ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Peter Donald <dona...@apache.org>
Subject Re: [PATCH] ant task with (nested) fileset(s)
Date Tue, 13 Nov 2001 08:52:57 GMT
On Tue, 13 Nov 2001 15:41, Jose Alberto Fernandez wrote:
> From: "Peter Donald" <donaldp@apache.org>
>
> > > I would like to see:
> > >
> > > (1) How to write <foreach> functionality over a fileset (or
> > > equivalent). (2) How to write <javaon> functionality.
> >
> > loaded questions. Assumes we need this functionality. While XSLT could
> > provide this I don't think they are needed given (3)
> >
> > > (3) How to write new pseudo-tasks that are defined as the sequential
> > >      execution of existing tasks with ways to pass arguments.
> >
> > See message with Subject "Re: Question about properties and ant2" on 14th
> > September.
>
> I took a look at that thread, and as I suspected (since I would have take
> notice otherwise) your example leaves out how to manage sets of files
> needed during expansion. How do I say the equivalent of "<include
> name='src/**/*.c' />" using XSTL? That is, how I create the lists of all
> files in the filesystem (relative to the basedir of the project) which
> match the include criteria above.

look above. reread. Ant is not a scripting language. Im not sure how many 
times that has to be said. Yes XSLT could do that easily enough (via 
extensions if you are wondering) but I don't think it should be. When I said 
"I don't think they are needed given (3)" I meant "I don't think they are 
needed given (3)". Any clearer?

> Without that things like <apply> or <execon> cannot be express via
> templates which is the point of the discussion. 

who said that ? I don't think the functionality you are desiring from 
apply/execon is desired behaviour. The functionality is better provided via 
alternate mechanisms.

> Even in the presence of (3)
> the point is how to do the expansion.

How is easy. The question is "if" or "should" not "how". And my answer is 
"no".

> > Right. But is "talk-talk" going to get us any closer to a solution to
> > this problem? Of course not. Only way to get closer to solution is
> > "walk-walk", people actually start working together and implementing
> > things. I have no problem doing the work - hell in worst case I would do
> > it all if I had too - problem is "design by committee" on a tool that is
> > "good enough" at the moment. No one agree or even trys to work together -
> > can you guess the end result ? I can. Egos are great things - no?
>
> I do not believe that using a separate template language is the solution to
> all problems. Since you are the one pushing for it and disallowing anything
> else that is different from your vision, then I think you have the burden
> of providing a solution that we all can assess and accept or refuse once we
> see what the implications are.

walk-walk not talk-talk!!!!!!!

I don't have to convince you of anything - you aren't a committer. I have to 
convince other committers or they me - considering the lack of agreement on 
simple things - do you honestly believe this is viable ?

-- 
Cheers,

Pete

-------------------------------------------------------
To fight and conquer in all your battles is not supreme 
excellence; supreme excellence consists in breaking the 
enemy's resistance without fighting. - Sun Tzu, 300 B.C.
-------------------------------------------------------

--
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