Lucene5学习之多索引目录查询以及多线程查询

     上一篇中我们使用多线程创建了索引，下面我们来试着采用不把多个索引目录里的数据合并到一个新的索引目录的方式去查询索引数据，当然你也可以合并(合并到一个索引目录查询就很简单了)，其实很多情况我们都是不合并到一个索引目录的，那多索引目录该如何查询呢，在Lucene5中使用的MultiReader类，在Lucene4时代，使用的是MultiSearcher类。至于Lucene多线程查询，只需要在构建IndexSearcher对象时传入一个ExecutorService线程池管理对象即可，具体请看下面贴出的示例代码：

package com.yida.framework.lucene5.index;

import java.io.IOException;
import java.util.ArrayList;
import java.util.List;
import java.util.concurrent.Callable;
import java.util.concurrent.ExecutionException;
import java.util.concurrent.ExecutorService;
import java.util.concurrent.Executors;
import java.util.concurrent.Future;

import org.apache.lucene.document.Document;
import org.apache.lucene.index.DirectoryReader;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.MultiReader;
import org.apache.lucene.index.Term;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.TermQuery;
import org.apache.lucene.store.Directory;

import com.yida.framework.lucene5.util.LuceneUtils;

/**
 * 多线程多索引目录查询测试
 * @author Lanxiaowei
 *
 */
public class MultiThreadSearchTest {
	public static void main(String[] args) throws InterruptedException, ExecutionException, IOException {
		//每个线程都从5个索引目录中查询，所以最终5个线程的查询结果都一样
		//multiThreadAndMultiReaderSearch();
		
		//多索引目录查询(把多个索引目录当作一个索引目录)
		multiReaderSearch();
	}
	
	/**
	 * 多索引目录查询
	 * @throws InterruptedException
	 * @throws ExecutionException
	 * @throws IOException
	 */
	public static void multiReaderSearch()  throws InterruptedException, ExecutionException, IOException {
		Directory directory1 = LuceneUtils.openFSDirectory("C:/lucenedir1");
		Directory directory2 = LuceneUtils.openFSDirectory("C:/lucenedir2");
		Directory directory3 = LuceneUtils.openFSDirectory("C:/lucenedir3");
		Directory directory4 = LuceneUtils.openFSDirectory("C:/lucenedir4");
		Directory directory5 = LuceneUtils.openFSDirectory("C:/lucenedir5");
		IndexReader reader1 = DirectoryReader.open(directory1);
		IndexReader reader2 = DirectoryReader.open(directory2);
		IndexReader reader3 = DirectoryReader.open(directory3);
		IndexReader reader4 = DirectoryReader.open(directory4);
		IndexReader reader5 = DirectoryReader.open(directory5);
		MultiReader multiReader = new MultiReader(reader1,reader2,reader3,reader4,reader5);
		
		IndexSearcher indexSearcher = LuceneUtils.getIndexSearcher(multiReader);
		Query query = new TermQuery(new Term("contents","volatile"));
		List<Document> list = LuceneUtils.query(indexSearcher, query);
		if(null == list || list.size() <= 0) {
			System.out.println("No results.");
			return;
		}
		for(Document doc : list) {
			String path = doc.get("path");
			//String content = doc.get("contents");
			System.out.println("path:" + path);
			//System.out.println("contents:" + content);
		}
	}
	
	/**
	 * 多索引目录且多线程查询，异步收集查询结果
	 * @throws InterruptedException
	 * @throws ExecutionException
	 * @throws IOException
	 */
	public static void multiThreadAndMultiReaderSearch()  throws InterruptedException, ExecutionException, IOException {
		int count = 5;
		ExecutorService pool = Executors.newFixedThreadPool(count);
		
		Directory directory1 = LuceneUtils.openFSDirectory("C:/lucenedir1");
		Directory directory2 = LuceneUtils.openFSDirectory("C:/lucenedir2");
		Directory directory3 = LuceneUtils.openFSDirectory("C:/lucenedir3");
		Directory directory4 = LuceneUtils.openFSDirectory("C:/lucenedir4");
		Directory directory5 = LuceneUtils.openFSDirectory("C:/lucenedir5");
		IndexReader reader1 = DirectoryReader.open(directory1);
		IndexReader reader2 = DirectoryReader.open(directory2);
		IndexReader reader3 = DirectoryReader.open(directory3);
		IndexReader reader4 = DirectoryReader.open(directory4);
		IndexReader reader5 = DirectoryReader.open(directory5);
		MultiReader multiReader = new MultiReader(reader1,reader2,reader3,reader4,reader5);
		
		final IndexSearcher indexSearcher = LuceneUtils.getIndexSearcher(multiReader, pool);
		final Query query = new TermQuery(new Term("contents","volatile"));
		List<Future<List<Document>>> futures = new ArrayList<Future<List<Document>>>(count);
		for (int i = 0; i < count; i++) {
			futures.add(pool.submit(new Callable<List<Document>>() {
				public List<Document> call() throws Exception {
					return LuceneUtils.query(indexSearcher, query);
				}
			}));
		}
		
		int t = 0;
		//通过Future异步获取线程执行后返回的结果
		for (Future<List<Document>> future : futures) {
			List<Document> list = future.get();
			if(null == list || list.size() <= 0) {
				t++;
				continue;
			}
			for(Document doc : list) {
				String path = doc.get("path");
				//String content = doc.get("contents");
				System.out.println("path:" + path);
				//System.out.println("contents:" + content);
			}
			System.out.println("");
		}
		//释放线程池资源
		pool.shutdown();
		
		if(t == count) {
			System.out.println("No results.");
		}
	}
}

当然你也可以把上面的代码改造成每个线程查询一个索引目录，我上面是每个线程都从5个索引目录中查询，所以结果会打印5次，看到运行结果请不要感到奇怪。

 

如果你还有什么问题请加我Ｑ-Q：7-3-6-0-3-1-3-0-5，

或者加裙
一起交流学习！

已有 0 人发表留言，猛击->> 这里<<-参与讨论


ITeye推荐

• —软件人才免语言低担保 赴美带薪读研！—