ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefan Bodewig <bode...@apache.org>
Subject Re: Order of attributes versus elements
Date Mon, 04 Mar 2002 12:27:03 GMT
On Sun, 03 Mar 2002, Conor MacNeill <conor@cortexebusiness.com.au>
wrote:
> Stefan Bodewig wrote:
>
>> So, I think Ant should give guarantee some behavior, which order is
>> to be preferred may be open for discussion in the Ant2 context, but
>> not really for Ant1.  Do you think the order is wrong?
> 
> I find it more natural the other way around.

I agree that it is totally confusing in the context of addConfigured
as the child elements get their attributes handled before the parent
element.

I'm not sure for the other cases, all objects would be created and
after that all attributes reflected in, which has its own kind of
logic 8-)

We can certainly change that for Ant2, I'm not against it at all.

Following Adam explaining how it work in Myrmidon, Myrmidon would have
a hard time to give any guarantee as the separation between child
elements and attributes is not really existent any longer.  You seem
to prefer the no-guarantee aproach as well.  I'm fine with it.

Is anybody against marking the order explicitly as undefinded in Ant2.

> Ah, OK. Unfortunately this particular behaviour makes it very
> difficult to replace a built-in task with an enhanced
> version.

I know, when I had to change the behavior of task overriding (shortly
after we added the manifest task.  FOP had its own mainfest task and
that didn't work any longer because Ant's built-in task was used
during the configuration phase) I explicitly stated that this wouldn't
work for task that defined child elements.

> The ids don't seem that useful anyway. If they refer to a non
> top-level object it probably won't have been configured anyway
> unless it has been evaluated. I feel perhaps we should just move to
> always using UnknownElement for non-toplevel tasks. That would do
> away with all the task invalidation hack.

They would - but in a non backwards-compatible way.

> What a tangled web we weave :-)

You don't expect me to disagree, I hope 8-)

Stefan

--
To unsubscribe, e-mail:   <mailto:ant-dev-unsubscribe@jakarta.apache.org>
For additional commands, e-mail: <mailto:ant-dev-help@jakarta.apache.org>


Mime
View raw message