Question

I need to post several (read: a lot) PDF files to the web but many of them have hard coded file:// links and links to non-public locations. I need to read through these PDFs and update the links to the proper locations. I’ve started writing an app using itextsharp to read through the directories and files, find the PDFs and iterate through each page. What I need to do next is find the links and then update the incorrect ones.

string path = "c:\\html";
DirectoryInfo rootFolder = new DirectoryInfo(path);

foreach (DirectoryInfo di in rootFolder.GetDirectories())
{
    // get pdf
    foreach (FileInfo pdf in di.GetFiles("*.pdf"))
    {
        string contents = string.Empty;
        Document doc = new Document();
        PdfReader reader = new PdfReader(pdf.FullName);

        using (MemoryStream ms = new MemoryStream())
        {
            PdfWriter writer = PdfWriter.GetInstance(doc, ms);
            doc.Open();

            for (int p = 1; p <= reader.NumberOfPages; p++)
            {
                byte[] bt = reader.GetPageContent(p);

            }
        }
    }
}

Quite frankly, once I get the page content I’m rather lost on this when it comes to iTextSharp. I’ve read through the itextsharp examples on sourceforge, but really didn’t find what I was looking for.

Any help would be greatly appreciated.

Thanks.

Answer 1

This one is a little complicated if you don’t know the internals of the PDF format and iText/iTextSharp’s abstraction/implementation of it. You need to understand how to use PdfDictionary objects and look things up by their PdfName key. Once you get that you can read through the official PDF spec and poke around a document pretty easily. If you do care I’ve included the relevant parts of the PDF spec in parenthesis where applicable.

Anyways, a link within a PDF is stored as an annotation (PDF Ref 12.5). Annotations are page-based so you need to first get each page’s annotation array individually. There’s a bunch of different possible types of annotations so you need to check each one’s SUBTYPE and see if its set to LINK (12.5.6.5). Every link should have an ACTION dictionary associated with it (12.6.2) and you want to check the action’s S key to see what type of action it is. There’s a bunch of possible ones for this, link’s specifically could be internal links or open file links or play sound links or something else (12.6.4.1). You are looking only for links that are of type URI (note the letter I and not the letter L). URI Actions (12.6.4.7) have a URI key that holds the actual address to navigate to. (There’s also an IsMap property for image maps that I can’t actually imagine anyone using.)

Whew. Still reading? Below is a full working VS 2010 C# WinForms app based on my post here targeting iTextSharp 5.1.1.0. This code does two main things: 1) Create a sample PDF with a link in it pointing to Google.com and 2) replaces that link with a link to bing.com. The code should be pretty well commented but feel free to ask any questions that you might have.

using System;
using System.Text;
using System.Windows.Forms;
using iTextSharp.text;
using iTextSharp.text.pdf;
using System.IO;

namespace WindowsFormsApplication1
{
    public partial class Form1 : Form
    {

        //Folder that we are working in
        private static readonly string WorkingFolder = Path.Combine(Environment.GetFolderPath(Environment.SpecialFolder.Desktop), "Hyperlinked PDFs");
        //Sample PDF
        private static readonly string BaseFile = Path.Combine(WorkingFolder, "OldFile.pdf");
        //Final file
        private static readonly string OutputFile = Path.Combine(WorkingFolder, "NewFile.pdf");

        public Form1()
        {
            InitializeComponent();
        }

        private void Form1_Load(object sender, EventArgs e)
        {
            CreateSamplePdf();
            UpdatePdfLinks();
            this.Close();
        }

        private static void CreateSamplePdf()
        {
            //Create our output directory if it does not exist
            Directory.CreateDirectory(WorkingFolder);

            //Create our sample PDF
            using (iTextSharp.text.Document Doc = new iTextSharp.text.Document(PageSize.LETTER))
            {
                using (FileStream FS = new FileStream(BaseFile, FileMode.Create, FileAccess.Write, FileShare.Read))
                {
                    using (PdfWriter writer = PdfWriter.GetInstance(Doc, FS))
                    {
                        Doc.Open();

                        //Turn our hyperlink blue
                        iTextSharp.text.Font BlueFont = FontFactory.GetFont("Arial", 12, iTextSharp.text.Font.NORMAL, iTextSharp.text.BaseColor.BLUE);

                        Doc.Add(new Paragraph(new Chunk("Go to URL", BlueFont).SetAction(new PdfAction("http://www.google.com/", false))));

                        Doc.Close();
                    }
                }
            }
        }

        private static void UpdatePdfLinks()
        {
            //Setup some variables to be used later
            PdfReader R = default(PdfReader);
            int PageCount = 0;
            PdfDictionary PageDictionary = default(PdfDictionary);
            PdfArray Annots = default(PdfArray);

            //Open our reader
            R = new PdfReader(BaseFile);
            //Get the page cont
            PageCount = R.NumberOfPages;

            //Loop through each page
            for (int i = 1; i <= PageCount; i++)
            {
                //Get the current page
                PageDictionary = R.GetPageN(i);

                //Get all of the annotations for the current page
                Annots = PageDictionary.GetAsArray(PdfName.ANNOTS);

                //Make sure we have something
                if ((Annots == null) || (Annots.Length == 0))
                    continue;

                //Loop through each annotation

                foreach (PdfObject A in Annots.ArrayList)
                {
                    //Convert the itext-specific object as a generic PDF object
                    PdfDictionary AnnotationDictionary = (PdfDictionary)PdfReader.GetPdfObject(A);

                    //Make sure this annotation has a link
                    if (!AnnotationDictionary.Get(PdfName.SUBTYPE).Equals(PdfName.LINK))
                        continue;

                    //Make sure this annotation has an ACTION
                    if (AnnotationDictionary.Get(PdfName.A) == null)
                        continue;

                    //Get the ACTION for the current annotation
                    PdfDictionary AnnotationAction = (PdfDictionary)AnnotationDictionary.Get(PdfName.A);

                    //Test if it is a URI action
                    if (AnnotationAction.Get(PdfName.S).Equals(PdfName.URI))
                    {
                        //Change the URI to something else
                        AnnotationAction.Put(PdfName.URI, new PdfString("http://www.bing.com/"));
                    }
                }
            }

            //Next we create a new document add import each page from the reader above
            using (FileStream FS = new FileStream(OutputFile, FileMode.Create, FileAccess.Write, FileShare.None))
            {
                using (Document Doc = new Document())
                {
                    using (PdfCopy writer = new PdfCopy(Doc, FS))
                    {
                        Doc.Open();
                        for (int i = 1; i <= R.NumberOfPages; i++)
                        {
                            writer.AddPage(writer.GetImportedPage(R, i));
                        }
                        Doc.Close();
                    }
                }
            }
        }
    }
}

EDIT

I should note, this only changes the actual link. Any text within the document won’t get updated. Annotations are drawn on top of text but aren’t really tied to the text underneath in anyway. That’s another topic completely.