> From: James Duncan Davidson [mailto:[EMAIL PROTECTED] > > > On 12/6/00 11:18 PM, "Peter Donald" <[EMAIL PROTECTED]> wrote: > > > Why ? You would prefer for ${fred} to potentially mean 7-8 > different things > > depending on where it is used. Currently we do that with different > > namespace for filesets/properties/other datatypes. I prefer > simplicty over > > complexity that you seem to be advocating ? > > Same thing exists with System.properties. "fred" means very > different things > when it is the result of "user.dir" or "home.dir" or > "user.name". Context > matters. I really don't see this as complexity that I'm > advocating. Adding a > bunch of typing on top of this is complex afaic. > > Filesets != properties. A property is a simple name=value > pair. Filesets is > a one-to-set mapping of files which may contain regexp patterns. >
I think we should point out that one of the needs for some sort of namespace consolidation is that, today, <ant> and <antcall> cannot pass data-types around. Only string properties. This means that paths, filesets, etc need to be redefined on every subproject. By consolidating the namespaces, we will have a common mechanism to pass information arround all with the same rules. BTW, I prefer allowing typed properties as considering properties another data-type. The reason being that properties carried the concepts of UserDefined which we use when a project calls a subproject which means that this will be carried naturally to passing other datatypes. And then there is the issue whether IDs should be always valis within the local XML document just like real ID attribute types in DTDs. Jose Alberto
