Skip to main content

4 posts tagged with "XLL"

View All Tags

· 6 min read

Excel-DNA version 1.6 is now available for testing as pre-release version 1.6.0-preview3 from the NuGet package repository. The extension libraries ExcelDna.Registration and ExcelDna.IntelliSense also have matching releases. ExcelDnaDoc should follow in the coming weeks.

The main focus for version 1.6 is to provide a first preview of the support for add-ins that target .NET 6. As a prerequisite to adding the .NET 6 target, we’ve also added support for new SDK-style project files and NuGet references (also great for projects that target .NET Framework). I’ve also taken the opportunity to tweak the Excel-DNA packing mechanism a bit in an attempt to avoid some of the false positive anti-virus detections we’ve seen recently. I’ll briefly discuss these (in reverse order) below, but first a word about our sponsors.

GitHub sponsors

Excel-DNA is now registered on GitHub sponsors – see https://github.com/sponsors/Excel-DNA. Thank you very much to everyone who has already signed up – your contributions are directly funding further development.

If you use Excel-DNA and would like to encourage future support and ongoing development, please sign up as a GitHub sponsor for the project. GitHub sponsors will also have access to a private repository of sample projects where I hope to add additional tools and documentation over time.

I had previously considered the .NET Core / .NET 6 support as a good point to switch to a commercial model for Excel-DNA. However, the benefits of being an open-source project with a permissive license have been significant. So, if possible, I hope to continue with the core library as free and open source software, where further development is funded by the sponsors.

For those in a corporate setting using Excel-DNA extensively or in a mission critical role, I also offer an annual corporate support agreement. Please contact me directly if you are interested in more details.

Anti-virus false positives

Over the last year we’ve seen a number of anti-virus and security programs identify Excel-DNA add-ins as security risks, including false positive detections that delete add-in files by the default Windows Defender service. It seems that these detections were triggered by some malicious Excel add-ins that have been built using Excel-DNA, with the anti-virus heuristics subsequently identifying all Excel-DNA add-ins as problematic. For some security vendors, including Microsoft, we’ve been able to report this false positive detection and they have updated their signatures accordingly. But in many cases the Excel-DNA version 1.5 add-ins are still being flagged.

It seems the main heuristic used to detect Excel .xll add-ins as problematic is the presence of executable assemblies as resources in the .xll library. Excel-DNA uses the packing of assemblies as resources in the .xll to simplify the add-in distribution, in many cases making possible a single file (or two files for 32-bit and 64-bit) add-in distribution.

For this version there are two changes which may help with these false positives.

The first is that packed files are now encoded, so that they are not detected as embedded executable code directly. In my testing that change removed the false positive detections at least in the Microsoft products. However, at least one other user has reported still seeing problems with this version, so it does not seem like bulletproof solution.

Another option has been contributed by user @Phundamentals (thank you!), to add a project property which disables the compression of packed files.

  <ExcelDnaPackCompressResources>false</ExcelDnaPackCompressResources>

Should these mitigations not be enough to reduce most false positives, we could also introduce an option where there are no packed files at all (currently the managed Excel-DNA assemblies are always packed).

There have also been some indications that signing the final add-in library helps reduce false positive detections.

In all cases I do encourage developers to report the false positives to their anti-virus or security vendor. Their software is mistakenly identifying and blocking legitimate add-ins from running.

Developers of malicious add-ins are welcome to contact me for sponsored licenses of an alternative add-in framework.

PackageReference and SDK-style project files

The main ExcelDna.AddIn NuGet package now supports references in SDK-style project files. For new projects, this means that a .dna file is no longer added to the project automatically, and if not present will be created and output at build time from project properties. Existing projects with customized .dna file(s) will work as they did before.

For a new project, a simple project file might look like this:

<Project Sdk="Microsoft.NET.Sdk">

<PropertyGroup>
<TargetFramework>net472</TargetFramework>

<!-- We don't need the extra 'ref' directory and reference assemblies for the Excel add-in -->
<ProduceReferenceAssembly>false</ProduceReferenceAssembly>

<!-- We need all dependencies to be copied to the output directory, as-if we are an 'application' and not a 'library'.
This property also sets the CopyLockFileAssemblies property to true. -->
<EnableDynamicLoading>true</EnableDynamicLoading>
</PropertyGroup>

<ItemGroup>
<PackageReference Include="ExcelDna.AddIn" Version="1.6.0-preview3" />
</ItemGroup>

</Project>

Various additional properties can be set in the project file – see the example here https://github.com/Excel-DNA/ExcelDna/blob/master/Source/Tests/ExcelDna.AddIn.Tasks.IntegrationTests.TestTarget/SDKProperties/SDKProperties.csproj

.NET 6 support

Excel-DNA version 1.6 (finally!) includes support for add-ins that target .NET 6. Most Excel-DNA features seem to work, including the COM-based features like RTD-based functions, ribbon and CTP extensions. However, there has been very limited testing and you should consider the .NET 6 support as an early preview.

Update the TargetFramework tag in an SDK-style project file to the .NET 6 (Windows) target:

    <TargetFramework>net6.0-windows</TargetFramework>

or build for both .NET Framework and .NET 6, which allows you to test both targets:

    <TargetFrameworks>net472;net6.0-windows</TargetFrameworks>

End users of the add-in will need to have the .NET 6 runtime installed. (To include the runtime files as part of the add-in will not be supported, but an installer program might check and install the runtime.)

A hard limitation of the .NET core series of runtimes (.NET 5 / 6 / 7 etc.) is that only one version of a .NET core runtime can be loaded into a specific process. The core runtime can still be loaded concurrently with a .NET Framework runtime (e.g. .NET Framework 4.8), although this is not a configuration officially supported by Microsoft. Excel-DNA will try to support add-ins targeting a .NET Framework running together with .NET 6 add-ins, but there is not planned to be any support for the future .NET 7 runtime. Fortunately .NET 6 is a ‘Long-term Support’ (LTS) release and will be formally supported until the end of 2024, so we can keep targeting .NET 6 for a few years. There is also some work on the horizon that make ahead-of-time compilation of .NET libraries (like the Excel-DNA add-ins) viable, bypassing the concurrent runtime issues. So I do expect a future path beyond .NET 6, though that is not an immediate concern.

I look forward to bug reports, questions and other feedback about the .NET 6 support (including whether it works at all). Support for modern .NET has been a long time coming and ensures that Excel-DNA can take part in the exciting future evolution of .NET.

After 16 years I am still amazed by and deeply appreciate the support and enthusiasm for Excel-DNA. Last but not least I want to thank Sergey Vlasov for his calm and consistent efforts that are now driving the project forward.

-Govert

· 2 min read

Functions and macros created in an Excel-DNA add-in can be called directly from Excel VBA by using Application.Run(...). However, .NET also supports creating rich object models that are exported as COM libraries, which can be Tools->Referenced in VBA. Excel-DNA has some advanced support to host COM-exported objects from Excel-DNA add-ins, giving some advantages over the regular .NET "Register for COM interop" hosting approach:

  • COM objects that are created via the Excel-DNA COM server support will be active in the same AppDomain as the rest of the add-in, allowing direct shared access to static variables, internal caches etc.

  • COM registration for classes hosted by Excel-DNA does not require administrative access (even when registered via RegSvr32.exe).

  • Everything needed for the COM server can be packed in a single-file .xll add-in, including the type library used for IntelliSense support in VBA.

Mikael Katajamäki has written some detailed tutorial posts on his Excel in Finance blog that explore this Excel-DNA feature, with detailed explanation, step-by-step instructions, screen shots and further links. See:

Note that these techniques would work equally well with code written in VB.NET, allowing you to port VB/VBA libraries to VB.NET with Excel-DNA and then use these from VBA.

Thank you Mikael for the great write-up!

· 6 min read

Update (21 June 2017): The most up-to-date version of the ArrayResizer utility can be found here: https://github.com/Excel-DNA/ExcelDna/blob/master/Distribution/Samples/ArrayResizer.dna

Update: To work correctly under Excel 2000/2002/2003, this sample requires at least version 0.29.0.12 of Excel-DNA.

A common question on the Excel-DNA group is how to automatically resize the results of an array formula. The most well-know appearance of this trick is in the Bloomberg add-in.

WARNING! This is a bad idea. Excel does not allow you to modify the sheet from within a user-defined function. Doing this breaks Excel's calculation model.

Anyway, here is my attempt at an Excel-DNA add-in that implements this trick. My approach is to run a macro on a separate thread that will check and if required will expand the formula to an array formula of the right size. This way nothing ugly gets done if the array size is already correct - future recalculations will not run the formula array resizer if the size is still correct.

The code below will register a function call Resize. You can either call Resize from within your function, or enter something like =Resize(MyFunction(…)) as the cell formula. The code also registers two sample functions, MakeArray and MakeArrayAndResize to play with, each take the number of rows and columns for the size of the returned array.

To test this:

  1. Get started with Excel-DNA.
  2. Copy the code and xml wrapper into a text file called Resizer.dna (the xml wrapper is at the end of this post).
  3. Copy the ExcelDna.xll in the Excel-DNA distribution to Resizer.xll (next to the Resizer.dna).
  4. File->Open the Resizer.xll in Excel and enter something like =MakeArrayAndResize(5,3) into a cell. See how it grows.

In the current version, the formula expansion is destructive, so anything in the way will be erased. One case I don't know how to deal with is when there is an array that would be partially overwritten by the expended function result. In the current version Excel will display an error that says "You cannot change part of an array.", and I replace the formula with a text version of it. I'd love to know how you think we should handle this case.

Any questions or comments (can if anyone can get it to work, or not?) can be directed to the [Excel-DNA Google group][excel-dna-group] or to me directly via e-mail. I'm pretty sure there are a few cases where it will break - please let me know if you run into any problems.

I'll try to gather the comments and suggestions for an improved implementation that might go into the next version of Excel-DNA.

Also, if you have any questions about how the implementation works, I'd be happy to write a follow up post that explains a bit more of what I'm doing. But first, let's try to get it working.

Here's the Resizer add-in code:

using System;
using System.Collections.Generic;
using System.Reflection;
using System.Runtime.InteropServices;
using System.Threading;
using ExcelDna.Integration;

public static class ResizeTest
{
public static object MakeArray(int rows, int columns)
{
object[,] result = new string[rows, columns];
for (int i = 0; i < rows; i++)
{
for (int j = 0; j < columns; j++)
{
result[i,j] = string.Format("({0},{1})", i, j);
}
}

return result;
}

public static object MakeArrayAndResize(int rows, int columns)
{
object result = MakeArray(rows, columns);
// Call Resize via Excel - so if the Resize add-in is not part of this code, it should still work.
return XlCall.Excel(XlCall.xlUDF, "Resize", result);
}
}

public class Resizer
{
static Queue<ExcelReference> ResizeJobs = new Queue<ExcelReference>();

// This function will run in the UDF context.
// Needs extra protection to allow multithreaded use.
public static object Resize(object[,] array)
{
ExcelReference caller = XlCall.Excel(XlCall.xlfCaller) as ExcelReference;
if (caller == null)
return array;

int rows = array.GetLength(0);
int columns = array.GetLength(1);

if ((caller.RowLast - caller.RowFirst + 1 != rows) ||
(caller.ColumnLast - caller.ColumnFirst + 1 != columns))
{
// Size problem: enqueue job, call async update and return #N/A
// TODO: Add guard for ever-changing result?
EnqueueResize(caller, rows, columns);
AsyncRunMacro("DoResizing");
return ExcelError.ExcelErrorNA;
}

// Size is already OK - just return result
return array;
}

static void EnqueueResize(ExcelReference caller, int rows, int columns)
{
ExcelReference target = new ExcelReference(caller.RowFirst, caller.RowFirst + rows - 1, caller.ColumnFirst, caller.ColumnFirst + columns - 1, caller.SheetId);
ResizeJobs.Enqueue(target);
}

public static void DoResizing()
{
while (ResizeJobs.Count > 0)
{
DoResize(ResizeJobs.Dequeue());
}
}

static void DoResize(ExcelReference target)
{
try
{
// Get the current state for reset later

XlCall.Excel(XlCall.xlcEcho, false);

// Get the formula in the first cell of the target
string formula = (string)XlCall.Excel(XlCall.xlfGetCell, 41, target);
ExcelReference firstCell = new ExcelReference(target.RowFirst, target.RowFirst, target.ColumnFirst, target.ColumnFirst, target.SheetId);

bool isFormulaArray = (bool)XlCall.Excel(XlCall.xlfGetCell, 49, target);
if (isFormulaArray)
{
object oldSelectionOnActiveSheet = XlCall.Excel(XlCall.xlfSelection);
object oldActiveCell = XlCall.Excel(XlCall.xlfActiveCell);

// Remember old selection and select the first cell of the target
string firstCellSheet = (string)XlCall.Excel(XlCall.xlSheetNm, firstCell);
XlCall.Excel(XlCall.xlcWorkbookSelect, new object[] {firstCellSheet});
object oldSelectionOnArraySheet = XlCall.Excel(XlCall.xlfSelection);
XlCall.Excel(XlCall.xlcFormulaGoto, firstCell);

// Extend the selection to the whole array and clear
XlCall.Excel(XlCall.xlcSelectSpecial, 6);
ExcelReference oldArray = (ExcelReference)XlCall.Excel(XlCall.xlfSelection);

oldArray.SetValue(ExcelEmpty.Value);
XlCall.Excel(XlCall.xlcSelect, oldSelectionOnArraySheet);
XlCall.Excel(XlCall.xlcFormulaGoto, oldSelectionOnActiveSheet);
}
// Get the formula and convert to R1C1 mode
bool isR1C1Mode = (bool)XlCall.Excel(XlCall.xlfGetWorkspace, 4);
string formulaR1C1 = formula;
if (!isR1C1Mode)
{
// Set the formula into the whole target
formulaR1C1 = (string)XlCall.Excel(XlCall.xlfFormulaConvert, formula, true, false, ExcelMissing.Value, firstCell);
}
// Must be R1C1-style references
object ignoredResult;
XlCall.XlReturn retval = XlCall.TryExcel(XlCall.xlcFormulaArray, out ignoredResult, formulaR1C1, target);
if (retval != XlCall.XlReturn.XlReturnSuccess)
{
// TODO: Consider what to do now!?
// Might have failed due to array in the way.
firstCell.SetValue("'" + formula);
}
}
finally
{
XlCall.Excel(XlCall.xlcEcho, true);
}
}

// Most of this from the newsgroup: http://groups.google.com/group/exceldna/browse_thread/thread/a72c9b9f49523fc9/4577cd6840c7f195
private static readonly TimeSpan BackoffTime = TimeSpan.FromSeconds(1);
static void AsyncRunMacro(string macroName)
{
// Do this on a new thread....
Thread newThread = new Thread( delegate ()
{
while(true)
{
try
{
RunMacro(macroName);
break;
}
catch(COMException cex)
{
if(IsRetry(cex))
{
Thread.Sleep(BackoffTime);
continue;
}
// TODO: Handle unexpected error
return;
}
catch(Exception ex)
{
// TODO: Handle unexpected error
return;
}
}
});
newThread.Start();
}

static void RunMacro(string macroName)
{
object xlApp;
try
{
object xlApp = ExcelDnaUtil.Application;
xlApp.GetType().InvokeMember("Run", BindingFlags.InvokeMethod, null, xlApp, new object[] {macroName});
}
catch (TargetInvocationException tie)
{
throw tie.InnerException;
}
finally
{
Marshal.ReleaseComObject(xlApp);
}
}

const uint RPC_E_SERVERCALL_RETRYLATER = 0x8001010A;
const uint VBA_E_IGNORE = 0x800AC472;
static bool IsRetry(COMException e)
{
uint errorCode = (uint)e.ErrorCode;
switch(errorCode)
{
case RPC_E_SERVERCALL_RETRYLATER:
case VBA_E_IGNORE:
return true;
default:
return false;
}
}
}

You can easily make a test add-in for this by wrapping the code into a .dna file with this around:

<DnaLibrary Language="CS">
<![CDATA[

<!--// Paste all of the above code here //-->

]]>
</DnaLibrary>

· 2 min read

Excel 2010 introduced support for offloading UDF computations to a compute cluster. The Excel blog talks about it http://blogs.msdn.com/b/excel/archive/2010/02/12/offloading-udf-s-to-a-windows-hpc-cluster.aspx, and there are some nice pictures on this TechNet article: http://technet.microsoft.com/en-us/library/ff877825(WS.10).aspx.

Excel-DNA now supports marking functions as cluster-safe, and I have updated the loader to allow add-ins to work under the XllContainer on the HPC nodes. There are some issues to be aware of:

  • The add-in does not create its own AppDomain when running on the compute node. One consequence is that no custom .xll.config file is used; configuration entries need to be set in the XllContainer configuration setup.
  • There are some limitations on the size of array data that can be passed to and from UDF calls - this limit is probably configurable in the WCF service.
  • Only the 32-bit host is currently supported.

To test this you will need an Windows HPC Server 2008 R2 cluster with the HPC Services for Excel installed. On the clients you need Excel 2010 with the HPC cluster connector installed. The latest check-in for Excel-DNA with this support is on GitHub: https://github.com/Excel-DNA/ExcelDna.

In the Microsoft HPC SDK there is a sample called ClusterUDF.xll with a few test functions. I have recreated these in C# in the samples file Distribution\Samples\ClusterSample.dna Basically functions just need to be marked as IsClusterSafe=true to be pushed to the cluster for computation. For example

[ExcelFunction(IsClusterSafe=true)]
public static int DnaCountPrimesC(int nFrom, int nTo)
{
// ...
}

As usual, any feedback on this feature - questions or reports on whether you use it - will be most appreciated.