Overview: Build a console to help migrate more than 80k PDF document internal links. The client used a DFS SMB file share to hold index PDFs and multiple documents that needed to be moved to a SharePoint document library.
Hypothesis: Loop through all PDFs in a folder; if there are links, identify the file server links and convert them to web links so they work in the new SharePoint document library. Various tools were identified as possible solutions, but came up short in the migration. Two good tools are Replace Magic & PDF-XChange Editor.
Resolution: Below is the C# code I wrote to update the links in VS Code. The debugger was helpful because there were many different link types across the plethora of PDFs.
using iText.Kernel.Pdf.Annot;
using iText.Kernel.Pdf.Action;
using iText.Bouncycastle.Crypto; // pdf fails at runtime periodically without the directive
using System.Text.RegularExpressions;
class Program
{
static void Main(string[] args)
{
string folderPath = @"C:\Users\PaulBeck\Downloads\Software\LinkConvert\ConvertLinksCsharp\"; // Replace with your folder path
Console.WriteLine("Please enter a folder Path: e.g. " + folderPath);
string inputPath = Console.ReadLine();
Console.WriteLine("Last Path: e.g. Childfolder2");
string inputPathVol = Console.ReadLine();
if (inputPath.Length>5) {
folderPath = inputPath; }
string[] pdfFiles = Directory.GetFiles(folderPath, "*.pdf");
foreach (string file in pdfFiles)
{
Uri fileUri = new Uri(file);
string directory = Path.GetDirectoryName(file);
string filename = Path.GetFileNameWithoutExtension(file);
string extension = Path.GetExtension(file);
string newFilename = $"{filename}_new{extension}"; // Create the new filename
string newFilePath = Path.Combine(directory, newFilename); // Combine the directory and new filename to form the new URL
UpdatePdf(fileUri.AbsoluteUri,newFilePath, inputPathVol);
}
}
private static void UpdatePdf(string inputFilePath, string outputFilePath, string lastPathPart)
{
var varInnerName = "";
PdfDocument pdfDoc = new PdfDocument(new PdfReader(inputFilePath), new PdfWriter(outputFilePath));
for (int i = 1; i <= pdfDoc.GetNumberOfPages(); i++) // Iterate through the pages
{
var page = pdfDoc.GetPage(i);
var annotations = page.GetAnnotations();
foreach (var annotation in annotations) // Iterate through the annotations
{
if (annotation.GetSubtype().Equals(PdfName.Link))
{
var linkAnnotation = (PdfLinkAnnotation)annotation;
var action = linkAnnotation.GetAction();
if (action is PdfDictionary dictionary)
{
foreach (var key in dictionary.KeySet())
{
var value = dictionary.Get(key);
Console.WriteLine($"{key}: {value}");
var varPdfNameF = dictionary.Get(PdfName.F);
if (varPdfNameF is PdfDictionary varF2Dict) {
varInnerName = varF2Dict.Get(PdfName.F).ToString(); }
}
}
else
{ Console.WriteLine("No URL found."); }
if (action != null && (action.Get(PdfName.S).Equals(PdfName.GoToR) || action.Get(PdfName.S).Equals(PdfName.Launch)))
{
var varF = action.GetAsString(PdfName.F)?.ToString() ?? "";
var uri = $"https://radimaging.sharepoint.com/sites/Documents/Standards/{lastPathPart}/{varInnerName}";
string pattern = @"\.\./";
if(uri.Contains("../")) // string cleanUrl1 = Regex.Replace(uri, pattern, string.Empty);
{
uri = $"https://radimaging.sharepoint.com/sites/Documents/Standards/{varInnerName}";
uri= Regex.Replace(uri, pattern, string.Empty);
}
var newAction = PdfAction.CreateURI(uri);
if (varF.Length > 20) {
newAction = PdfAction.CreateURI(varF); }
linkAnnotation.SetAction(newAction);
}
}
}
}
pdfDoc.Close();
Console.WriteLine("PDF links updated successfully!");
}
}
 |
| Quick way to use the PDF X-Change Editor tool to change links. |
Example of how to correctly set internal links within a pdf:
Example of how to set links to pages or files on the Internet/extranet:
Update Nov 2025: User Playwright MCP to download an index pdf, and then download or open files and pages to validate the links worked. Updated the index pdf using C# Console built using Github Copilot (GHCP) and updated into SharePoint using MS Graph the updated fixed pdf.
