 1. Make sure your Com is public and visible. To do this, add these
    tags to your Com class:

        [ClassInterface(ClassInterfaceType.AutoDual)]
        [Guid(&quot;41E90F3E-56C1-4633-81C3-6E8BAC8BDD70&quot;)]
        [ProgId(&quot;..........&quot;)]
        [ComVisible(true)] 
        public class MyCom
        {

 2. Make sure your COM has been registered. You can do this using the command line:

 

        C:\WINDOWS\Microsoft.Net\Framework\v4.0.30319\regasm &quot;C:\.......\xxx.dll&quot;

Once before, I was certain that you couldn&#39;t do this, but the other day I was playing around with some code and it seemed to compile and work. I just want to verify that I am not just getting lucky. Can a template class have a pure virtual function - which I guess would also mean just plain virtual methods would be valid as well for the destructor?

    template &lt;typename WordType&gt; class DataSource
    {
    public:
        DataSource();
        DataSource(DataSource const&amp; other);
        virtual ~DataSource();

        virtual void Put(
            WordType const* const data,
            unsigned int const wordCount) = 0;
    }

I&#39;ve tried looking it up online and all that I&#39;ve been able to find is that you cannot have a virtual method (pure or otherwise) in a normal class such as this:

    class DataSource
    {
    public:
        DataSource();
        DataSource(DataSource const&amp; other);
        virtual ~DataSource();

        template &lt;typename WordType&gt;
        virtual void Put(
            WordType const* const data,
            unsigned int const wordCount) = 0;
    }

And that this is due to the imposibility of managing a virtual table to reference all the different types of possible types this method would be instanciated with.

However, when it came to a virtual member function of a template class, it seems to be different because the whole class itself is &quot;created&quot; via the template parameter when the template class variable is instanciated. At this point, the virtual method is just like any other virual method of a class due to the &quot;find-and-replace&quot; nature of templates.

Anyway, stating the question again in case it got lost in there:
Are virtual (pure and/or normal) virtual functions allowed within a tempate class?

Are pure virtual methods allowed within a template class?

With Grails there are several ways to do the same thing.

Finds all of domain class instances:

    Book.findAll()
    Book.getAll()
    Book.list()
    
Retrieves an instance of the domain class for the specified id:

    Book.findById(1)
    Book.get(1)

When do you use each one? Are there significant differences in performance?

Difference between findAll, getAll and list in Grails

I had created a COM+ domain partition then mapped it to a Windows 2008 server machine and imported a COM+ application into it.

I tried using the following C# code to activate an object from that specific partition on the server remotely:

&lt;!-- language: lang-csharp --&gt;

    //partition guid
    Guid guidMyPartition = new Guid(&quot;41E90F3E-56C1-4633-81C3-6E8BAC8BDD70&quot;);
    //parition moniker
    string uri= &quot;partition:{&quot; + guidMyPartition + &quot;}/new:MyObject&quot;;
    Type t = Type.GetTypeFromProgID(&quot;MyObject&quot;, &quot;MyServer&quot;);
    MyObject obj = (MyObject)Activator.GetObject(t, uri);

But I get this exception:
&gt; Cannot create channel sink to connect to URL &#39;partition:{41e90f3e-56c1-4633-81c3-6e8bac8bdd70}/new:MyObject&#39;. An appropriate channel has probably not been registered.

Does anybody know how such an activation can be accomplished?



COM+ object activation in a different partition

I had created a COM+ domain partition then mapped it to a Windows 2008 server machine and imported a COM+ application into it.
I tried using the following C# code to activate an object from that specific partition on the server remotely:

<pre><code class="hljs language-ini">//partition guid
Guid guidMyPartition = new Guid("41E90F3E-56C1-4633-81C3-6E8BAC8BDD70");
//parition moniker
string uri= "partition:{" + guidMyPartition + "}/new:MyObject";
Type t = Type.GetTypeFromProgID("MyObject", "MyServer");
MyObject obj = (MyObject)Activator.GetObject(t, uri);
</code></pre>
But I get this exception:
> Cannot create channel sink to connect to URL 'partition:{41e90f3e-56c1-4633-81c3-6e8bac8bdd70}/new:MyObject'. An appropriate channel has probably not been registered.
Does anybody know how such an activation can be accomplished?

I am holding some information in static variables defined in page class (not in Global.asax). I only declare variable in code like:

    protected static int SomeGlobalUnsecureID;
    protected static string SomeGlobalUnsecureString;

and define the variable in PageLoad event. For example, I check ID from the database, if it&#39;s different from SomeGlobalUnsecureID, I update SomeGlobalUnsecureID and String from somewhere else, otherwise leave them as is. This is perfectly safe in my app. logic (i.e. those data are not secure, everybody can access them, no problem); only thing I want to accomplish is 

 1. Hold the same amount of memory regardless of users connected
    
 2.   Change if and only if persistent info is different from the one in
    &#39;memory&#39; (because actually reading the string is time consuming for
    me.

Now, since I make the check in PageLoad, I have no problems in reloaded pages. However my page is full of WebMethods, and sometimes I see that the static variables are zeroed. And the strange part is; the session is still active even when the static variables are zeroed (so-&gt; no server or app. pool restart etc.)

This is really strange for me. I assume that static variable will hold its value until the application (somehow) ends. But even the Session did not expire, the static variable is zeroed. What do you suggest? Is using application variables a better choice? All documents I&#39;ve read on web suggest static variables instead of application variables, do I need to declare them somehow different?




Lifetime of ASP.NET Static Variable

I have a Web.config with several ConnectionStrings

    &lt;connectionStrings&gt;
        &lt;add name=&quot;connStr1&quot; connectionString=&quot;...
        &lt;add name=&quot;ConnStr2&quot; connectionString=&quot;...
        &lt;add name=&quot;connStr3&quot; connectionString=&quot;...

Is there a way using config transformations to remove a specific connectionstring? Something Like:

    &lt;connectionStrings&gt;
        &lt;xdt:Remove connStr2?

Obviously no where near the correct syntax, but you get my drift...

How to remove a ConnectionString using Config Transformations

One of the ConfigurationProperty I have in my ConfigurationSection is an ENUM. When .net parses this enum string value from the config file, an exception will be thrown if the case does not match exactly. 

Is there away to ignore case when parsing this value?

.net Custom Configuration How to case insensitive parse an enum ConfigurationProperty

I have a C# ASP.NET application which starts about 25 different threads running some methods in a class called SiteCrawler.cs. 

In `HttpContext.Current.Session` I want to save the result of the search made by the user and present it to the user when all the threads are finished running. My problem is that the `HttpContext.Current` object is null in the spawned threads because it doesn&#39;t exist there. 

What other options do I have to save user/session specific data without using session because of the limitations when the application is multithreaded?

I have tried to search around every inch of Stackoverflow to find a solution but without any luck....

Access HttpContext.Current from different threads

    public interface IDic
    {
        int Id { get; set; }
        string Name { get; set; }
    }
    public class Client : IDic
    {

    }


How can I cast `List&lt;Client&gt;` to `List&lt;IDic&gt;`?

Cast List&lt;T&gt; to List&lt;Interface&gt;

I&#39;m having an issue with a COM based client-server setup. The COM server is written in C# (.NET 4.0) and runs as a (registered) local server.

Depending on which application connects to the server, other clients will receive a **Server execution failed (Exception from HRESULT: 0x80080005 (CO_E_SERVER_EXEC_FAILURE)**

The underlying issue is explained [here (in the section COM is integrity aware)][1]. The way I understand it, it is being caused by the fact that an elevated application creates the server with a higher integrity level. When another non-elevated application then connects, it is not allowed to connect to the same instance. The same happens when a non-elevated application creates the process, followed an elevated application connecting.

I&#39;ve tried to implement the solution described on the [page][2]: modifying the registry to set a security descriptor that should allow all clients to connect. There is a [code sample][3] in C++, but this does effectively the same thing in .NET:

    // Security Descriptor with NO_EXECUTE_UP
    var sd = new RawSecurityDescriptor(&quot;O:BAG:BAD:(A;;0xb;;;WD)S:(ML;;NX;;;LW)&quot;);
    byte[] securityDescriptor = new Byte[sd.BinaryLength];
    sd.GetBinaryForm(securityDescriptor, 0);

    RegistryKey key = Registry.ClassesRoot.OpenSubKey(&quot;AppID\\{APP-ID-GUID}&quot;, true);
    if (key == null)
    {
        key = Registry.ClassesRoot.CreateSubKey(&quot;AppID\\{APP-ID-GUID}&quot;);
    }

    using (key)
    {
        key.SetValue(&quot;LaunchPermission&quot;, securityDescriptor, RegistryValueKind.Binary);
    }

However, this does not have the desired effect. When the second client tries to create an instance of the object in question, Windows tries to launch a separate instance of my COM Server, but the server prevents two instances from running as the same user. Given the permissions I&#39;ve set, I would not expect a second instance to launch in the first place.

Since one of the client applications is running in Medium IL and the other in High IL, I also experimented with variants on the [mandatory label][4], like:

    O:BAG:BAD:(A;;0xb;;;WD)S:(ML;;NX;;;ME)
    O:BAG:BAD:(A;;0xb;;;WD)S:(ML;;NX;;;LW)(ML;;NX;;;ME)(ML;;NX;;;HI)

I&#39;ve also tried setting the `ROTFlags` registry key to 0x1 (ROTFLAGS_ALLOWANYCLIENT) as suggested on the page, still no change in behavior.

I&#39;ve established that the LaunchPermission registry value is being used in some way. I cannot discover where it&#39;s being read using Process Monitor, but when I use the `dcomcnfg.exe` tool to set the same key, I can force the server to fail loading by denying launch permissions.

I would like to point out that my server process does not need elevation. How do I make both elevated and non-elevated processes capable of connecting to a single server instance?


  [1]: http://msdn.microsoft.com/en-us/library/bb625962.aspx
  [2]: http://msdn.microsoft.com/en-us/library/bb625962.aspx
  [3]: http://msdn.microsoft.com/en-us/library/ms679687.aspx
  [4]: http://msdn.microsoft.com/en-us/library/bb625958.aspx
  [5]: http://msdn.microsoft.com/en-us/library/ms679687.aspx

Allowing connection to .NET COM server with mismatching integrity level

I can fire up Excel with the following script. But in ghci (7.4.1) I have a segmentation fault when I run it. 

I don&#39;t know where to search from now. I don&#39;t have this error if I remove the line 

    workSheets &lt;- workBook #  propertyGet_0 &quot;Worksheets&quot;

Here is the code. May be I forgot something. I read the source code of com.hs [here][1], but it doesn&#39;t give me any clue.  

&lt;!-- language: lang-hs --&gt;

    import System.Win32.Com 
    import System.Win32.Com.Automation
    --
    -- createObjectExcel 
    -- coming from Automation.hs and com.hs
    --

    iidIDispatch_unsafe  = mkIID &quot;{00020400-0000-0000-C000-000000000046}&quot;

    createObjExl :: IO (IDispatch ()) 
    createObjExl = do
        clsidExcel &lt;- clsidFromProgID &quot;Excel.Application&quot;
        pExl &lt;- coCreateInstance clsidExcel  Nothing LocalProcess iidIDispatch_unsafe
        return pExl


    fichierTest2 = &quot;E:/Programmation/haskell/Com/qos1.xls&quot;

    main = coRun $ do 
        pExl &lt;- createObjExl
        workBooks &lt;- pExl #  propertyGet_0 &quot;Workbooks&quot;
        workBook &lt;- workBooks #  propertyGet_1 &quot;Open&quot; fichierTest2
        workSheets &lt;- workBook #  propertyGet_0 &quot;Worksheets&quot;

        workBooks # method_1_0 &quot;Close&quot; (0::Int)
        pExl # method_0_0 &quot;Quit&quot;

        mapM release [workSheets,workBook, workBooks, pExl]

-------
**Edit** with Gonzalez&#39;s advice I tried debugging, but no information arose. I tried the code by hand in ghci, and it seemed that the guilty party was the release function. 

When I entered these in ghci, I got the segmentation fault: 

    *Main&gt; coInitialize
    *Main&gt; pExl &lt;- createObjExl
    *Main&gt; release pExl
    0

Now if I hit &quot;pExl&quot; I have a reference. Shouldn&#39;t that be set to Null? 

    *Main&gt; pExl
    &lt;interface pointer = 0x020844cc&gt;

    *Main&gt; coUnInitialize
    *Main&gt; :q
    leaving Ghci
    Segmentation Fault/access violation ...


  [1]: http://hackage.haskell.org/package/com-1.2.3.1

Excel Automation with Haskell gives a seg fault

I want to be able to have two instances which are completely independent in the sense that I can be working on two separate unrelated projects in different folders without any interference.

Content Type	Original Author	Original Content on Stackoverflow
Question	Khaled Saleh	View Question on Stackoverflow
Solution 1 - C#	mtz1406	View Answer on Stackoverflow

COM+ object activation in a different partition

C# Problem Overview

C# Solutions

Solution 1 - C#

Difference between findAll, getAll and list in Grails

Are pure virtual methods allowed within a template class?

Attributions