アットウィキロゴ

p5449

package pdf;
 
import java.io.FileInputStream;
import java.io.InputStream;
import org.apache.tika.parser.pdf.PDFParser;
import org.apache.tika.metadata.Metadata;
import org.apache.tika.parser.ParseContext;
import org.apache.tika.sax.BodyContentHandler;
import org.xml.sax.ContentHandler;
 
 
import java.io.File;
import tool.*;
 
public class pro {
 
    String[] data=new String[5000];
    int s,datanumber;
 
 
    public static void main(String[] args) {
       pro test=new pro();
    }
 
    pro(){
 
        File f=new File("data");
       String[] ff=f.list();
 
 
       int tr;
       for(tr=0;tr<ff.length;tr++){
          String str="";
 
       InputStream is = null;
    try {
 
      is = new FileInputStream("data/"+ff[tr]);
      ContentHandler contenthandler = new BodyContentHandler();
      Metadata metadata = new Metadata();
      PDFParser pdfparser = new PDFParser();
      pdfparser.parse(is, contenthandler, metadata, new ParseContext());
      str=contenthandler.toString();
 
 
    }catch (Exception e) {e.printStackTrace();}
 
 
 
 
    String[] x=str.split("\n");
 
    datanumber=x.length;
 
    for(s=1;s<datanumber+1;s++){
    data[s]=x[s-1];
    }
 
    String nn=ff[tr].replace("pdf","");
 
 
    writefile sub2=new writefile();
    sub2.data=data;
    sub2.datanumber=datanumber;
    sub2.makedata("gdata/"+nn+"txt");
 
 
 
 
 
 
 
}
}
 
最終更新:2017年01月01日 02:34