On Tue, 26 Mar 2002, Steve Loughran <[EMAIL PROTECTED]> wrote:
> 1. the old behaviour is a bug, pure and simple, right?
No. A missing feature, maybe. <script> doesn't expand properties,
<sql> doesn't (but there is a request to change that) ...
> So we will only have a problem if the user is using ${something} and
> <isset property="something"> holds. Which is kind of insidious; that
> could introduce intermittent failures in obscure cases.
I agree these cases are obscure (and that I cannot see any other
possibility that changing the behavior is breaking anything). I'm
sure somebody will face this obscure situation as I've always been
wrong when I thought that nobody would use a given feature. 8-)
> So, I am fairly tempted to move to a model where we always expand
> properties in the telnet task, we doc it as a 'this may break stuff'
> flag and move on.
I wouldn't veto that, but also document how to write a build file that
works on 1.4.1 and 1.5, namely:
<write>${PATH}</write>
needs to be changed to
<write string="$${PATH}" />
I think this is what you hinted at yourself, just make sure it is
documented somewhere.
> How about we move the translate properties to the <telnet> task
> itself? I cant see anyone adding it on a case by case c
????
Stefan
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
