我們自己的lusecn搜索引擎

日誌 > 技術交流

轉載到我空間 複製本文網址 隱藏簽名檔 小字體 選用該信紙

上一篇|下一篇|返回日誌列表

設置置頂 | 編輯 | 刪除

我們自己的lusecn搜索引擎

發表於：2008年2月26日 14時32分36秒閱讀(5)評論(0)本文鏈接：http://user.qzone.qq.com/592433424/blog/1204007556

我們自己的lusecn搜索引擎

package com.shzscq.searchpatent.search;
import java.util.StringTokenizer;
import org.apache.lucene.search.*;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.queryParser.QueryParser;
import org.apache.lucene.queryParser.ParseException;
import org.apache.lucene.document.Document;

/**
* Created by IntelliJ IDEA.
* User: qiao
* Date: 2008-1-23
* Time: 15:51:29
* To change this template use File | Settings | File Templates.
*/
public class Search
{
   private int iHits = 0;
   private String sMultiHits = "";
   private String sDatabase = "";
   private String sQuery = "";
   private String sSortField = "";
   private boolean bSort = false;
   private String sResult = "";
   private String sPath = "";
   private String sFieldList = "";
   public int getHits()
   {
        return this.iHits;
   }
   public String getArrHits()
   {
       return this.sMultiHits;
   }
   public String getResult()
   {
       return this.sResult;
   }
   public void initial(String sPath,String sIndexDatabase,String sQueryInput,String sSortField,boolean bSort,String sFieldList)
   {
       this.sPath = sPath;
       this.sDatabase =  sIndexDatabase;
       this.sQuery = sQueryInput;
       this.sSortField = sSortField;
       this.bSort = bSort;
       this.sFieldList = sFieldList;
   }
   public void searchAllWithResult(int iPage, int iPageSize) {
       String sErrorMessage = "";
       boolean bError = false;
       this.iHits = 0;
       StringTokenizer sDbToken = new StringTokenizer(this.sDatabase, ";");        //錕矯碉拷錕斤拷錕叫碉拷錕斤拷菘錕?
       int iDbNum = sDbToken.countTokens();                        //錕斤拷菘錕斤拷錕僥?
       IndexSearcher[] _arrSearchers = new IndexSearcher[iDbNum];
       Query _query = null;                     //the Query created by the QueryParser
       Hits _hits = null;
       int iDocIndex = (iPage-1) * iPageSize;
       try {
           for (int i = 0; i < iDbNum; i++) {
               String sDbName = sDbToken.nextToken();
               String sIndexName = this.sPath + "//" + sDbName;
               _arrSearchers[i] = new IndexSearcher(IndexReader.open(sIndexName));     //create an indexSearcher for our page
           }
       } catch (Exception e) {                         //any error that happens is probably due
           //to a permission problem or non-existant
           //or otherwise corrupt index
           bError = true;                                  //don't do anything up to the footer
           sErrorMessage = "ERROR opening the Index - contact sysadmin!" + e.getMessage();
           System.out.println(sErrorMessage);
       }
       if (this.sQuery == null) {
           bError = true;
           sErrorMessage = "Query string is null!";
       }
       if (bError == false) {
           try {
               Analyzer _analyzer = new StandardAnalyzer();        //construct our usual analyzer
               QueryParser _parser = new QueryParser("AB", _analyzer);
               _query = _parser.parse(this.sQuery);
           } catch (ParseException e) {
               bError = true;
               sErrorMessage = "Error While parsing query:" + e.getMessage();
           }
       }
       if (bError == false) {
           try {
               MultiSearcher _multiSearcher = new MultiSearcher(_arrSearchers);
               if (sSortField.equalsIgnoreCase("")) {
                   _hits = _multiSearcher.search(_query);
               } else {
                   Sort _st = new Sort();
                   _st.setSort(sSortField, bSort);   //錕斤拷錕斤拷錕斤拷錕斤拷
                   _hits = _multiSearcher.search(_query, _st);
               }
               this.iHits = _hits.length();
               if (_hits.length() == 0) {
                   bError = true;
                   sErrorMessage = "I'm sorry I couldn't find what you were looking for.";
               } else {
                   int iBegin = (iPage - 1) * iPageSize;
                   int iEnd = 0;
                   if (iPage * iPageSize > this.iHits) {
                       iEnd = _hits.length();
                   } else {
                       iEnd = iPage * iPageSize;
                   }
                   String[] sArrFields = this.sFieldList.split(";");
                   this.sResult = "<docs>";
                   for (int i = iBegin; i < iEnd; i++) {
                       Document doc = _hits.doc(i);
                       this.sResult = this.sResult + "<doc>";
                       sResult = sResult + "<id>" + (iDocIndex+1)  + "</id>";
                       for (int j = 0; j < sArrFields.length; j++) {
                           this.sResult = this.sResult + "<fld name=/"" + sArrFields[j] + "/">";
                           this.sResult = this.sResult + doc.get(sArrFields[j]);
                           this.sResult = this.sResult + "</fld>";
                       }
                       this.sResult = this.sResult + "</doc>";
                       iDocIndex++;
                   }
               }
               _multiSearcher.close();
           } catch (Exception ex) {
               System.out.println(ex.getMessage());
               bError = true;
               sErrorMessage = "I'm sorry I couldn't find what you were looking for.";
           }
       }
       for(int i=0;i<_arrSearchers.length;i++)
       {
          try
          {
           _arrSearchers[i].close();
           _arrSearchers[i] = null;
          }
          catch(Exception ex)
          {
              System.out.print(ex.getMessage());
          }
       }
       _hits = null;
   }
   public void searchEvyWithResult(int iPage,int iPageSize)
   {
       this.iHits = 0;
       this.sMultiHits = "";
       StringTokenizer sDbToken = new StringTokenizer(this.sDatabase, ";");        //錕矯碉拷錕斤拷錕叫碉拷錕斤拷菘錕?
       int iDbNum = sDbToken.countTokens();
       int iBegin = (iPage-1) * iPageSize;
       int iEnd = iPage * iPageSize;
       int iDocIndex = (iPage-1) * iPageSize;
       String [] sArrFields = this.sFieldList.split(";");
       this.sResult = "<docs>";
       for(int i=0;i<iDbNum;i++)
       {
           String sDbName = sDbToken.nextToken();
           String sIndexName = this.sPath + "//" + sDbName;
           IndexReader _reader = null;
           Searcher _searcher = null;
           Analyzer _analyzer = new StandardAnalyzer();
           QueryParser _parser = new QueryParser("AB", _analyzer);
           try
           {
               _reader = IndexReader.open(sIndexName);
               _searcher = new IndexSearcher(_reader);
                Query _query = _parser.parse(sQuery);
                Hits _hits = _searcher.search(_query);
                int iLen = _hits.length();
                if (iLen != 0)
                {
                    int iBIn = iBegin;
                    int iEIn = iEnd;
                    if (iBegin == 0)
                    {
                        if (iLen <= iEnd)
                        {
                            iEIn = iLen;
                            iBegin = iLen;
                        }
                    } else
                    {
                        if (iBegin >= this.iHits && iBegin < (this.iHits + iLen))
                        {
                            iBIn = (iBegin - this.iHits);
                            if ((this.iHits + iLen) <= iEnd)
                            {
                               iEIn = iLen;
                               iBegin = this.iHits + iLen;
                            }
                            else
                            {
                                iEIn = iEnd - this.iHits;
                            }
                        }
                    }
                   if(iLen > iBIn)
                   {
                   for (int j = iBIn; j < iEIn; j++) {
                            Document doc = _hits.doc(j);
                            sResult = sResult + "<doc>";
                            sResult = sResult + "<id>" + (iDocIndex + 1)  + "</id>";
                            for (int m = 0; m < sArrFields.length; m++) {
                                sResult = sResult + "<fld name=/"" + sArrFields[m] + "/">";
                                sResult = sResult + doc.get(sArrFields[m]);
                                sResult = sResult + "</fld>";
                            }
                            sResult = sResult + "</doc>";
                            iDocIndex ++ ;
                        }
                   }
                }
                this.iHits = this.iHits + iLen;
                this.sMultiHits = this.sMultiHits + ";" + iLen;
                _searcher.close();
                _reader.close();
                _searcher = null;
                _reader = null;
           }
           catch(Exception ex)
           {
               System.out.println(ex.getMessage());
           }

       }
       this.sMultiHits = this.sMultiHits.substring(1);
       this.sResult = this.sResult + "</docs>";
   }
   public void searchEvyWithoutResult()
   {
       this.iHits = 0;
       this.sMultiHits = "";
       StringTokenizer sDbToken = new StringTokenizer(this.sDatabase, ";");        //錕矯碉拷錕斤拷錕叫碉拷錕斤拷菘錕?
       int iDbNum = sDbToken.countTokens();
       for(int i=0;i<iDbNum;i++)
       {
           String sDbName = sDbToken.nextToken();
           String sIndexName = this.sPath + "//" + sDbName;
           IndexReader _reader = null;
           Searcher _searcher = null;
           Analyzer _analyzer = new StandardAnalyzer();
           QueryParser _parser = new QueryParser("AB", _analyzer);
           try
           {
               _reader = IndexReader.open(sIndexName);
               _searcher = new IndexSearcher(_reader);
                Query _query = _parser.parse(sQuery);
                Hits _hits = _searcher.search(_query);
                int iLen = _hits.length();
                this.sMultiHits = this.sMultiHits + ";" + iLen;
                this.iHits = this.iHits + iLen;
                _searcher.close();
                _reader.close();
                _searcher = null;
                _reader = null;
           }
           catch(Exception ex)
           {
               System.out.println(ex.getMessage());
           }

       }
       this.sMultiHits = this.sMultiHits.substring(1);
   }
   public void search(int iPage,int iPageSize)
   {
       if(this.sSortField.equalsIgnoreCase(""))
       {
           searchEvyWithResult(iPage,iPageSize);
       }
       else
       {
           searchAllWithResult(iPage,iPageSize);
           searchEvyWithoutResult();
       }
   }
    public static void main(String[] args)
    {
        Search _sch = new Search();
        _sch.initial("D://TestShanghai//server//data","MedicineStandardCN","TI:錕斤拷血壓 OR TI:hypertension","PN",false,"PN;TI;PA");
        _sch.search(1,20);
        //_sch.searchEvyWithoutResult();//(8,100);
        System.out.println(_sch.getHits());
        System.out.println(_sch.getArrHits());
        System.out.print(_sch.getResult());
    }
}

評論列表

聲明：騰訊不會在QQ空間以回覆方式通知用戶中獎，請大家不要相信這些虛假信息以免上當受騙。更多虛假案例請點擊->

發表評論

第一頁 上一頁 下一頁 最後頁

共2篇評論，第1頁/共1頁

全選

批量刪除 取消批量刪除

 

第一頁 上一頁 下一頁 最後頁

共2篇評論，第1頁/共1頁

使用簽名檔 請選擇道具 請選擇道具隱身草彩虹炫天使之愛

<textarea class="content" id="commentEditor" style="BORDER-RIGHT: #ccc 1px solid; BORDER-TOP: #ccc 1px solid; BORDER-LEFT: #ccc 1px solid; COLOR: gray! important; BORDER-BOTTOM: #ccc 1px solid" onfocus="getUBBeditor(this)" rows="13" cols="50" name="content">點擊這裏發表評論</textarea>

溫馨提示：點擊驗證碼輸入框，以獲取驗證碼

請輸入驗證碼：

      悄悄話 | (以小紙條形式發送) 查看今日免費發送數量

上一篇|下一篇|返回日誌列表

loading......