On 11/13/06, Blšták Peter <[EMAIL PROTECTED]> wrote:
Hi Xavier I have two questions on you. 1. I would like to ask you, if it make sense (or is possible) at this time to make contributions to Ivy 1.4.1 (I work on IVY-324 and IVY-325). Is it possible that my contribution will be added to project (if it will be acceptable in technical scope)?
Patches contributed to current Ivy version can be incorporated in future versions, we will just have to migrate them to the new packages, this is not a big work. 2. I don't understand variable substitutions made on patterns. For example
tasks publish, retrieve, ... In this tasks, there is a getProperty(....) method called for pattern. It does variables substitution. Then Ivy.publish(...) or similar methods are called. These methods does varaibles substitution again.
In some cases substitutions are done too often, some code hass not been fully updated since the introduction of the getProperty method on IvyTask. Why is it done in this way? Can't happen that there will be substitution
done twice?
Yes, substitution can happen twice, this is bad but not too bad, the second time is only a small waste of time (but I agree it would be better without) Is it necessary to do substitutions in Ant tasks? The best would be to do variable substitution in all Ivy methods, and not in the callers (e.g. Ant tasks). The code needs some cleaning/refactoring in this area (and many others), that's what I will do with the first version on apache (if everybody agree about that). Xavier Thanks
Peter
