I am working in a legacy java application, where Ghostscript is being used from the command line to repair pdfs:

private static File repairLinuxGhostScript(String filePath) throws IOException{
    PerfLog.snapShot();

    tempFile = createTempFileInTeradactorTempDirectory(filePath.substring(0,filePath.length()-4) + "-repaired", ".pdf");

    CommandLine cmd = new CommandLine("gs");
    cmd.addArgument("-o");
    cmd.addArgument(tempFile.getAbsolutePath());
    cmd.addArgument("-dPDFSETTINGS=/prepress"); //selects output similar to Acrobat Distiller "Prepress Optimized" (up to version X) setting. https://ghostscript.com/docs/9.54.0/VectorDevices.htm
    cmd.addArgument("-sDEVICE=pdfwrite"); //The pdfwrite, ps2write and eps2write devices create PDF or PostScript files whose visual appearance should match, as closely as possible, the appearance of the original input file https://ghostscript.com/docs/9.54.0/VectorDevices.htm
    cmd.addArgument("-dSubsetFonts=false"); //
    cmd.addArgument(filePath);

    DefaultExecutor exec = new DefaultExecutor();
    ByteArrayOutputStream outputCapturer = new ByteArrayOutputStream();
    exec.setStreamHandler(new PumpStreamHandler(outputCapturer, System.err));
    System.out.println(cmd.toString());
    int exitValue = exec.execute(cmd);
    LOG.info(outputCapturer.toByteArray());
    if(exitValue != 0) {
        throw new ProcessorServiceException("Ghostscript gs exited with error code " + exitValue);
    }
    PerfLog.log(PerfLog.Phase.IMPORT, PerfLog.SubPhase.REPAIR,"PDF Repair","GhostScript","PdfRepairer");
    return tempFile;
}

We want to get rid of temporary files so I rewrote the method to this:

public static byte[] repairPDF(byte[] inputPDF) throws IOException {
    // Prepare Ghostscript command
    CommandLine cmd = new CommandLine("gs");
    cmd.addArgument("-dPDFSETTINGS=/prepress");
    cmd.addArgument("-sDEVICE=pdfwrite");
    cmd.addArgument("-dSubsetFonts=false");
    cmd.addArgument("-sOutputFile=%stdout"); // Output to stdout

    cmd.addArgument("-");

    // Set up input and output streams
    ByteArrayInputStream inputStream = new ByteArrayInputStream(inputPDF);
    ByteArrayOutputStream outputStream = new ByteArrayOutputStream();
    ByteArrayOutputStream errorStream = new ByteArrayOutputStream();
    PumpStreamHandler streamHandler = new PumpStreamHandler(outputStream, errorStream, inputStream);

    DefaultExecutor executor = new DefaultExecutor();
    executor.setStreamHandler(streamHandler);

    // Execute Ghostscript command
    try {
        executor.execute(cmd); // Send input PDF via stdin
    } catch (ExecuteException e) {
        System.err.println("Execution failed: " + e.getMessage());
        throw new IOException("Ghostscript execution failed", e);
    } catch (IOException e) {
        System.err.println("IO Error: " + e.getMessage());
        throw new IOException("Error processing PDF", e);
    }

    // Handle errors if any
    String errorOutput = errorStream.toString();
    if (!errorOutput.isEmpty()) {
        System.err.println("Ghostscript Error: " + errorOutput);
    }

    // Get the result as a byte array
    return outputStream.toByteArray();
}

However, both methods return different results. The pdf from method 1 is not equal to the pdf of method 2 (some text disappears, some images,…).

I also read somewhere here on stack overflow that Ghostscript needs to read the files from disk anyway, so it doesn’t really make a difference whether I would remove temporary files on my end, since they get created anyway (source: /a/49880072/12171869).

I am wondering: is there a way to achieve the same result as the initial method without creating temporary files? I have taken a look at PDF box but haven’t gotten the desired result (note that I am working with pdf box 1.8.11, since legacy application)

Thank you.