Clipping pdf with itextsharp 5 results in a corrupt PDF

Question

I want to read a PDF file using itextsharp 5, and create a new PDF file with the left half of each page of the original PDF, discarding the right half. I wrote the following code, but the result is corrupt and cannot be opened with Acrobat Reader (opening it with Chrome works though). How can I make a PDF that works with Acrobat Reader?

private byte[] ObtenerMitadPdf(byte[] Contenido)
    {
        using (var pdfReader = new PdfReader(Contenido))
        {
            using (var memoryStream = new MemoryStream())
            {
                using (var documento = new Document())
                using (var pdfWriter = PdfWriter.GetInstance(documento, memoryStream))
                {
                    pdfWriter.CloseStream = false;
                    documento.Open();
                    for (int i = 1; i <= pdfReader.NumberOfPages; i++)
                    {
                        var pagina = pdfWriter.GetImportedPage(pdfReader, i);
                        var tamañoPagina = pdfReader.GetPageSizeWithRotation(i);
                        var nuevoTamaño = new Rectangle(tamañoPagina.Left, tamañoPagina.Bottom,
                            tamañoPagina.Left + (tamañoPagina.Width / 2), tamañoPagina.Top, tamañoPagina.Rotation);
                        if (tamañoPagina.Rotation == 90)
                        {
                            pdfWriter.DirectContent.AddTemplate(pagina, 0, -1, 1, 0, 0, tamañoPagina.Height);
                        }
                        else
                        {
                            pdfWriter.DirectContent.AddTemplate(pagina, 0, 0);
                        }
                        documento.SetPageSize(nuevoTamaño);
                        documento.NewPage();
                    }
                }
                return memoryStream.ToArray();
            }
        }
    }

Please don't put var pdfWriter = PdfWriter.GetInstance(documento, memoryStream) into a using block. Closing the Document will implicitly close the PdfWriter. Other than that your approach is unnecessarily complicated, you can simply change the crop boxes of the PdfReader and write that version using a ` PdfStamper`. — mkl
How do I set the crop box with PdfReader? Does it take rotation into account? — Alejandro

Alejandro Alejandro · Accepted Answer · 2018-09-26T20:56:08

I finally managed to get it working!

private byte[] ObtenerMitadPdf(byte[] Contenido)
    {
        using (var pdfReader = new PdfReader(Contenido))
        {
            using (var memoryStream = new MemoryStream())
            {
                using (var stamper = new PdfStamper(pdfReader, memoryStream))
                {
                    for (int i = 1; i <= pdfReader.NumberOfPages; i++)
                    {
                        var pagina = pdfReader.GetPageN(i);
                        var tamañoPagina = pdfReader.GetPageSizeWithRotation(i);
                        var mediaBox = new PdfRectangle(tamañoPagina.Left, tamañoPagina.Bottom,
                            tamañoPagina.Left + (tamañoPagina.Width / 2), tamañoPagina.Top, tamañoPagina.Rotation);
                        pagina.Put(PdfName.MEDIABOX, mediaBox);
                        stamper.MarkUsed(pagina);
                    }
                }
                return memoryStream.ToArray();                   
            }
        }
    }

Clipping pdf with itextsharp 5 results in a corrupt PDF

1 Answers