Yeah, yeah, I know. :-) Mostly, I was just curious if such a thing were possible. But I think there is a good argument to be made for this kind of task. After all, if one has to break up a functionally "atomic" target into a whole bunch of sub-targets which will never be re-used, that's a lot more work and organizational overhead. If I truly need this capability (and I suspect I will), I'll definitely take a look at the <trycatch> task in ant-contrib.
- mps --- Mark Lundquist <[EMAIL PROTECTED]> wrote: > > On Mar 17, 2005, at 10:04 AM, Michael Pelz Sherman wrote: > > > I'd like to be able to do something like this: > > > > <target name="foo"> > > <some-task/> > > <another-task/> > > > > <condition property="should.exit"> > > (condition check here....) > > </condition> > > > > <exit if="should.exit" message="some message"/> > > > > (do more-tasks...) > > > > </target> > > > There's a general pattern that may be in play here... whenever you have > a target that's starting to look like a program (or a script, if you > prefer), it may be a sign that you aren't fully grokking the "Ant Way". > Try to break it down into sub-targets with the right dependencies > expressed between them and use the 'if' or 'unless' attribute when you > need to control whether a target actually executes. > > —ml— > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
