When Excel-like filtering is enabled and the user selects an item from the "Available Filters" drop down menu, a CompositeFilterForm is shown.

RadVirtualGrid is created to support a million records or more. Still it contains some "heavy" operations like CopySelection/CutSelection/Paste which can take a long time.

I would like to intercept these methods so I can show a "Please wait"-window during the operation. Furthermore I have the need to set a flag to true when such a operation is executing. 

Currently, after installing Visual Studio 2022 and Telerik UI for WinForms suite on a Windows 11 Arm I cannot see any extension installed on Extension menu:

Implement support for content controls (a.k.a. Structured document tags), which will allow inserting editing controls in the document:

- Rich Text
- Plain Text
- Check Box
- Combo Box
- Drop-down list
- Date picker

What I would like is to be able to export to a named sheet and keep the other sheets in file. For example:
"Sheet 1" (with new values), "Sheet2" (with old values)

Using your current naming convention, the Option might be named FileExportMode.CreateOrOverrideSheet

This functionality will decrease the size of the exported document.

When the columns are auto-generated, the way to modify the columns is to use the Columns collection. We could expose an event that will be called when the columns are auto-generating. From the event arguments, we could get the newly created column and replace with if needed or modify it. Similar to the CreateRow event of the control.

We work with a RadGridView with 145000 rows and 4 columns. We use copy-paste to move data around from other apps and the application with build with Telerik.

When we copy a flat file (with tabs delimited fields) and we paste it to the RadGridView, the whole process is painfully slow. The function to retrieve the data from the clipboard takes minutes (maybe hours, I cancelled it). It tracked the cause down to the StringTokenizer class. The tokenizer splits the string up into separate fields. But after extracting a field it creates a new copy of that string (containing about 10MB of data) minus the field. I patched it (with HarmonyX) and now it takes only one second:

static class StringTokenizerPerformancePatch
{
	static private readonly InstanceFieldAccessor<StringTokenizer, LinkedList<string>> _tokens = new InstanceFieldAccessor< StringTokenizer, LinkedList<string>>("tokens");
	static private readonly InstanceFieldAccessor<StringTokenizer, string> _sourceString = new InstanceFieldAccessor<StringTokenizer, string>("sourceString");
	static private readonly InstanceFieldAccessor< StringTokenizer, string> _delimiter = new InstanceFieldAccessor<StringTokenizer, string>("delimiter");
	static private readonly InstanceFieldAccessor< StringTokenizer, IEnumerator<string>> _enumerator = new InstanceFieldAccessor<StringTokenizer, IEnumerator<string>>("enumerator");

	[HarmonyPatch(typeof(StringTokenizer), "Tokenize")]
	staticclassPatch_StringTokenizer_Tokenize
	{
		static bool Prefix(StringTokenizer __instance)
		{
			var tokens = _tokens.GetValue(__instance);
			var sourceString = _sourceString.GetValue(__instance);
			var delimiter = _delimiter.GetValue(__instance);
			Tokenize(tokens, sourceString, delimiter);
			_enumerator.SetValue(__instance, tokens.GetEnumerator());
			returnfalse;
		}

		static private void Tokenize(LinkedList<string> tokens, string text, string delimiter)
		{
			tokens.Clear();
			
			if (string.IsNullOrEmpty(text))
				return;

			int index = 0;
			while(true)
			{
				var index2 = text.IndexOf(delimiter, index, StringComparison.Ordinal);
				if (index2 < 0)
				{
					tokens.AddLast(text.Substring(index));
					break;
				}

				string token = text.Substring(index, index2 - index);
				tokens.AddLast(token);
				index = index2 + delimiter.Length;
			}
		}
	}
}

• If delimiters are one character, why not use String.Split?
• Why use a LinkedList?

Is there a way to load this RadFixedDocument into a RadPdfViewer without writing the document to a file first and then loading it in the viewer?

Workaround:

You can achieve the desired functionality by exporting the RadFixedDocument to a MemoryStream instead of a file in the file system and then load it in the PdfViewer: 

RadFixedDocument document = CreateRadFixedDocument();

PdfFormatProvider pdfFormatProvider = new PdfFormatProvider();
Stream ms = new MemoryStream();
pdfFormatProvider.Export(document, ms);

radPdfViewer1.LoadDocument(ms);

Add heatmap control as in the WPF suite http://www.telerik.com/products/wpf/heatmap.aspx

According to PDF format specification, there are three valid encoding name values (MacRomanEncoding, MacExpertEncoding and WinAnsiEncoding). There are documents that instead of skipping the optional Encoding property, are writing invalid /NULL name value in the font dictionary. Currently, in this invalid document scenario RadPdfViewer throws and catches Exception and this results in missing text content. 

We may handle this invalid document scenario by ignoring the invalid Encoding value.

Our clients can download different product files from their Telerik account:

https://docs.telerik.com/devtools/winforms/installation-and-upgrades/download-product-files 

Currently, when Digitally Signed Assemblies are necessary, they submit support tickets and the support engineers provide the signed assemblies. It would be nice to be directly uploaded to the Telerik account like the rest of the product files.