Actually, what I would like to suggest is that we reify the idea of a "task context". I am still writing some notes on it, but the basic idea is that you can define various aspects about how to call a task or what to do if a task fails. You could then install a context at the build file level, target level or task level. You could reuse them as well using id/refid.
I also like it because it better defines the contract between task and environment for the task developer. It also seems as semi-natural place for me to introduce my idea of a auto-Set (from a week or so back) where the context can automatically call a task multiple times for a set of parameters. It also allows a further level of customization for the aggressive developer/user without changing the "core". Also, it seems like a good place going parallel or distribution also. d On Sat, 17 Mar 2001 21:41:34 -0600, Ken Wood wrote: >Actually, I'd like to take this one step farther. > >I'd like to see more consistency across ALL tasks. >So, if 'ant' task and 'javac' task have failonerror, let's >provide it in ALL tasks. > >Similarly, let's get some consitency in names. Some tasks >use "dir" and "todir", others use "src" and "dest". >Some tasks use explict filesets, some use implicit >file sets, and some tasks don't use filesets at all. >(javadoc, for example). > >It would be nice to know that if you learn to do things >a certain way in one task, that some thing is done the >same way in all other tasks, instead of a different >way each time. > >I suspect these issues relate to my comment the other >day about too much cut and paste duplication between tasks. >When a 'new and improved' way comes along, it's hard to go back and >cut and paste the NEW way into every tasks that had a copy >of the OLD way.... > >Scott Sanders wrote: > >> I do not know if someone has already noted this, but it would be nice >> if the ant task had a failonerror attribute. >> >> I am using an ant build file to call the build files of all jakarta >> projects. When one fails, the entire process fails currently. The >> workaround that I am thinking of is to use the <java/> task to call >> ant. Any other suggestions? >> >> Maybe possibly every task could have a failonerror attribute. >> >> Thanks >> Scott Sanders >> >> >
