RE: Would/Can Property's override be configable insteads of hardc ode?

[Jose Alberto Fernandez <JF...@viquity.com>]

One kind of override blocking that makes little sense is
blocking on sub-project calls. I see no reason why sub-projects
should not be able to override (locally) properties set on the caller
project. After all the meaning of those properties may be quite different
in the subprojects.

The only possible exclussion to this are the properties set as part of
the <ant> task which should behave as a -D definition for the subproject.

With this change, a lot of the structured property manipulation that some
people are trying to do can be done fairly easy. Without changing the core 
at all.

Jose Alberto


> -----Original Message-----
> From: Peter Donald [mailto:donaldp@mad.scientist.com]
> Sent: Thursday, August 03, 2000 9:20 PM
> To: ant-dev@jakarta.apache.org
> Subject: Re: Would/Can Property's override be configable insteads of
> hardcode?
> 
> 
> At 11:55  4/8/00 +0800, you wrote:
> >Would/Can Property's override be configable insteads of hardcode?
> 
> debate about this raged on dev list a while back. I think the 
> consensus was
> that everything gets moved to interpretation time rather than 
> parse time. I
> don't know what the opinion was on overiding properties but I believe
> people wanted to overide them.
> 
> It should be resolved in next release :P
> 
> Cheers,
> 
> Pete
> 
> *------------------------------------------------------*
> | "Nearly all men can stand adversity, but if you want |
> | to test a man's character, give him power."          |
> |       -Abraham Lincoln                               |
> *------------------------------------------------------*
>