java - 从 apache lucene 索引中搜索并按组计算结果

Question

我正在尝试从 lucene 索引中搜索，但我想过滤此搜索。有两个字段内容和类别。假设我想搜索具有“sports”的文件，并且我还想计算有多少文件属于 a 和 b 类别。我正在尝试使用以下代码来实现此目的。但问题是，如果有数百万条记录，那么由于循环执行，速度会变慢，请建议我另一种方法来完成任务。

try { File indexDir= new File("文件路径")

           Directory directory = FSDirectory.open(indexDir);

                IndexSearcher searcher = new IndexSearcher(directory, true);
                int maxhits=1000000;
                QueryParser parser1 = new QueryParser(Version.LUCENE_36, "contents",

                  new StandardAnalyzer(Version.LUCENE_36));

          Query qu=parser1.parse("sport");

                TopDocs topDocs = searcher.search(, maxhits);
                ScoreDoc[] hits = topDocs.scoreDocs;


          len = hits.length;

       JOptionPane.showMessageDialog(null,"found times"+len);

                 int docId = 0;
                Document d;





 String category="";

int ctr=0,ctr1=0;

for ( i = 0; i<len; i++) {
docId = hits[i].doc;
d = searcher.doc(docId);
category= d.get(("category"));
if(category.equals("a"))
ctr++;
if(category.equals("b"))
ctr1++;


}

  JOptionPane.showMessageDialog("wprd found in category a times"+ctr);
   JOptionPane.showMessageDialog("wprd found in category b times"+ctr1);
  }

 catch(Exception ex)

 {

  ex.printStackTrace();
 }

Answer 1

您可以只查询您要查找的每个类别并获取totalHits。更好的是使用 TotalHitCountCollector ，而不是获取 TopDocs 实例:

Query query = parser1.parser("+sport +category:a")
TotalHitCountCollector collector = new TotalHitCountCollector();
search.search(query, collector); 
ctr = collector.getTotalHits();
query = parser1.parser("+sport +category:b")
collector = new TotalHitCountCollector();
search.search(query, collector); 
ctr1 = collector.getTotalHits();