv('ISO-8859-1', 'ASCII//TRANSLIT', $docText);
     $doc->addField(Zend_Search_Lucene_Field::UnStored('contents', $docText));
     $query = iconv('', 'ASCII//TRANSLIT', $query);$hits = $index->find($query);
  
  简单例子
     $index = new Zend_Search_Lucene('my-index', true);
     
     $doc = new Zend_Search_Lucene_Document();
     
     // Store document URL to identify it in search result.
     $doc->addField(Zend_Search_Lucene_Field::Text('url', 'http://www.eyuwo.com'));
     
     // Index document content
     $doc->addField(Zend_Search_Lucene_Field::UnStored('contents', 'this is just a test of Zend_Search_lucene'));
     
     // Add document to the index.
     $index->addDocument($doc);
     
     // Write changes to the index.
     $index->commit();
     //$query='test';//查询的关键字
     $term  = new Zend_Search_Lucene_Index_Term('test');   
       $query = new Zend_Search_Lucene_Search_Query_Term($term);   
     $hits = $index->find($query);//得到查询的结果
     if(!empty($hits)){
      foreach ($hits as $hit) {
          echo 'Score:'.$hit->score.'
  ';
          echo 'Url:'.$hit->url.'
  ';
          echo 'ID:'.$hit->id.'
  ';
      }
     }else {
      error_message("没有相关的信息!");
     }