I'm puzzled as to way maven would want to avoid it. The current behavior is: Flexible Easy to understand Easy to communicate Easy to read/follow poms Natural maven feel Works well with existing (and quantity of) maven phases
Why one would trade that for the opposite doesn't make sense to me. It would definitely supply material for the 'why is maven so complicated and limiting' arguments. -Dave On Tue, Dec 23, 2014 at 3:27 PM, Bernd Eckenfels <e...@zusammenkunft.net> wrote: > Am Tue, 23 Dec 2014 15:10:02 -0700 > schrieb David Hoffer <dhoff...@gmail.com>: > > > On the other hand if I can only have one declaration of A then I'm > > forced into gymnastics with phase binding. First I have to have > > different phases for the first 5 and the later 5 goals and then to > > get plugin B to run before the later half I have to carefully set > > this (pom location and phase) so that it runs after the first set and > > before the later. I'm not saying it can't be done...but it feels > > very unnatural and because it forces one to pick extra > > normally-unneeded phases one might run into cases where its not even > > possible. > > > > Am I missing something? > > I just want to point out, that it is actually a feature to have > data dependencies modeled as phases. So whenever something has to run > after something else it should be in later phases. > > However with the very rigid number of phases which have their own > semantic this can be rather painfull for a multi stage process like > yours. > > So I can understand both sides, yours if you want to model a imerative > dependency and mavens if it wants to avoid it. Not sure whats a good > middle ground. > > Gruss > Bernd > > --------------------------------------------------------------------- > To unsubscribe, e-mail: users-unsubscr...@maven.apache.org > For additional commands, e-mail: users-h...@maven.apache.org > >
