>
> if( lOleError == S_OK )
> lOleError = CoCreateInstance( HB_ID_REF( ClassID ), NULL,
> CLSCTX_SERVER, fIID ? HB_ID_REF( iid ) : HB_ID_REF( IID_IDispatch ), (
> void** ) ( void * ) &pDisp );
> }
> + else if( ISNUM( 1 ) )
> + {
> + pDisp = ( IDispatch * ) ( HB_PTRDIFF ) hb_parnint( 1 );
> +#if HB_OLE_C_API
> + lOleError = ( HRESULT ) pDisp->lpVtbl->AddRef( pDisp );
> +#else
> + lOleError = ( HRESULT ) pDisp->AddRef();
> +#endif
> + }
> else
> lOleError = CO_E_CLASSSTRING;
> hb_setOleError( lOleError );
>
> but there is a question if we want to still tolerate code with
> C pointer <-> HVM number item conversions and keep workarounds
> for it. I'm leaving the decision to Mindaugas.
IMO for compatibility we should try to leave it for now and remove
it later when 3rd parties have adapted.
I've checked whole codebase and found a few occurrences of these
constructs in hbwin (misc/dll/prn) and a lot in GTWVG and WAPI
headers.
I'd be glad to delete them from hbwin. If there are no objections,
I'll do it, so that we can see what effect will it have on users.
Brgds,
Viktor
_______________________________________________
Harbour mailing list
Harbour@harbour-project.org
http://lists.harbour-project.org/mailman/listinfo/harbour
