How to Test If Associated PDF Viewer Has Been Opened?

Check the registry key for .pdf extension to see what application is associated with it

HKEY_CURRENT_USER\Software\Microsoft\Windows\CurrentVersion\Explorer\FileExts\.pdf\UserChoice

The ProgId key will show the application e.g. FoxitPhantomPDF.Document

If you then search for the ProgId (in this case it's "FoxitPhantomPDF.Document")  in the registry i.e.

HKEY_CLASSES_ROOT\FoxitPhantomPDF.Document\shell\open\command

You will see the full execution path name

You can then use TestComplete' Process method to see if the application is running.

Hi KRogersX,

This may help on the registry bit...

How To: Read data from the Windows Registry | SmartBear Community

As for ..."if it's been opened successfully with a specific PDF?" You could check the object tree for the viewer process with some code after you retrieve the associated viewer.  It may take some clever code to check for the various viewers but it should be possible.  Opening via a browser would be a different process as well.  

... If you find my posts helpful drop me a like! 👍 Be sure to mark or post the solution to help others out and/or to credit the one who helped you. 😎

This code should open the file in the specified path and you can use the registry code to determine the viewer to check for...

function OpenAndVerifyPDF(pdfPath) {
  var shell = Sys.OleObject("WScript.Shell");
  shell.Run('"' + pdfPath + '"', 1, false); // Launch PDF with default viewer

  // Wait a few seconds for the viewer to open
  Delay(3000);

  // Check if the Adobe Reader or another viewer process is running
  if (Sys.WaitProcess("AcroRd32", 5000).Exists || Sys.WaitProcess("Acrobat", 5000).Exists) {
    Log.Message("PDF viewer opened successfully.");
    return true;
  } else {
    Log.Error("PDF viewer did not open.");
    return false;
  }
}

P.S. The bit of code in this example that checks the tree for the running process could be useful after your app has opened the file.  You will need to check the associated  process and window caption for the filename if available.

After your application under test launches the required PDF file and to verify that the file is opened correctly in its associated application even when another instance of the viewer might already be running or in Browser / Embedded viewers.

This can be achieved by reading the Windows Registry to determine the default PDF viewer, and checking window titles for the expected file name within a 10-second timeout.

I have not tried it out but here's a TestComplete example in JavaScript:

function VerifyPDF(pdfFilePath) {
  if (!aqFileSystem.Exists(pdfFilePath)) {
    Log.Error("The specified PDF file does not exist: " + pdfFilePath);
    return;
  }

  var fileName = aqFileSystem.GetFileName(pdfFilePath).toLowerCase();
  var fileBase = fileName.replace(/\.[^/.]+$/, "");

  // === STEP 1: Determine default viewer from registry
  var exeName = null;
  var useFallback = false;

  var progID = aqRegistry.GetRegValue("HKEY_CLASSES_ROOT\\.pdf", "");
  if (progID) {
    var commandPath = "HKEY_CLASSES_ROOT\\" + progID + "\\shell\\open\\command\\";
    var openCommand = aqRegistry.GetRegValue(commandPath, "");
    var exeMatch = openCommand.match(/"?(.*?\.exe)"/i);
    var exePath = exeMatch ? exeMatch[1] : null;
    exeName = exePath ? aqFileSystem.GetFileName(exePath).toLowerCase() : null;
  }

  // === STEP 2: Define fallback process list (common browsers and embedded containers)
  var fallbackProcesses = [
    "msedge", "chrome", "firefox", "iexplore", // browsers
    "electron", "yourapp", "customviewer" // embedded apps (customize here)
  ];

  if (!exeName) {
    Log.Warning("Could not detect default PDF viewer from registry. Falling back to known browser/embedded types.");
    useFallback = true;
  }

  var timeout = 10000; // 10 seconds
  var start = aqDateTime.GetTickCount();
  var found = false;
  var anyViewerFound = false;

  while ((aqDateTime.GetTickCount() - start) < timeout) {
    for (var i = 0; i < Sys.ProcessCount; i++) {
      var proc = Sys.Process(i);
      var procName = proc.Name.toLowerCase();

      // Match default viewer or fallback list
      var isTargetViewer = (!useFallback && procName === exeName.replace(".exe", "")) ||
                           (useFallback && fallbackProcesses.indexOf(procName) >= 0);

      if (!isTargetViewer) continue;
      anyViewerFound = true;

      // === Check top-level windows
      for (var j = 0; j < proc.WindowCount; j++) {
        var wnd = proc.Window("*", "*", j);
        if (!wnd.Exists) continue;

        var title = wnd.WndCaption.toLowerCase();

        // For browsers, title may show full path or filename
        if (title.includes(fileBase) || title.includes(fileName)) {
          Log.Message("PDF file '" + fileName + "' is opened in: " + procName + " - Window: " + wnd.WndCaption);
          found = true;
          break;
        }
      }

      if (found) break;
    }

    if (found) break;
    Delay(500);
  }

  // === Final result
  if (!found) {
    if (!anyViewerFound) {
      Log.Error("No PDF viewer process (application or browser) was started.");
    } else {
      Log.Warning("Viewer is running, but no window was found showing the PDF file: " + fileName);
    }
  }
}

💬 If a response helped you out, don’t forget to Like it! And if it answered your question, mark it as the solution so others can benefit too.

This part is a bit confusing from an automation test perspective,

I need to test if the PDF has actually successfully opened in the associated PDF viewer. I don't want to assume the user has a specific PDF viewer like Adobe Reader. 

I am not sure how this could be done unless you check every possible reader.  A developer could build an error check into the application under test but I don't know how you would test this.

(P.S. Maybe you are spinning up various test environments multiple VMs? )