ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jose Alberto Fernandez <JFernan...@viquity.com>
Subject RE: [PATCH] New <case> task
Date Wed, 11 Oct 2000 18:43:52 GMT
> From: Stefan Bodewig [mailto:bodewig@bost.de]
> 
> Jose Alberto, I think I understand your needs, but really, this is
> something that would be stretching my picture of Ant way to far.
> 
> What you propose might be purely declarative but it points into a
> direction I really don't want to see Ant traveling. I share Pete's
> view that soon thereafter somebody would come and say
> <if>/<elseif>/<else> is yet another way to write <case>. After
that
> comes "why do we only set properties, we could call a target like
> antcall does as well" ....
> 
> No, please lets not get onto that road.
> 

Well, Stefan, I actually wrote <case> following your comments about my
modifications to the "if" attribute. You proposed that this kind of thing
should be better in a task. Fair enough, I made a task.

Now a task, in core, is being frown at. Well, how about making it an
optional task. Unless there is something viceral about it, I do not think
there could be any reason to be against that. Since everything else has been
allowed there.

Would you please just move the task there, or do you want me to re-submit
it?

> I think what you - together with others who ask for procedural
> features, template mechanisms and the like - want to achieve, could be
> the driving force of a front end to Ant that encapsulates all this,
> creates a (possible ugly) build file and feeds it to Ant. Preferably
> this would work without actually writing that file - generating SAX
> events for example.
> 
> Would you be interested in helping to develop such a front end? This
> might as well help us moving some parts out of Ant's core that don't
> really seem to belong there (though I doubt Pete and I agree on which
> parts that might be).
> 

Well, I am not sure how much time commitment I can make on this. I would
really like something on this regard, however, my views are a little
different. I would like ANTs architecture to allow for plugin preprocessors
to be installed at different stages of the ANT pipeline (SAX preprocessors,
I mean). Otherwise, things like <ant> or <antcall> will not work as
expected.

Would ANTs architects be willing to allow for this kind of thing?

Jose Alberto

Mime
View raw message