>>>>> "SR" == Stuart Roebuck <sr@adolos.com> writes:
SR> On Thursday, November 2, 2000, at 04:42 PM, Stefan Bodewig wrote:
>> If you can give a good heuristic how to recognize that there is a
>> chmod command on a platform that is either a Mac (not MacOS X) or
>> uses a ; as PATH separator, we can make the check smarter.
SR> Would there be any interest in sticking a standardised platform
SR> check heuristic into the abstract Task?
Of course, I always wanted a way to say os="Unix" in <exec> instead of
listing all the different little beasts I expect the build to be
performed on.
>> Ant 2.0 will have a very different notion of what is core and
>> what is optional,
What I should have said here is, that we are still at the state of
deciding how Ant 2.0 is supposed to work and putting together a spec
for it. Discussion about this happens at ant-dev right now. I wouldn't
hold my breath waiting for 2.0, expect another release of the Ant 1
series before that.
SR> Although I agree with the idea of making as much plug-in as
SR> possible, I also think that Ant needs to have a 'Standard
SR> Edition'
Well, we will have a repository of tasks where everybody can come and
pick his collection of tasks from. I can even envision something like
XEmacs' sumo tarball which in effect includes all elisp packages
hosted at the XEmacs site - in our case it would be a one stop place
to get all the contributed tasks.
SR> On a slightly connected but otherwise unconnected point...Is
SR> there anyplace to put my rejected :-( telnet task.
Don't think it has been rejected as such. It takes a little time for
one of the committers to look at a contribution, evaluate, comment on
and finally commit it. Ask Glenn McAllister, he had a patch I
committed three months after he had sent it in, just fast enough to do
it before he became a committer himself 8-).
Stefan
|