ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jose Alberto Fernandez" <j_a_fernan...@yahoo.com>
Subject RE: Property setting Policy.
Date Mon, 18 Jun 2001 15:13:53 GMT
> From: Stefan Bodewig [mailto:bodewig@apache.org]
>
>
> Peter Donald <donaldp@apache.org> wrote:
>
> > ie I think most people agree that immutable properties are better
> > but some people have requested mutability of properties.
>
> We could solve it the same way we do in Ant1 - they are mutable in
> Ant1, but people don't know it 8-)
>
> > If were to enforce the immutability in core they could not get
> > around that when they think we are wrong ;)
>
> We can be wrong?
>
> Seriously, what are the parts of a "property setting policy" that we
> need to enforce and which can we deal with more freely?  Maybe we
> should define the policy first?
>

Once we allow ${foo} to refer to things other than properties (i.e.,
<typedef>s) users should be able to define their own objects with watever
behavior they may want to have. It is up to their code. I see no reason to
abandon ANT's property inmutability.

Actually, I would vote for enforcing it at the programatic API level. So
that we have a clear understanding of how regular properties are supposed to
act. If users on some obscure escenario need something else, it can be
provided by its own type definition with its own rules.

Jose Alberto

> Stefan
>


Mime
View raw message