[boost] Re: In-place construction for variant

From: Eric Friedman (ebf_at_[hidden])
Date: 2003-10-28 22:15:10

David Abrahams wrote:

  > Before we go adding new components just to satisfy discomfort at the
> idea that heap storage might be used in some cases, we should ask
> ourselves whether anyone has a use case for fixed_variant.  The only
> thing it provides over the existing variant design is a runtime check
> that you're not changing the type it contains.  It doesn't even
> provide a compile-time check.  IMO that is not worth implementing a
> new component for.

What you describe isn't what I had in mind for fixed_variant.

There will in fact be a compile-time check because fixed_variant will 
*not* provide any operator=. Instead, what I meant by the content 
remaining mutable is that it will still be possible to change the 
content by visitation. (Contrast this with a const variant<...>.)

Sorry if I wasn't clear. Anyway, this isn't going to happen before 1.31, 
so there will be plenty of time to discuss the idea further.

Eric