1

I have good knowledge in C# and C++. But I am totally new to COM programming. I am converting some code from VC++ to C#. But VC++ code involves a lot of COM keywords like STDMETHODIMP, STDMETHODCALLTYPE, __declspec, HRESULT, and other COM Programming concepts. I didn't get what are these COM terminologies.

But since I can understand C++, I am able to understand the logic involved in the VC++ code and convert this logic to C# implementation IGNORING the COM terminologies.

Some functions in my VC++ code are having the returning type STDMETHODIMP and they are just returning S_OK or E_FAIL etc.. How shall I proceed with my conversion to C#? What should I return in my C# code for these VC++ functions which are returning S_OK E_FAIL and what return type should I use in my C# code instead of STDMETHODIMP?

In fact there are many such COM terminologies in my VC++ code like "skippedEntity()" "processingInstruction()" etc.. In fact this VC++ code is making use of a DLL called "msxml6.dll". And implementing the interface methods in it. So, is it ok if I IGNORE these COM Terminologies and proceed implementing the logic invloved in VC++ code to C#.

Kindly let me know same..

Since COM is an Alien concept to me, sorry if I am asking some silly doubt or meaningless doubt..

Thanks In Advance..

Improve this question
0

2 Answers 2

Reset to default
2

The uppercase identifiers are macros, they ensure that the COM interface function has the proper signature. A COM function that's compatible with automation must use the correct calling convention, __stdcall, and must return a long (an hresult) that indicates the error or success status. Additional requirements are that the COM interface must implement IUnknown and that the function argument types should restrict themselves to automation compatible types.

The COM interop support in the CLR ensures that these requirements are met. It generates a v-table for a C# interface that automatically implements IUnknown, you don't have to implement it yourself. The HRESULT return value is automatically generated, it is mapped from an Exception. Function arguments are automatically marshaled from their unmanaged type to the equivalent native type.

The argument type translation is a rather large topic and covered in any decent book about the subject (like Adam Nathan's). Suffice it to say that any automation compatible type has a corresponding managed type. The harder ones are object to VARIANT, string to BSTR and array to SAFEARRAY.

You use the [ComVisible] attribute in C# to create a COM server. Several other ones are relevant, like [Guid] and [InterfaceType]. This is well covered in any examples. You create the type library with Tlbexp.exe or, more commonly, with Regasm.exe /tlb. The latter tool is the one that's required to register the server, you cannot use Regsvr32.exe anymore.

Improve this answer
Sign up to request clarification or add additional context in comments.

Comments

0

Most of that stuff is handled automatically if you add a com-reference in your project (or import with TLBIMP).

S_OK and S_FAIL are converted to exceptions. See MSDN.

Improve this answer

Comments

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.