Re: [jamboost] Property utilities

From: Vladimir Prus (ghost_at_[hidden])
Date: 2002-05-03 07:56:53

David Abrahams wrote:
> ----- Original Message -----
> From: "Vladimir Prus" <ghost_at_[hidden]>
>
> > I'm now thinking that compose-requirements should really be the same
>
> as
>
> > apply-requrements. I believe that we should treat inherited
>
> requirements in
>
> > the same "soft" manner. For example, if parent has <optimization>off
> > and current project require <optimization>space, we just ignore
> > <optimization>off. If parent has <rtti>on and current project require
> > <rtti>off, we just refuse to accept such requirements. This is the
>
> semantics
>
> > I suggest in
> > http://groups.yahoo.com/group/jamboost/message/819
> >
> > Anybody has any objections?
>
> No, that sounds fine. However, we need a better name. I would like
> something which suggests which of the requirements lists gets precedence
> in case of link-compatible options. Ideas:
>
> override-requirements
> refine-requirements
>
> I like #2. At least there's a vague implication that the old properties
> come first and the additional ones come 2nd (but only vague). Can we do
> better?

Now that compose-requirements is found to be the same as apply-requirements, 
can we go with 'apply-requirements' only?

- Volodya