ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From ru...@us.ibm.com
Subject Re: Generalization of available (setif and setunless)
Date Sun, 26 Mar 2000 19:46:11 GMT


Stefan Bodewig wrote:
>Kind of embarrassing to introduce myself with a not too well informed
>proposal, sorry.

Nothing ill-informed about your proposal.  Others have proposed that
parameters be bound later, and have even offered suggestions on how to code
this.  This is within the realm of possibility.

And, quite frankly, the current state is a bit confusing.  What is slowing
us down at the moment is that we don't yet agree on where we want to take
this.  A few things that we agree on:

1) If you take a look at XSLT, you will see what happens when you try to
make a full programming language out of XML.  Whether or not that is the
right thing for XSLT, there is consensus that we do _NOT_ want to do that
with Ant.

2) It should be easy to integrate in logic from another programming
language.  Given the history of this project, understandably most of the
focus to date has been on Java.  Recently, I've begun integrating in
scripting languages (take your pick: JavaScript, Perl, Python,  VBScript,
TCL, NetRexx...).  And, of course, you can always execute standalone
programs.

So, net, we don't want to make a programming language out of Ant.  But we
do want to make it easy to include logic in the language of your choice.
Is there some way that you can restructure your proposal so that it takes
advantage of other languages instead of duplicating them?

- Sam Ruby



Mime
View raw message