Re: Subobject constructors

• To: cxx-abi@xxxxxxxxxxxx, Mark Mitchell <mark@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>
• Subject: Re: Subobject constructors
• From: Jim Dehnert <dehnert@xxxxxxxxxxxxxxxxxxxx>
• Date: Wed, 5 Jul 2000 01:28:00 -0700 (PDT)

> From: Mark Mitchell <mark@xxxxxxxxxxxxxxxx>
> 
> The wording in 3.3 about subobject constructors is misleading:
> 
>   If the not-in-charge constructor calls constructors for base class
>   subobjects that do not need construction vtables, the construction
>   vtable parameter is not passed to the base class subobject
>   constructor, and the base class subobject constructors use their
>   complete object vtables for initialization.
> 
> This is also true when an in-charge constructor is making the call.
> There is never a need to pass a VTT to a base class subobject that
> doesn't have virtual bases.

See if it looks better now...

Jim

-	    Jim Dehnert		dehnert@xxxxxxx
				(650)933-4272

• Follow-Ups:
  • Re: Subobject constructors
    • From: Mark Mitchell

• Prev by Date: Re: Pure virtual functions
• Next by Date: Re: Covariant returns
• Previous by thread: Subobject constructors
• Next by thread: Re: Subobject constructors
• Index(es):
  • Date
  • Thread