I am not so sure we will gain a lot by making everything dynamic. The main reason for this, if I remember correctly was to allow things like <call target=... > with their own properties. But as it has been shown, we can do that today without any need to change the way properties work.
Classloaders I think are really fundamental and it would be nice if we can come up with some nice mechanism for loading tasks using classloaders. So that users defined tasks can be used even though they are not in the original classpath of ANT. Jose Alberto > -----Original Message----- > From: Peter Donald [mailto:[EMAIL PROTECTED] > Sent: Wednesday, August 16, 2000 7:32 AM > To: [EMAIL PROTECTED] > Subject: RE: When's the next stable release? > > > Well property expansion is already in there. It is just a matter of > changing it from parse time to interpret time. So that is > prolly not the > best place to start. > > Changine everything from parse time to execute time would > take a lot longer > but is a veryworth while task to attempt :P. However it is > also quite large :/ > > At 09:44 16/8/00 -0400, you wrote: > >Greetings... > > > >Is anyone working on the property expansion? Is that part of > something > >else someone is doing? > > > >In other words: Would it be useful for me to contribute in that area? > > > >- Paul Philion > > > >> -----Original Message----- > >> From: Peter Donald [mailto:[EMAIL PROTECTED] > > > >> * Everything gets interpreted at runtime - including > >> property expansion. > > > > > Cheers, > > Pete > > *------------------------------------------------------* > | "Nearly all men can stand adversity, but if you want | > | to test a man's character, give him power." | > | -Abraham Lincoln | > *------------------------------------------------------* >
