1) When trying to copy the given source code you’ll mention that all the classes are unknown to Visual Studio, because all a lot of references are missing. So far I have no clue about the R#-class-architecture. So I started .NET Reflector, open ALL the .dlls in the R#/bin-folder into Reflector and use the search to find the .dlls to reference in my project.

2) When you have your first R#-plugin ready you want to test it. To do so you have to copy the dll containing your plugin in the R#-plugin folder. You may want to do this in a post-build event like this:


copy $(TargetPath) "C:\Program Files (x86)\JetBrains\ReSharper\v5.1\Bin\PlugIns\WobTest"


When you build your project with this post-build event it fails. This doesn’t work for two reasons. First: Win 7 denies access to that folder. Second: after another start of Visual Studio R# has loaded the plugin so you can’t copy a newer version over it.

My solution to this is a .cmd that starts Visual Studio without R# (devenv /safemode). I run this .cmd as administrator.

Now at least I can debug my plugins by starting another instance of Visual Studio (not in safemode!)

This solution is not cool, because developing R#-plugin without using R# in the development-process sucks…. any ideas?

a better solution…

Some minutes after I published this post, Benjamin pointed me to this page. Forget about my lines to 2). You don’t need a silly .cmd run as administrator not a post-build event. All you need to do is to start your debugging Visual Studio the the command line option /ReSharper.Plugin followed by the path to your plugin-dll.

/ReSharper.Plugin "R:\sandbox\ReSharperPlugInTest\PublicGoesVirtual\PublicGoesVirtual\bin\Debug\PublicGoesVirtual.dll"

I’m not sure what append, maybe it happed on a project I imported from an older Visual Studio version. But for this project the property I needed so much was just missing.

If the same happens to you, you can edit the DslPackage.csproj and add the following line:

<ProjectTypeGuids>{82b43b9b-a64c-4715-b499-d71e9ca2bd60};     {FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}</ProjectTypeGuids>

It worked on my machine and I got this property back.

I created a setup and changed the prerequisites from Client Profile to the Full Framework. When testing this on a virgin machine the setup told me that I need to install the .NET 4.0 framework (that’s right) but then it directs me to the download of the Client Profile only.

To fix this, you have to change the InstallUrl of the .NET Framework Launch Condition:

• right-click on the setup project in solution explorer and choose View –> Launch Conditions
• here you will find a Launch Condition for the .NET Framework
• select this Launch Condition and open the Properties Window
• there you will find the InstallUrl property

For the full .net 4.0 Framework change this URL to http://go.microsoft.com/fwlink/?LinkId=186913

For the Client Profile of the .net 4.0 Framework the URL can stay as it is: http://go.microsoft.com/fwlink/?LinkId=131000

First of all I have to explain to everybody what I’m talking about:

With Visual Studio 2010 Microsoft introduced a new way to deploy extensions to Visual Studio. Instead of creating a setup.exe that copies files, writes to the registry and calls devenv.exe with the /setup option one only needs such a VSIX file that does all that is necessary to install an extension.

When working with the T4 (Template Transformation Toolkit) system one creates tt files with some kind of script that is used to generate source code within a user’s Visual Studio project. The Microsoft DSL Tools do the same: every DSL project contains a number of tt files that will generate all the code. If you take a look at such a file, you will only see two lines:

<#@ Dsl processor="DslDirectiveProcessor" requires="fileName='..\DslDefinition.dsl'" #>
<#@ include file="Dsl\Diagram.tt" #>

The magic happens with the include command. The real template code sits in the included file and not within every single project. When installing the DSL Tools extension to Visual Studio all these include files get installed somewhere where they can be found by the tt engine.

For my own DSL Languages I would like do the same: the user projects should only contain such two line tt files and include the rest of the template from other files.

How to deploy tt template files?

The following steps will show you how to deploy additional files with you DSL extension.

I will start from a newly created Domain-Specific Language Designer project. Such a project contains a Dsl and a DslPackage project where the DslPackage project creates besides of the DslPackage.dll a vsix file to install the designer onto another machine.

Our goal will be to add some tt files to the vsix and make them includable on the installed machine.

1. Add a folder to the DslPackage project and name it TextTemplates (or any other name)
2. Add the tt files to this folder.
   • In the properties windows clear the “Custom Tool” property; otherwise Visual Studio will try to execute these templates within the DslPackage project.
   • set “Build Action” to “Content”
   • set “Include in VSIX” to “True”
3. Add a file called Additional.pkgdef to the DslPackage project with the following content and
   • set “Build Action” to “Content”
   • set “Include in VSIX” to “True”
     

     [$RootKey$\TextTemplating\IncludeFolders] [$RootKey$\TextTemplating\IncludeFolders\.tt] "IncludeMyDsl"="$PackageFolder$\TextTemplates"

     Replace “IncludeMyDsl” by a unique name for your DSL but make sure it starts with the word “Include”.

4. Edit the source.extension.tt in the DslPackage project and add the following line within the Content tag:

   <VsPackage>Additional.pkgdef</VsPackage>

With these changes the tt files from the TextTemplates folder get packaged into the vsix file and will be accessible using the <include> tag on the machine where this extension is installed.

Try switching off the win32manifest-switch in Visual Studio.

Here is the long story.

Hi! As you may already know we are working on the best .NET – ARIS interface there is so far. It’s an interface that makes heavy use of an old C-Api, using interop and p/invoke.

In Mid-December Benjamin found a very strange behavior that almost spoiled my Christmas. We had just release Version 1.1 when he called me: “Arisan crashes with Vista64.” I was shocked. “And with Vista32 as well”. I felt annihilated. We had done so much testing with various versions of XP, Vista and it worked fine with any of those OSes.

After some testing I saw that ARIS didn’t deliver any pointers at all. That’s rather bad for an interface to C that uses pointer in every call.  After one more day we had more evidence: We had switched to .NET3.5 and VS2008 some month ago and our old versions, that were compiled using the .NET2.0-compiler, still worked.

I was so terrified. The 3.5-framework should be similar to the 2.0-framework, just some (not so) little enhancements. A bug in the 3.5-compiler?

To provide a quick solution we re-wrote parts of the fancy 3.5-syntax, to make our newest enhancement compile with the 2.0-compiler again. That worked (at least we thought so) and we released version 1.1.1.

Then a very frustrating time began. We read articles over and over and quickly found this very interesting post. The problems described there where exactly what we saw.  So we followed the track of the nxcompat-flag and DEP, but it didn’t lead us anywhere.

The C-Api we interface is well documented, but it calls hundreds of other components and one of them crashed deep inside an undiscovered country.

We read pages of differences between 2.0- and 3.5-compiler, new compiler-flags (should have paid more attention here), any documentation we could find, but nothing seemed to help.

We had a feeling it could have something to do with Vista UAC, but we didn’t find any matching explanations to our problem. Being really desperate we started this thread at the MS forums. As you can read there, no one could provide a solution that worked.

I was really sad, because the last thing I wanted to do was to switch the entire development of arisan back to VS2005. I got prepared for a very sad life from now on.

But five days ago my life turned to happiness again. One more time I started of by googling words like “.NET compiler 2.0 3.5 differences” and one more time I read the new features of the 3.5-compiler. This time I obviously paid more attention to the compiler-flags and read about the /win32manifest and /nowin32manifest – switches. I didn’t have much hope left, but tried to set the /nowin32manifest switch.

It worked! Everything worked! The C-Api delivered pointers again. I tried switching the win32manifest on and off and a day later Benjamin confirmed my results. We had found it. But after a long time of suffering (for me it felt like half a year) we still were skeptical and thought we needed a deeper understanding of the problem.

The win32manifest and UAC

Finally we had something to look for and quickly found valuable information on the great I’m just sayin’ blog. The documentation of the switches helped as well, so here is a summary:

The win32manifest is not the .NET-manifest.

The win32manifest is part of the UAC (user access control). UAC was introduced with Vista as an answer to more and more security threads. One of its ideas is to keep malicious software from secretly starting to work. If you work with Vista you may have noticed a lot of dialogs that ask you for permission to run certain programs. THAT is UAC.

Now the win32manifest inside an application tells the OS what level of permission it needs to run. There are levels like “asInvoker” (default), “highestAvailable” or “requireAdministrator”. In case the OS (and the current user) can provide the requested permission, the program runs as a trusted process (according to the granted access level). Of course, 90+% (just my estimation) of the programs on you Vista-machine don’t have a win32mianifest,  because it’s a new feature and older applications just don’t have it. (And I think a majority of newer applications won’t have it either). In order to be compatible with no-win32manifest applications on one hand and providing security on the other hand, Vista can do a miracle called “virtualization”. It’s a bit like running a process in a sandbox: Vista provides anything the process may need to run: disk space, memory, a registry etc. So the process feels cosy.

But anything provided this way is separated from the main system. The process uses a copy of the registry, writes to a special parts of the HDD drives and accesses the memory in a controlled way (not sure about the memory thing…). For me this is a great achievement of Vista, almost magic. I can’t tell in detail how it works (mainly because I don’t have a clue), but it works good and doesn’t cost notable performance.

So on your Vista machine you have processes that run virtualized and others that run non-virtualized. “Non-virtualized” is also called “UCA-compatible” or just “compatible”, so I’ll adopt this wording here.

You could say that Vista trusts compatible processes a little bit more than virtualized processes. That doesn’t mean virtualized processes are evil, but.. well… maybe… under certain circumstances… not 100% trustworthy. (Here is another nice article)

Note three facts here:

1. The state (virtualized/compatible) always counts for a process and is set when the process starts. When you start an .exe-file, Vista decides how to run it.
2. Because of 1) there is no sense in attaching a win32manifest to components that do not start a process, like .dlls. Thus .dlls don’t have a win32manifest.
3. The state (virtualized/compatible) of a process doesn’t change during its lifetime.

So if a .dll is used by a virtualized process, its code also runs virtualized. And if the same .dll is used by a compatible process, its code runs compatible.

Interaction

With some of that in my mind, I could explain the problems we had with our interface software arisan: arisan is a .dll you reference in your .NET-application and use it. As mentioned above arisan calls a native (so non-.NET) .dll using p/invoke-interop-techniques. Now this native .dll starts a bunch of new processes: a database-server, a database-driver, some middle-layers, etc. I am pretty sure non of this newly started processes is based on applications with a win32manifest – so they all run virtualized.

But the .NET-applications we use to test the arisan.dll started running compatible as we switched to VS2008 and the 3.5-compiler. So our test application (compatible) called arisan.dll (compatible), arisan.dll (compatible) called C-Api.dll (compatible). C-Api.dll started new processes (virtualized !!) and called functions there. And in the end a compatible (fully trusted) process asked a virtualized (less trusted) process to fill some pointers… and that’s the point where Vista interferes and says: “Dudes, are you drunk? No way!”

It was a bit of bad luck that the clash of a virtualized and a compatible process happened so deep inside a software we couldn’t debug and only threw some strange, meaningless error-messages, but on the other hand Vista could have told us more. Maybe there is some place in Vista where the clash is logged. If someone knows – please let me know, too. Let me get this straight: Vista is absolutely right here, but I’d wish more information when it happens. I don’t exactly know how, but you could do harmful things if you could easily exchange all sorts of pointers between virtualized and compatible processes…

Solution

So finally the solution was easy. Both processes have either to run virtualized or non-virtualized. Since we can’t make a third-party software run compatible, we have to make our .NET-application run virtualized: We just need to prevent .NET from inserting a win32manifest to executables.

Using the compiler from the command-line the compiler-flag /nowin32manifest does it.

In VS2008 same switch can be found on the “Application”-tab of the project properties.

The manifest-area is enabled for ‘console applications’ and ‘Windows applications’. It is disabled for class libraries.

Another way to handle this problem is to switch off UAC completely. But we strictly advice you against this solution. UAC is one of Vista’s main features and it is designed to provide shelter for your precious data in times of evil viruses and brutal intrusion attempts

Epilog- bad fix

After we understood the problem, we saw that the ‘fix’ we provided as V1.1.1 couldn’t do much. Ok – it made the demos run, because the executable compiled with the 2.0- come without a win32manifest and thus run virtualized. But users trying it with VS2008 still had the same issues.

Actually we cannot provide a technical solution with the arisan.dll, because if the running mode depends on the executables that use arisan. So we’ll write detailed information and everyone lived happily ever after.

Today I will introduce Flow. With Flow you can model the behavior of nodes in a wireless sensor network (WSN) in a data driven and event driven manor. I developed Flow as part of my diploma thesis.

Flow is build as a Visual Studio Addon (a VS Package) on top of the DSL-Tools containing three different domain specific languages to describe different parts of the software running on a sensor node. These different DSLs are working together and using most of the techniques I described earlier in this blog. E.g. my library JaDAL was initially build to support this diploma thesis.

You will find screenshots, more explanations and Flow itself on http://flow.irgendwie.net. While the webpage and the software are available in English the thesis itself is a pdf-document and can only be downloaded in a German version.

The software is fully working and if no wireless sensor network is available you can use a node simulator to evaluate Flow. The code is released under the new BSD license and also available for download.

Screenshots:

Dataflows modeled with Flow looks like that:

The simulated sensor nodes are represented as Windows forms programmed by such dataflows:

For more information just visit http://flow.irgendwie.net.

Let’s start with a short introduction to DirectiveProcessors. DirectiveProcessors are used in .tt-files to provide data to the template. The DirectiveProcessor can take parameters from the declaration in the .tt-file and adds some code to the compiled template. This code (often properties) can be used from within the template. For example the DSL Tools are generating one DirectiveProcessor for each DSL model. In a template form the DSL Tools you will see a line like the following:

<#@ Language1 processor=“Language1DirectiveProcessor”
requires=“fileName=’Sample.mydsl1′” #>

This line uses the Language1DirectiveProcessor and provides it with one parameter (fileName). The DirectiveProcessor adds code to the template to open the given file and creates (in this case) a ExampleModel-property to use in the template code.

I created two DirectiveProcessors to use the data of simple Xml-files and Visual Studio Project files in the templates. The code of the two DirectiveProcessors is very straight forward and maybe one could advance it (e.g. make it compatible with templates written in Visual Basic).

Using the XmlFileDirectiveProcessor

Add a line like the following to your .tt-file:

<#@ XmlFile processor=“XmlFileDirectiveProcessor” FileName=“example.xml” #>

Inside this template you can access the (full qualified) filename via the this.XmlFileName-Property and the Content of this file (as a XDocument) via the this.XmlFile-Property.

Using the VsProjectFileDirectiveProcessor

Add this line to your template:

<#@ ProjectFile processor=“VsProjectFileDirectiveProcessor” FileName=“x.csproj” #>

A property named this.ProjectFile will be added to the template. This property provides you with an instance of the VsProjectFile-class containing the project file contents. This class contains only very little functionality (feel free to add some more and submit it back to me!). Just take a look at the source code of this class. The method GetAllFiles() returns a string array of all files found in the project (supporting C++ and C# project files – .vcproj and .csproj).

Setup

An entry in the registry is needed to allow the T4 system to find custom DirectiveProcessors. Just add the following entries to your registry and don’t forget to point to the right location of the JaDAL.dll.

[HKEY_CURRENT_USER\Software\Microsoft\VisualStudio\9.0Exp\
Configuration\TextTemplating\DirectiveProcessors\VsProjectFileDirectiveProcessor]
“Class”=”BenjaminSchroeter.Dsl.DirectiveProcessors.VsProjectFileDirectiveProcessor”
“CodeBase”=”D:\\JaDAL\\bin\\Debug\\JaDAL.dll”

[HKEY_CURRENT_USER\Software\Microsoft\VisualStudio\9.0Exp\
Configuration\TextTemplating\DirectiveProcessors\XmlFileDirectiveProcessor]
“Class”=”BenjaminSchroeter.Dsl.DirectiveProcessors.XmlFileDirectiveProcessor”
“CodeBase”=”D:\\JaDAL\\bin\\Debug\\JaDAL.dll”

These registry keys are for the Experimental Hive of Visual Studio. To register the DirectiveProcessors for the normal Visual Studio instance the registry key starts with HKEY_LOCAL_MACHINE\Software\Microsoft\VisualStudio\9.0
\TextTemplating\DirectiveProcessors\

I could not find any option in the DSL design to remove the model explorer, but if you look at some of the .tt files you will find somewhere a query for this.Dsl.Explorer != null. For example in the package.tt file that generates the package.cs which is responsible for registering the tool window for the model explorer:

<#
    if(this.Dsl.Explorer != null)
    {
#>
    [VSShell::ProvideToolWindow(         typeof(<#= dslName #>ExplorerToolWindow),          MultiInstances = false,          Style = VSShell::VsDockStyle.Tabbed,          Orientation = VSShell::ToolWindowOrientation.Right,          Window = "{3AE79031-E1BC-11D0-8F78-00A0C9110057}")]
    [VSShell::ProvideToolWindowVisibility(         typeof(<#= dslName #>ExplorerToolWindow),          Constants.<#= dslName #>EditorFactoryId)]
<#
    }
#>

Even if I did not find any option to set the Dsl.Explorer property of the model to null (you did not see the Explorer property anywhere in the DSL diagram) it seems that the developers of the DSL Tools had this use case in mind.

To remove the Dsl.Explorer from your DSL model open the .dsl file with a text editor and go to the end. There you will find some XML tags like the following:

<Explorer ExplorerGuid="6c276297-6acd-4e9a-8740-b61ba834004b"  Title="HardwareDescription Explorer">
    <ExplorerBehaviorMoniker      Name="HardwareDescription/HardwareDescriptionExplorer" />
</Explorer>

Just delete these three lines and generate your code once again. Maybe you should reset your Experimental Hive, too.

I tested a DSL with a removed Model Explorer without any problems. It seems the developers of the DSL Tools did a very good job on the code generation templates . The generated ModelExplorer.cs file contains only a single line:

// This source file is empty because    this DSL does not define a model explorer.

That should be a good proof that a DSL can run without the model explorer even if it might be not supported by the DSL Tools.

1.10 TypeConverters and TypeDescriptors are not picked up during the build process or during toolbox initialization.
When adding a custom TypeConverter or TypeDescriptor and then building the DSL, the TypeConvertor or TypeDescriptor is not picked up. The workaround is to rebuild the solution with a clean build.

[see Known Issues for Microsoft Visual Studio 2008 SDK 1.0]

Some time ago I posted a workaround for this problem with the TypeConverter, but today a realized that this workaround does not work with the same issue for TypeDescriptors.

Imagine some class with the following attribute:

[TypeDescriptionProvider(typeof(MyClassTypeDescriptionProvider))]
public partial class MyClass : ModelElement
{}

The corresponding TypeDescriptor (provided by the MyClassTypeDescriptionProvider class) is loaded only the first time you build or rebuild your solution. Every time you start Visual Studio after that, this attribute seems to be ignored.

Fortunately there is another way to glue TypeDescriptors to your classes using a static method of the TypeDescriptor class at runtime:

TypeDescriptor.AddProvider(
   new MyClassTypeDescriptionProvider(), typeof(MyClass));

I think a good place for this code is the static constructor of the MyClass type:

partial class MyClass
{
    static MyClass()
    {
        TypeDescriptor.AddProvider(
            new MyClassTypeDescriptionProvider(),
            typeof(MyClass));
    }
}

Another interesting point is: Not only custom TypeDescriptors you write yourself are affected by this problem, also the TypeDescriptors that are generated by the DSL-Tools from your DslDefintion have to struggle with it:

In the DSL Explorer you can define custom TypeDescriptors for each Domain Class and each Shape. Even these TypeDescriptors will not be loaded after the first run. In other words: the definition of TypeDescriptors in the DSL Explorer is pretty useless as long as you do not add the three lines of code to each class. Of course this is something one can automate! I stole some code from PropertiesGrid.tt and added a few lines to create the RegisterTypeDescriptor.tt. Just add this file to your Dsl Project in the GeneratedCode folder and all TypeDescriptors defined in the DSL Explorer will be loaded every time you start your project.

You don’t find bugs in CodeVeil when you try to find them, but half a year ago we found this one by chance. CodeVeil veils well, but the resulting code throws a mighty exception.

You don’t need spectacular code to lure the bug out of it’s hole – here is the prelude:

private static void Main(string[] args)
{
    // Get list and fill it with values
    List<Capsule<int>> l = new List<Capsule<int>>();
    l.Add(new Capsule<int>(23));
    l.Add(new Capsule<int>(42));

    // Have it sorted
    List<Capsule<int>> l2 = GetSortedReturnList(l);

    // Show result
    foreach (Capsule<int> ci in l2)
        Console.WriteLine(ci.Item);

    Console.ReadLine();
}

A simple list of two ints that are encapsulated in a generic type. The generic type just makes sure that two items can be compared:

internal class Capsule<T> : IComparable<Capsule<T>>
{
    public T Item;

    public Capsule(T i)
    {
        Item = i;
    }

    public int CompareTo(Capsule<T> other)
    {
        return Item.ToString().CompareTo(other.Item.ToString());
    }
}

And this code causes the exception when being executed by the veiled assembly:

private static List<Capsule<T>> 

    GetSortedReturnList<T>(List<Capsule<T>> list)
{           
    int sortOrder = 1;
    list.Sort(
        delegate(Capsule<T> a, Capsule<T> b) 

          { return sortOrder * a.CompareTo(b); });

    return (list);
}

The sore point is the sortOrder variable being multiplied with the result of a.CompareTo(b) used in an anonymous method. Okay, you wouldn’t expect this code from a beginner, but it’s not rocket-science either. So we were very astonished and it took us quite some time to track this down and to believe our eyes. Maybe the problem with CodeVeil is related to the internal representation of anonymous method. When you use a constant instead of a variable

const int sortOrder = 1;

the problem disappears. I guess the compiler eliminates the constant and CodeVeil will not see it. But it remains, when you use a Linq-notation here:

list.Sort((a, b) => a.CompareTo(b)*sortOrder);

As said before we found this bug half a year ago. We contacted Xheo and received a very quick and friendly reply, but no solution. Seasons went by and several new versions of CodeVeil were released; but this bug is still present. At least in the stand-alone version.

But there is hope: The new beta preview of the DeployLX suite contains an obviously new version of CodeVeil and the bug is finally gone. And – by the way – this new version seems to offer a nice package of new feature and improved power! So, please, Mr. Xheo, release a new stand-alone-version of CodeVeil based on the new beta preview!