索引��d��器_FSDiractory模式

HanLab — Sat, 31 May 2008 21:11:00 GMT

package indexer;
//package ch2.lucenedemo.process;

import java.io.File;

import java.io.IOException;
import java.util.ArrayList;
import java.util.logging.Level;
import java.util.logging.Logger;
import jeasy.analysis.MMAnalyzer;

import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.FSDirectory;
import parameters.Param;
import pretreat.FileControl;

public class IndexOnFS implements IIndexTool{
    // 成员变量存储创徏的烦引文件存攄��位置
    private String INDEX_STORE_PATH = Param.INDEX_STORE_PATH;
        //建立索引的目标文�?/span>
        private String INDEX_WANTED_PATH = "e:\\";
        //目录数组��L��
        private int NumOfDir = 0;
        //存放根目录下的所有子目录
        private ArrayList<String> DirList = new ArrayList<String>();
        //地址映射
        private Directory dir = null;

        private IndexWriter writer;

        public IndexOnFS(String path){
                try {
                    dir = FSDirectory.getDirectory(INDEX_STORE_PATH);
                } catch (IOException ex) {
                    Logger.getLogger(IndexOnFS.class.getName()).log(Level.SEVERE, null, ex);
                }
                INDEX_WANTED_PATH = path;
                makeSegments();
                searchDirectorys(path);
        }

        //建立索引之前遍历所有目录�ƈ存放�Q�这是�ؓ了迎合IndexWriter的同步机�?/span>
        public void searchDirectorys(String rootDir){

                File rootfile = new File(rootDir);
                File[] files = rootfile.listFiles();
                if(files!=null)
                for (int i = 0; i < files.length; i++){
                    if(files[i].isDirectory()){
                       DirList.add(files[i].getPath());
                       searchDirectorys(files[i].getPath());
                    }
                }
        }
        public void printAllDirectorys(){
                for(int i = 0;i<DirList.size();i++)
                       System.out.println(DirList.get(i));
        }
        public void createIndexs() {
                createIndex(INDEX_WANTED_PATH);
                for(int k = 0;k<DirList.size();k++)
                    createIndex(DirList.get(k));
        }
        public Document preIndexWrite(File file){
                // 创徏一个新的Document
            Document doc = new Document();
            // 文�g名对应的Field
            Field field = new Field("filename", file.getName(),
                                 Field.Store.YES, Field.Index.TOKENIZED);
        doc.add(field);
        // 文�g内容对应的Filed
        field = new Field("content", FileControl.fileToString(file),//转到控制�?/span>
                   Field.Store.NO, Field.Index.TOKENIZED);
        doc.add(field);
                //文�g路径对应的Filed
                field = new Field("filepath", file.getPath(),
                           Field.Store.YES, Field.Index.TOKENIZED);
        doc.add(field);

                return doc;
        }
    /*单目录创建烦�?/span>*/
    public void createIndex(String inputDir) {
        try {
            /*MMAnalyzer作�ؓ分词工具创徏一个IndexWriter*/
                writer = new IndexWriter(dir,new MMAnalyzer(), false); /*�W�一�ơ创建烦引时为true*/
            File filesDir = new File(inputDir);
            /*取得所有需要徏立烦引的文�g数组*/
            File[] files = filesDir.listFiles();
            /*遍历数组*/
                        if(files!=null)
            for (int i = 0; i < files.length; i++) {
                              /*判断是否为文�?/span>*/
                              if(files[i].isFile()){
                    /*把Document加入IndexWriter*/
                    writer.addDocument(preIndexWrite(files[i]));
                                        System.out.println( files[i].getPath());
                        }
                                    }
                        writer.optimize(); /*索引优化*/

        } catch (Exception e) { e.printStackTrace(); }

                finally{
                    try{writer.close();
                    }catch(Exception ee){ ee.printStackTrace(); }
                }

    }
        //初始化空索引�?/span>
        public void makeSegments(){
            if(new File(INDEX_STORE_PATH).list().length==0){
                try {
                    IndexWriter iw = new IndexWriter(dir, new MMAnalyzer(), true);
                    writer.addDocument(preIndexWrite(new File(Param.INITFILE_PATH)));
                } catch (Exception ex) { ex.printStackTrace(); }

                finally{
                      try{writer.close();
                      }catch(Exception ee){ ee.printStackTrace(); }
                  }
            }
        }
        public ArrayList getDirs(){
            return this.DirList;
        }

        public void startIndex() {
            makeSegments();
            createIndexs();
        }

    public static void main(String[] args) {
        IndexOnFS processor = new IndexOnFS("e:\\毕业论文");
                //processor.searchDirectorys("e:\\1");
                processor.startIndex();
    }

}

HanLab 2008-06-01 05:11 发表评论

Lucene源码�l�构

HanLab — Sun, 06 Apr 2008 09:02:00 GMT

�q�是我做毕业设计时画的，感觉�q�可以就放过来共享，有什么不��的地方�Q�请多指炏V�?br />

看Lucene代码也算是中享受�Q�根据下囑֏�以先看看关键�cȝ��代码�?br /> Lucene2.3.1发布不久�Q�源代码下蝲地址�Q?a >http://apache.mirror.phpchina.com/lucene/java/

org.apache.Lucene.search/	搜烦入口
org.apache.Lucene.index/	索引入口
org.apache.Lucene.analysis/	语言分析�?/span>
org.apache.Lucene.queryParser/	查询分析�?/span>
org.apache.Lucene.document/	存储�l�构
org.apache.Lucene.store/	底层IO/存储�l�构
org.apache.Lucene.util/	一些公用的数据�l�构

HanLab 2008-04-06 17:02 发表评论

97在线超碰,国产精品一区二区a,91国在线观看

索引��d��器_FSDiractory模式

Lucene源码�l�构