用 Lucene 做一個簡單的 Java 搜索工具

?

初學 Lucene ，剛接觸搜索引擎。知道了一點點，想做個小工具，實現根據 “ 單詞 ” 搜索某個 java 源文件。比如輸入 “String” 去查詢某些 java 源文件里用到了這個類。

這個想法的來源是，在以前剛學 java 時，有一本 java 基礎教程的書的附帶光盤里有作者寫的一個程序，可以方便初學者查找某些類在哪個實例里出現。當時沒有太在意，覺得作者的代碼很長。所以現在想自己也寫一個這樣的小程序。

?

開發工具與運行環境：

使用 Lucene2.0 的包， jdk1.5 ，在 WindowsXP 下運行。

思路分析與設計：

整個程序里，除了 Lucene 的必要操作外，就是 IO 的基本操作了。因為要對某目錄下及其子目錄下的所有 Java 源文件進行索引，就要用到遞歸，同時要過濾掉非 Java 源文件。根據這種情況，設計了以下 5 個類。

異常類不是必要的，特意設計來包裝 IO 異常、文件異常和 Lucene 的異常。文件過濾工廠類的出現并不是故弄玄虛，只是不想太多代碼集中一起，就把文件過慮器的設計放到一個類里。下面是程序的完整代碼及注釋。

?

IndexJavaFiles.java

/**

? * index the java source files

? */

package powerwind;

?

import java.io.*;

import java.util.Date;

?

import org.apache.lucene.document.*;

import org.apache.lucene.index.IndexWriter;

?

/**

? * @author Powerwind

? * @version 1.0

? */

public class IndexJavaFiles {

?

??? /**

??? ? * 默認構造方法

??? ? */

??? public IndexJavaFiles() {

??? }

?

??? /**

??? ? * 這個私有遞歸方法由 index 方法調用，保證 index 傳入的 file 是目錄不是文件

??? ? *

??? ? * @param writer

??? ? * @param file

??? ? * @param ff

??? ? * @throws IndexException

??? ? */

??? private void indexDirectory(IndexWriter writer, File file, FileFilter filter) throws IndexException {

?????? if (file.isDirectory()) {

?????????? // 有選擇地（過濾）獲取目錄下的文件和目錄

?????????? File[] files = file.listFiles(filter);

?????????? // 非空目錄

?????????? if (files != null ) {

????????????? for ( int i = 0; i < files. length ; i++) {

????????????????? indexDirectory(writer, files[i], filter);

????????????? }

?????????? }

?????? } else {

?????????? try {

?????????? 　　 // 這里的 file 經過先前的過濾

????????????? writer.addDocument(parseFile(file));

????????????? System. out .println( " 增加文件： " + file);

?????????? } catch (IOException ioe) {

????????????? throw new IndexException(ioe.getMessage());

?????????? }

?????? }

??? }

?

??? /**

??? ? * 傳參數是文件就直接索引，若是目錄則交給 indexDirectory 遞歸

??? ? *

??? ? * @param writer

??? ? * @param file

??? ? * @param ff

??? ? * @throws IndexException

??? ? */

??? public void index(IndexWriter writer, File file, FileFilter filter) throws IndexException {

?????? // 確定可讀

?????? if (file.exists() && file.canRead()) {

?????????? if (file.isDirectory()) {

????????????? indexDirectory(writer, file, filter);

?????????? } else if (filter.accept(file)) {

????????????? try {

????????????????? writer.addDocument(parseFile(file));

????????????????? System. out .println( " 增加文件： " + file);

????????????? } catch (IOException ioe) {

????????????????? throw new IndexException(ioe.getMessage());

????????????? }

?????????? } else {

????????????? System. out .println( " 指定文件或目錄錯誤，沒有完成索引 " );

?????????? }

?????? }

??? }

?

??? /**

??? ? * @param file

??? ? *

??? ? * 把 File 變成 Document

??? ? */

??? private Document parseFile(File file) throws IndexException {

?????? Document doc = new Document();

?????? doc.add( new Field( "path" , file.getAbsolutePath(), Field.Store. YES ,

???????????????????? Field.Index. UN_TOKENIZED ));

?????? try {

?????????? doc.add( new Field( "contents" , new FileReader(file)));

?????? } catch (FileNotFoundException fnfe) {

?????????? throw new IndexException(fnfe.getMessage());

?????? }

?????? return doc;

??? }

}

?

index(IndexWriter writer, File file, FileFilter filter) 調用私有方法 indexDirectory(IndexWriter writer, File file, FileFilter filter) 完成文件的索引。

下面是 IndexException 異常類。

IndexException.java

package powerwind;

?

public class IndexException extends Exception {

?

??? public IndexException(String message) {

?????? super ( "Throw IndexException while indexing files: " + message);

??? }

?

}

下面是 FileFilterFactory 類，返回一個特定的文件過濾器（ FileFilter ）。

FileFilterFactory.java

package powerwind;

?

import java.io.*;

?

public class FileFilterFactory {

??? /**

??? ? * 靜態匿名內部類

??? ? */

??? private static FileFilter filter = new FileFilter() {

?????? public boolean accept(File file) {

?????????? long len;

?????????? return file.isDirectory()||

????????????????? ?(file.getName().endsWith( ".java" ) &&

????????????????? ?((len = file.length()) > 0) && len < 1024 * 1024);

?????? }

??? };

??? public static FileFilter getFilter() {

?????? return filter ;

??? }

}

?

main 方法

??? /**

??? ? * ????? main 方法

??? ? */

??? public static void main(String[] args) throws Exception {

?????? IndexJavaFiles ijf = new IndexJavaFiles();

?????? Date start = new Date();

?????? try {

?????????? IndexWriter writer = IndexWriterFactory.newInstance().createWriter( "./index" , true );

?????????? System. out .println( "Indexing ..." );

?????????? ijf.index(writer, new File( "." ), FileFilterFactory.getFilter());

?????????? System. out .println( "Optimizing..." );

?????????? writer.optimize();

?????????? writer.close();

?

?????????? Date end = new Date();

?????????? System. out .println(end.getTime() - start.getTime() + " total milliseconds" );

?

?????? } catch (IOException e) {

?????????? System. out .println( " caught a " + e.getClass() + "\n with message: " + e.getMessage());

?????? }

??? }

?

?

SearchJavaFiles.java

package powerwind;

?

import java.io.*;

?

import org.apache.lucene.analysis.Analyzer;

import org.apache.lucene.analysis.standard.StandardAnalyzer;

import org.apache.lucene.document.Document;

import org.apache.lucene.index.IndexReader;

import org.apache.lucene.queryParser.*;

import org.apache.lucene.search.*;

?

public class SearchJavaFiles {

??? private IndexSearcher searcher ;

?

??? private QueryParser parser ;

?

??? /**

??? ? *

??? ? * @param searcher

??? ? */

??? public SearchJavaFiles(IndexSearcher searcher) {

?????? this . searcher = searcher;

??? }

?

??? /**

??? ? *

??? ? * @param field

??? ? * @param analyzer

??? ? */

??? public void setParser(String field, Analyzer analyzer) {

?????? setParser( new QueryParser(field, analyzer));

??? }

?

??? /**

??? ? * @param parser

??? ? */

??? public void setParser(QueryParser parser) {

?????? this . parser = parser;

??? }

?

??? /**

??? ? *

??? ? * @param query

??? ? * @return Hits

??? ? * @throws SearchException

??? ? */

??? public Hits serach(Query query) throws SearchException {

?????? try {

?????????? return searcher .search(query);

?????? } catch (IOException ioe) {

?????????? throw new SearchException(ioe.getMessage());

?????? }

??? }

?

??? /**

??? ? *

??? ? * @param queryString

??? ? * @return Hits

??? ? * @throws SearchException

??? ? */

??? public Hits serach(String queryString) throws SearchException {

?????? if ( parser == null )

?????????? throw new SearchException( "parser is null!" );

?????? try {

?????????? return searcher .search( parser .parse(queryString));

?????? } catch (IOException ioe) {

?????????? throw new SearchException(ioe.getMessage());

?????? } catch (ParseException pe) {

?????????? throw new SearchException(pe.getMessage());

?????? }

??? }

?

??? /**

??? ? *

??? ? * 輸出 hits 的結果，從 start 開始到 end ，不包括 end

??? ? *

??? ? * @param hits

??? ? * @param start

??? ? * @param end

??? ? * @throws SearchException

??? ? */

??? public static Hits display(Hits hits, int start, int end) throws SearchException {

?????? try {

?????????? while (start < end) {

????????????? Document doc = hits.doc(start);

????????????? String path = doc.get( "path" );

????????????? if (path != null ) {

????????????????? System. out .println((start + 1) + "- " + path);

????????????? } else {

????????????????? System. out .println((start + 1) + "- " + "No such path" );

????????????? }

????????????? start++;

?????????? }

?????? } catch (IOException ioe) {

?????????? throw new SearchException(ioe.getMessage());

?????? }

?????? return hits;

??? }

main 方法

??? /**

??? ? * @param args

??? ? */

??? public static void main(String[] args) throws Exception {

?

?????? String field = "contents" ;

?????? String index = "./index" ;

?????? final int rows_per_page = 2;

?????? final char NO = 'n' ;

?

?????? SearchJavaFiles sjf = new SearchJavaFiles( new IndexSearcher(IndexReader.open(index)));

?????? sjf.setParser(field, new StandardAnalyzer());

?????? BufferedReader in = new BufferedReader( new InputStreamReader(System. in , "UTF-8" ));

?

?????? while ( true ) {

?????????? System. out .println( "Query: " );

?????????? String line = in.readLine();

?????????? if (line == null || line.length() < 2) {

????????????? System. out .println( "eixt query" );

????????????? break ;

?????????? }

?????????? Hits hits = sjf.serach(line);

?????????? System. out .println( "searching for " + line + " Result is " );

?

?????????? int len = hits.length();

?????????? int i = 0;

?????????? if (len > 0)

????????????? while ( true ) {

????????????????? if (i + rows_per_page >= len) {

???????????????????? SearchJavaFiles.display(hits, i, len);

???????????????????? break ;

????????????????? } else {

???????????????????? SearchJavaFiles.display(hits, i, i += rows_per_page);

???????????????????? System. out .println( "more y/n?" );

???????????????????? line = in.readLine();

???????????????????? if (line.length() < 1 || line.charAt(0) == NO)

???????????????????????? break ;

????????????????? }

????????????? }

?????????? else

????????????? System. out .println( "not found" );

?????? }

??? }

}

?

SearchException.java

package powerwind;

?

public class SearchException extends Exception {

?

??? public SearchException(String message) {

?????? super ( "Throw SearchException while searching files: " + message);

??? }

?

}

?

完善設想：

1 、文件格式：

能夠處理 Zip 文件 Jar 文件，索引里面的 java 源文件。

通過反射機制索引 class 類文件。

2 、輸入輸出：

除控制臺輸入輸出外，還可以選擇從文件讀取查詢關鍵字，輸出查詢結果到文件。

3 、用戶界面：

圖形界面操作，雙擊查詢結果的某條記錄可以打開相應文件。

4 、性能方面

索引文件時，用緩存和多線程處理

?

?