Directx clone function (com/ole knowledge)

Posted on 1998-12-21
Last Modified: 2013-11-25
i added the documentation of the clone function of directx
i don't understand the third parameter.
it might be connected to com/ole stuff. ( i assume it does.)

can you help ?


Creates a copy of an object.


*ppvObj  );

Pointer to the outer IUnknown interface that allows COM aggregation
Identifier of the object being copied.
Address that will contain the copy of the object when the method returns.

Question by:hed
  • 2
  • 2
LVL 86

Accepted Solution

jkr earned 50 total points
ID: 1180772
It indeed has to do with COM. The mechanism is like this:
IDispatch* m_pObj;

    m_sc    =   ::CoCreateInstance  (   m_clsid,    
                                        ( void**) &m_pObj

In other words: You pass in the address of a pointer in which the address of the newly created object will be stored...

Author Comment

ID: 1180773
quite stupid of me.
the third parameter is obvious.
it's the first parameter is the one that puzzles me.
what is
Pointer to the outer IUnknown interface that allows COM aggregation features.
sorry again.
raising to 30.
LVL 86

Expert Comment

ID: 1180774
The 'outer' IUnknown is used for 'aggregation' (that's the way COM realizes inheritance).
From the VC5 docs:
Begin quote-->
Aggregation is the other, richer, reuse mechanism, in which the outer object exposes interfaces from the inner object as if they were implemented on the outer object itself. This is useful when the outer object would always delegate every call to one of its interfaces to the same interface in the inner object. Aggregation is actually a specialized case of containment/delegation, and is available as a convenience to avoid extra implementation overhead in the outer object in these cases.

Aggregation is almost as simple to implement as containment is, except for the three IUnknown functions: QueryInterface, AddRef, and Release. The catch is that from the client’s perspective, any IUnknown function on the outer object must affect the outer object. That is, AddRef and Release affect the outer object and QueryInterface exposes all the interfaces available on the outer object. However, if the outer object simply exposes an inner object’s interface as its own, that inner object’s IUnknown members called through that interface will behave differently than those IUnknown members on the outer object’s interfaces, an absolute violation of the rules and properties governing IUnknown.

The solution is that aggregation requires an explicit implementation of IUnknown on the inner object and delegation of the IUnknown methods of any other interface to the outer object’s IUnknown methods.

Creating Aggregable Objects
Creating objects that can be aggregated is optional; however, it is simple to do and to do so has significant benefits. The following rules apply to creating an aggregable object:

· The aggregable (or inner) object’s implementation of IUnknown::QueryInterface, AddRef, and Release controls the inner object’s reference count, and this implementation must not delegate to the outer object’s unknown (the controlling IUnknown).

· The QueryInterface, AddRef, and Release methods of all other interfaces implemented on the inner object must delegate to the controlling IUnknown and not directly affect the inner object’s reference count.

· The inner IUnknown must implement QueryInterface only for the inner object.

· The aggregable object must not call AddRef when holding a reference to the controlling IUnknown pointer.

· When the object is created, if any interface other than IUnknown is requested, the creation must fail with E_UNKNOWN.

The code fragment below illustrates a correct implementation of an aggregable object using the nested class method of implementing interfaces:

// CSomeObject is an aggregable object that implements
// IUnknown and ISomeInterface
class CSomeObject : public IUnknown
        DWORD        m_cRef;         // Object reference count
        IUnknown*    m_pUnkOuter;    // Controlling IUnknown, no AddRef
        // Nested class to implement the ISomeInterface interface
        class CImpSomeInterface : public ISomeInterface
            friend class CSomeObject ;
                DWORD    m_cRef;    // Interface ref-count, for debugging
                IUnknown*    m_pUnkOuter;    // controlling IUnknown
                CImpSomeInterface() { m_cRef = 0;   };
                ~ CImpSomeInterface(void) {};
                // IUnknown members delegate to the outer unknown
                // IUnknown members do not control lifetime of object
                STDMETHODIMP     QueryInterface(REFIID riid, void** ppv)
                {    return m_pUnkOuter->QueryInterface(riid,ppv);   };
                STDMETHODIMP_(DWORD) AddRef(void)
                {    return m_pUnkOuter->AddRef();   };
                STDMETHODIMP_(DWORD) Release(void)
                {    return m_pUnkOuter->Release();   };
                // ISomeInterface members
                STDMETHODIMP SomeMethod(void)
                {    return S_OK;   };
        } ;
        CImpSomeInterface m_ImpSomeInterface ;
        CSomeObject(IUnknown * pUnkOuter)
            // No AddRef necessary if non-NULL as we're aggregated.
        } ;
        ~CSomeObject(void) {} ;
        // Static member function for creating new instances (don’t use
        // new directly).Protects against outer objects asking for interfaces
        // other than IUnknown
        static HRESULT Create(IUnknown* pUnkOuter, REFIID riid, void **ppv)
            CSomeObject*        pObj;
            if (pUnkOuter != NULL && riid != IID_IUnknown)
                return CLASS_E_NOAGGREGATION;
            pObj = new CSomeObject(pUnkOuter);
            if (pObj == NULL)
                return E_OUTOFMEMORY;
            // Set up the right unknown for delegation (the non-aggregation case)
            if (pUnkOuter == NULL)
                pObj->m_pUnkOuter = (IUnknown*)pObj ;
            HRESULT hr;
            if (FAILED(hr = pObj->QueryInterface(riid, (void**)ppv)))
                delete pObj ;
            return hr;
        // Inner IUnknown members, non-delegating
        // Inner QueryInterface only controls inner object
        STDMETHODIMP QueryInterface(REFIID riid, void** ppv)
            if (riid == IID_IUnknown)
            if (riid == IID_ISomeInterface)
            if (NULL==*ppv)
                return ResultFromScode(E_NOINTERFACE);
            return NOERROR;
        } ;
        STDMETHODIMP_(DWORD) AddRef(void)
        {    return ++m_cRef; };
        STDMETHODIMP_(DWORD) Release(void)
            if (--m_cRef != 0)
                return m_cRef;
            delete this;
            return 0;
Aggregating Objects
When developing an aggregable object, the following rules apply:

· When creating the inner object, the outer object must explicitly ask for its IUnknown.

· The outer object must protect its implementation of Release from reentrancy with an artificial reference count around its destruction code.

· The outer object must call its controlling IUnknown’s Release if it queries for a pointer to any of the inner object’s interfaces. To free this pointer, the outer object calls its controlling IUnknown’s AddRef, followed by Release on the inner object’s pointer.

// Obtaining inner object interface pointer
pUnkInner->QueryInterface(IID_ISomeInterface, &pISomeInterface);
// Releasing inner object interface pointer
· The outer object must not blindly delegate a query for any unrecognized interface to the inner object, unless that behavior is specifically the intention of the outer object.
<<-- end quote

Author Comment

ID: 1180775
Adjusted points to 50

Featured Post

Salesforce Has Never Been Easier

Improve and reinforce salesforce training & adoption using WalkMe's digital adoption platform. Start saving on costly employee training by creating fast intuitive Walk-Thrus for Salesforce. Claim your Free Account Now

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

With most software applications trying to cater to multiple user needs nowadays, the focus is to make them as configurable as possible. For e.g., when creating Silverlight applications which will connect to WCF services, the service end point usuall…
I made this because I wanted to get e-mail with a attached csv file so I'd would be able to import user input into a MS Excel template, but I also wanted to register/save all inputs from each day in a file on the server. 1st - It creates a temp C…
The purpose of this video is to demonstrate how to set up the WordPress backend so that each page automatically generates a Mailchimp signup form in the sidebar. This will be demonstrated using a Windows 8 PC. Tools Used are Photoshop, Awesome…
The viewer will be introduced to the technique of using vectors in C++. The video will cover how to define a vector, store values in the vector and retrieve data from the values stored in the vector.

740 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question