On Tue, Oct 25, 2011 at 7:12 PM, Adam Murdoch
<[email protected]>wrote:
>
> On 26/10/2011, at 7:05 AM, Hans Dockter wrote:
>
> Wouldn't it be cool if the output of an aggregating task, e.g.
> assemble.output.files, would aggregate the outputs of the tasks it
> aggregates?
>
>
> Maybe. What would you use it for?
>
If I want to create an uberzip for all published archives.
task uberZip(type: Zip) {
from assemble
}
Right now I have to do something like:
task uberZip(type: Zip) {
from { tasks.withType(Jar) } // which is not the same as assemble. To do
the real assemble I have to work with the configurations build dependencies.
}
In any case, the above would be much nicer. You could argue that we need
something in the domain model to make this easier for this case. Some output
property for the published archivs I can simply depend on. But in general I
think it would be very helpful and correct behavior that aggregate tasks
behave like proposed. It is another piece of improving our autowiring story
and making the inputs and outputs more relevant.
Hans
>
>
> --
> Adam Murdoch
> Gradle Co-founder
> http://www.gradle.org
> VP of Engineering, Gradleware Inc. - Gradle Training, Support, Consulting
> http://www.gradleware.com
>
>
