Re: cvs commit: jakarta-ant-myrmidon/antlib/src/java/org/apache/antlib/core ForEachTask.java

[Peter Donald <pe...@apache.org>]

At 02:40 PM 6/7/2002 +1000, you wrote:
>That's probably true - but we don't know that yet.  Let's keep the task until
>we actually *do* have templating.  Then, we can compare them for usability.
>It may well turn out that they both have a place.

I would prefer not to put another hack in unless it is furthering 
development which I don't believe the task would.


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: cvs commit: jakarta-ant-myrmidon/antlib/src/java/org/apache/antlib/core ForEachTask.java

[Adam Murdoch <ad...@apache.org>]

On Fri, 7 Jun 2002 17:00, Peter Donald wrote:
> At 04:46 PM 6/7/2002 +1000, you wrote:
> >On Fri, 7 Jun 2002 14:51, Peter Donald wrote:
> > > At 02:40 PM 6/7/2002 +1000, you wrote:
> > > >That's probably true - but we don't know that yet.  Let's keep the
> > > > task until we actually *do* have templating.  Then, we can compare
> > > > them for usability. It may well turn out that they both have a place.
> > >
> > > I would prefer not to put another hack in unless it is furthering
> > > development which I don't believe the task would.
> >
> >I don't understand why the <for-each> task is a hack.  Seems like a
> >straightforward operation, really:  Do some stuff for each member of a
> > list.
>
> Essentially as its use in any real context basically requires recursive
> property resolution which has been vetoed many a time on ant-dev ;)

With mutable properties, and the improved property scoping we have now, you 
should be able to do all kinds of good stuff without resorting to recursive 
property resolution.   

I do agree that <for-each> is a crappy solution to templating.  But templates 
and iteration aren't the same thing, there's just a big overlap in the kinds 
of problems they can solve.  I want to keep <for-each>, because I think it 
will continue to prove useful even after we have templating in place.

-- 
Adam

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: cvs commit: jakarta-ant-myrmidon/antlib/src/java/org/apache/antlib/core ForEachTask.java

[Peter Donald <pe...@apache.org>]

At 04:46 PM 6/7/2002 +1000, you wrote:
>On Fri, 7 Jun 2002 14:51, Peter Donald wrote:
> > At 02:40 PM 6/7/2002 +1000, you wrote:
> > >That's probably true - but we don't know that yet.  Let's keep the task
> > > until we actually *do* have templating.  Then, we can compare them for
> > > usability. It may well turn out that they both have a place.
> >
> > I would prefer not to put another hack in unless it is furthering
> > development which I don't believe the task would.
>
>I don't understand why the <for-each> task is a hack.  Seems like a
>straightforward operation, really:  Do some stuff for each member of a list.

Essentially as its use in any real context basically requires recursive 
property resolution which has been vetoed many a time on ant-dev ;)


--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: cvs commit: jakarta-ant-myrmidon/antlib/src/java/org/apache/antlib/core ForEachTask.java

[Adam Murdoch <ad...@apache.org>]

On Fri, 7 Jun 2002 14:51, Peter Donald wrote:
> At 02:40 PM 6/7/2002 +1000, you wrote:
> >That's probably true - but we don't know that yet.  Let's keep the task
> > until we actually *do* have templating.  Then, we can compare them for
> > usability. It may well turn out that they both have a place.
>
> I would prefer not to put another hack in unless it is furthering
> development which I don't believe the task would.

I don't understand why the <for-each> task is a hack.  Seems like a 
straightforward operation, really:  Do some stuff for each member of a list.

-- 
Adam

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>