Itext Pdf 可以提取文本，但无法使用所有者密码从 pdf 复制页面？

Question

I'm using itext pdf for the java programming language to extract text from a pdf document. With a PdfReaderContentParser approach, it is possible to extract the desired textual content. But my PdfCopy approach results in an IllegalArgumentException .

This is the example of my PdfReaderContentParser approach:

import java.io.File;
import java.io.FileOutputStream;
import java.io.PrintWriter;

import com.itextpdf.text.pdf.PdfReader;
import com.itextpdf.text.pdf.parser.PdfReaderContentParser;
import com.itextpdf.text.pdf.parser.SimpleTextExtractionStrategy;
import com.itextpdf.text.pdf.parser.TextExtractionStrategy;

public class PdfExtract {

    public PdfExtract()  throws Exception{
        File f = new File( "D:/disertasi orang penting/OSD/disertasi doktor OSD.pdf" );
        PdfReader reader = new PdfReader(f.getPath()); 
        PdfReaderContentParser parser = new PdfReaderContentParser(reader);
        TextExtractionStrategy strategy;
        PrintWriter out2 = new PrintWriter(new FileOutputStream( "D:/disertasi orang penting/OSD/hasil-1.txt" ));
        for(int i = 1; i  <=  reader.getNumberOfPages(); i++) {
            strategy = parser.processContent(i , new SimpleTextExtractionStrategy()); 
            String result = strategy.getResultantText(); 
            out2.println(result);
            out2.flush();
            
        }
        out2.close();
    }
    
    public static void main(String[] args ) throws Exception {
        new PdfExtract();
    }
}

And this is an example of method PdfCopy approach:

import java.io.FileOutputStream;

import com.itextpdf.text.Document;
import com.itextpdf.text.pdf.PdfCopy;
import com.itextpdf.text.pdf.PdfImportedPage;
import com.itextpdf.text.pdf.PdfReader;

public class TestExtractPdf {

    public static void main(String[] args) {
        try {
            String dirHasil = "D:/disertasi orang penting/OSD/hasil/";
            PdfReader reader = new PdfReader("D:/disertasi orang penting/OSD/disertasi doktor OSD.pdf");
            int n = reader.getNumberOfPages();
            System.out.println("Number of pages : " + n);
            
            for (int i = 0; i < n; i++) {
                String outfile = dirHasil + Integer.toString(i + 1) + ".pdf";
                System.out.println("Writing " + outfile);
                Document document = new Document(reader.getPageSizeWithRotation(1));
                PdfCopy copy = new PdfCopy(document, new FileOutputStream(outfile));
                document.open();
                PdfImportedPage page = copy.getImportedPage(reader, i );
                copy.addPage(page);
                document.close();
                copy.close();
            }
        } catch (Exception e) {
            System.out.println("eror");
            e.printStackTrace();
        }
    }
}

and it is raise an eror:

java.lang.IllegalArgumentException: PdfReader not opened with owner password
    at com.itextpdf.text.pdf.PdfReaderInstance.getImportedPage(PdfReaderInstance.java:80)
    at com.itextpdf.text.pdf.PdfCopy.getImportedPageImpl(PdfCopy.java:388)
    at com.itextpdf.text.pdf.PdfCopy.getImportedPage(PdfCopy.java:255)
    at fjr.cpns.kemenkeu.TestExtractPdf.main(TestExtractPdf.java:25)

Answer 1

Itext Pdf can extract text but can not copy page from pdf with owner password?

Indeed, the iText 5 implementation of permissions of encrypted files is a bit weird: for some functionalities (in particular stamping and page copying) the owner password is required in case of encrypted files, for most other functionalities not.

As the PDF permission structure is designed for PDF viewers and editors with a GUI and not for programming libraries, you cannot seriously implement them in iText. The implementation as done could have served to demonstrate to Adobe that iText does respect this part of the PDF references back when the format PDF still was Adobe proprietary.

iText 5 since version 5.0.2 provides a way, though, to override these restrictions, simply initialize

PdfReader.unethicalreading = true;

before your code to make iText code assume you have opened arbitrary encrypted PDFs with the owner password and, therefore, have full permissions.

---

There is an actual error in your code, though:

Itext page numbering is 1-based, ie the first page is numbered 1.

In your working text extraction you respect this and start with 1:

for(int i = 1; i  <=  reader.getNumberOfPages(); i++) {
    strategy = parser.processContent(i , new SimpleTextExtractionStrategy()); 
    ...

In your not-working copying code, though, you don't respect it and start with 0:

for (int i = 0; i < n; i++) {
    ...
    PdfImportedPage page = copy.getImportedPage(reader, i );
    ...

To fix this, start with 1.