如何分割pdf文件的结果在java pdfbox

Question

我有一个pdf文件，包含60页。在每一页中，我都有唯一和重复的发票编号。我使用Apache PDFBOX。

import java.io.*;
import org.apache.pdfbox.pdmodel.*;
import org.apache.pdfbox.util.*;
import java.util.regex.*;
public class PDFtest1 {
public static void main(String[] args){
PDDocument pd;
try {
     File input = new File("G:\Sales.pdf");
     // StringBuilder to store the extracted text
     StringBuilder sb = new StringBuilder();
     pd = PDDocument.load(input);
     PDFTextStripper stripper = new PDFTextStripper();
     // Add text to the StringBuilder from the PDF
     sb.append(stripper.getText(pd));

     Pattern p = Pattern.compile("Invoice No.\s\w\d\d\d\d\d\d\d\d\d\d");
     // Matcher refers to the actual text where the pattern will be found
     Matcher m = p.matcher(sb);
     while (m.find()){
         // group() method refers to the next number that follows the pattern we have specified.
         System.out.println(m.group());
     }
     if (pd != null) {
         pd.close();
     }
   } catch (Exception e){
     e.printStackTrace();
    }
 }
 }  

我能够使用java regex读取所有发票号码。最后，结果如下

run:
Invoice No. D0000003010
Invoice No. D0000003011
Invoice No. D0000003011
Invoice No. D0000003011
Invoice No. D0000003011
Invoice No. D0000003012
Invoice No. D0000003012
Invoice No. D0000003012
Invoice No. D0000003013
Invoice No. D0000003013
Invoice No. D0000003014
Invoice No. D0000003014
Invoice No. D0000003015
Invoice No. D0000003016

我需要根据第5号发票拆分pdf。例如:发票编号:D0000003011，所有pdf页面应该合并为一个pdf等等。

Answer 1

public static void main(String[] args) throws IOException, COSVisitorException
{
    File input = new File("G:\Sales.pdf");
    PDDocument outputDocument = null;
    PDDocument inputDocument = PDDocument.loadNonSeq(input, null);
    PDFTextStripper stripper = new PDFTextStripper();
    String currentNo = null;
    for (int page = 1; page <= inputDocument.getNumberOfPages(); ++page)
    {
        stripper.setStartPage(page);
        stripper.setEndPage(page);
        String text = stripper.getText(inputDocument);
        Pattern p = Pattern.compile("Invoice No.(\s\w\d\d\d\d\d\d\d\d\d\d)");
        // Matcher refers to the actual text where the pattern will be found
        Matcher m = p.matcher(text);
        String no = null;
        if (m.find())
        {
            no = m.group(1);
        }
        System.out.println("page: " + page + ", value: " + no);
        PDPage pdPage = (PDPage) inputDocument.getDocumentCatalog().getAllPages().get(page - 1);
        if (no != null && !no.equals(currentNo))
        {
            saveCloseCurrent(currentNo, outputDocument);
            // create new document
            outputDocument = new PDDocument();
            currentNo = no;
        }
        if (no == null && currentNo == null)
        {
            System.out.println ("header page ??? " + page + " skipped");
            continue;
        }
        // append page to current document
        outputDocument.importPage(pdPage);
    }
    saveCloseCurrent(currentNo, outputDocument);
    inputDocument.close();
}
private static void saveCloseCurrent(String currentNo, PDDocument outputDocument)
        throws IOException, COSVisitorException
{
    // save to new output file
    if (currentNo != null)
    {
        // save document into file
        File f = new File(currentNo + ".pdf");
        if (f.exists())
        {
            System.err.println("File " + f + " exists?!");
            System.exit(-1);
        }
        outputDocument.save(f);
        outputDocument.close();
    }
}

注意:

• 这还没有测试过你的文件(因为我没有);
• 代码假设相同的发票号码总是在一起;
• 你的正则表达式被稍微改变了;
• 确保第一个和最后一个PDF文件是正确的，并随机检查几个，如果可用，使用不同的查看器;
• 验证文件总数是否符合预期;
• 所有文件的汇总大小会比源文件大，这是由于字体资源的原因;
• 使用1.8.10版本。不要同时使用PDFBox 0.7.3.jar !
• 错误处理非常基本，你需要改变它;

更新19.8.2015:

• 它现在支持没有发票号码的页面，这些将被附加。