［原创］java解析非结构化文本 doc xls pdf ppt xml html

［原创］java解析非结构化文本 doc xls pdf ppt xml html

    在开發中，經常遇到需要用去解析各类非结构化文本，像doc,xls,pdf,ppt,xml,html.
    本人在开發站内搜索时，需要加入对各类文件类型的支持，以方便建立索引。解析各类文档调用了几個开源的包。像dom4j-1.6.1.jar，FontBox-0.1.0-dev.jar，htmllexer.jar，htmlparser.jar，PDFBox-0.7.3.jar，poi-3.5-FINAL-20090928.jar，poi-scratchpad-3.5-FINAL-20090928.jar。这些开源的包可以讓我們很方便去解析各类非结构化文本。

jar包的下载地址：  http://www.ziddu.com/download/7017588/devlib.rar.html

代码如下：

package  com.ducklyl;

import  java.io.File;
import  java.io.FileInputStream;
import  java.util.Iterator;
import  org.apache.poi.hslf.model.Slide;
import  org.apache.poi.hslf.model.TextRun;
import  org.apache.poi.hslf.usermodel.SlideShow;
import  org.apache.poi.hssf.usermodel.HSSFCell;
import  org.apache.poi.hssf.usermodel.HSSFRow;
import  org.apache.poi.hssf.usermodel.HSSFSheet;
import  org.apache.poi.hssf.usermodel.HSSFWorkbook;
import  org.apache.poi.hwpf.HWPFDocument;
import  org.apache.poi.hwpf.usermodel.Paragraph;
import  org.apache.poi.hwpf.usermodel.Range;
import  org.dom4j.Document;
import  org.dom4j.Element;
import  org.dom4j.io.SAXReader;
import  org.htmlparser.Parser;
import  org.htmlparser.filters. * ;

import  org.htmlparser. * ;
import  org.htmlparser.nodes.TextNode;
import  org.htmlparser.util. * ;

import  org.pdfbox.pdfparser.PDFParser;
import  org.pdfbox.pdmodel.PDDocument;
import  org.pdfbox.util.PDFTextStripper;



public   class  HandleFile {
     public   static   void  main(String args[]){
        String str = " e:\\test.HTML " ;
        System.out.println(handleFile(str));
    }

     public   static  String handleFile(String filename){
        String result = "" ;
        String fileType = filename.substring(filename.lastIndexOf( " . " ) + 1 , filename.length());
         if (fileType.equalsIgnoreCase( " pdf " ))
            result = handlePdf(filename);
         else   if (fileType.equalsIgnoreCase( " xls " ))
            result = handleExcel(filename);
         else   if (fileType.equalsIgnoreCase( " doc " ))
            result = handleDoc(filename);
         else   if (fileType.equalsIgnoreCase( " xml " ))
            result = handleXml(filename);
         else   if (fileType.equalsIgnoreCase( " ppt " ))
            result = handlePPT(filename);
         else   if (fileType.equalsIgnoreCase( " htm " ) || fileType.equalsIgnoreCase( " html " ))
            result = handleHtml(filename);
         return  result;
    }
/**
 * 解析HTML
 *  @param  filename
 *  @return
  */
     public   static  String handleHtml(String filename){
        String content = "" ;
         try {
            File file = new  File(filename);
             if ( ! file.exists())  return  content;
            
            Parser parser = new  Parser(filename);
            parser.setEncoding( " UTF-8 " );
            NodeFilter textFilter = new  NodeClassFilter(TextNode. class );
            NodeList nodes = parser.extractAllNodesThatMatch(textFilter);
             for ( int  i = 0 ;i < nodes.size();i ++ ){
                TextNode textnode = (TextNode)nodes.elementAt(i);
                String line = textnode.toPlainTextString().trim();
                 if (line.equals( "" ))  continue ;
                content = content + line;
            }
        } catch (Exception e){
            e.printStackTrace();
        }
         return  content;
    }
     /**
     * 解析PPT
     *  @param  filename
     *  @return
      */
     public   static  String handlePPT(String filename){
        StringBuffer content  =   new  StringBuffer( "" );
         try {
            File file = new  File(filename);
             if ( ! file.exists()) {
                 return  content.toString();
            }
            FileInputStream instream = new  FileInputStream(file);
            SlideShow ppt  =   new  SlideShow(instream);
            Slide[] slides  =  ppt.getSlides();
             for ( int  i = 0 ;i < slides.length;i ++ ){
                TextRun[] t  =  slides[i].getTextRuns(); // 为了取得幻灯片的文字内容，建立TextRun
                 for ( int  j = 0 ;j < t.length;j ++ ){
                    content.append(t[j].getText()); // 这里会将文字内容加到content中去
                }
                content.append(slides[i].getTitle());
            }
        } catch (Exception e){
            e.printStackTrace();
        }
         return  content.toString();
    }
     /**
     * 解析XML
     *  @param  filename
     *  @return
      */
     public   static  String handleXml(String filename){
        String content = "" ,value = "" ,text = "" ;
         try {
            File file = new  File(filename);
             if ( ! file.exists()) {
                 return  content;
            }
              SAXReader saxReader  =   new  SAXReader();
              Document document  =  saxReader.read(file);
              Element root  =  document.getRootElement() ;
              
              Iterator iter = root.elementIterator() ;
                while (iter.hasNext()){
                  Element element = (Element)iter.next();
                  value = element.getStringValue();
                   if ( ! value.trim().equals( "" ))    content = content + value;
               }
        } catch (Exception e){
                e.printStackTrace();
        }
         return  content;
    }
     /**
     * 解析DOC
     *  @param  filename
     *  @return
      */
     public   static  String handleDoc(String filename){
        String content = "" ;
         try {
            File file = new  File(filename);
             if ( ! file.exists()) {
                 return  content;
            }
            FileInputStream instream = new  FileInputStream(file);
            HWPFDocument doc = new  HWPFDocument(instream);
            Range range = doc.getRange();
            String text = range.text();
             for ( int  i = 0 ;i < range.numParagraphs();i ++ ){
                Paragraph p = range.getParagraph(i);
                content = content + p.text().trim() + " \n " ;
            }
        } catch (Exception e){
            e.printStackTrace();
        }
         return  content;
    }
     /**
     * 解析PDF
     *  @param  filename
     *  @return
      */
     public   static  String handlePdf(String filename){
        String contenttxt = "" ;
         try {
            File file = new  File(filename);
             if ( ! file.exists()){
                 return  contenttxt;
            }
            FileInputStream instream = new  FileInputStream(file);
            PDFParser parser = new  PDFParser(instream);
            parser.parse();
            PDDocument pdfdocument = parser.getPDDocument();
            PDFTextStripper pdfstripper = new  PDFTextStripper();
            contenttxt = pdfstripper.getText(pdfdocument);
        } catch (Exception e){
            e.printStackTrace();
        }
         return  contenttxt;
    }
     /**
     * 解析EXCEL
     *  @param  filename
     *  @return
      */
     public   static  String handleExcel(String filename){
        String content = "" ;
         try {
            File file = new  File(filename);
             if ( ! file.exists()) {
                 return  content;
            }
            HSSFWorkbook workbook = new  HSSFWorkbook( new  FileInputStream(file));
            HSSFSheet sheet = workbook.getSheetAt( 0 );
            
             for ( int  i = 0 ;i < workbook.getNumberOfSheets();i ++ ){
                sheet = workbook.getSheetAt(i);
                 if (sheet != null ){
                     for ( int  m = 0 ;m < sheet.getLastRowNum();m ++ ){
                        HSSFRow row = sheet.getRow(m);
                         if (row == null )  break ;
                        
                         for ( int  n = 0 ;n < row.getLastCellNum();n ++ ){
                            HSSFCell cell = row.getCell(n);
                             if (cell == null )  break ;
                             int  type = cell.getCellType();
                             switch (type){
                                 case   0 :
                                    content = content + cell.getNumericCellValue();
                                     break ;
                                 case   1 :
                                    content = content + cell.getStringCellValue();
                                     break ;
                                 case   3 :
                                     break ;
                                 default :
                                    ;
                            }
                        }
                        content = content + " \n " ;
                    }
                }
                content = content + " \n " ;
            }

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

不想拷贝的朋友可以直接下载源代码： http://www.ziddu.com/download/7017614/src.txt.html

以上代码比较简单，就不作说明，希望能幫到需要用的朋友。当然上面只是一個简单的例子，如果要具体应用，大家可以自己再改写。如果你有其它的想法，欢迎分享你的精彩想法。


转载请注明出处