Le samedi 24 mai 2014 05:15:31 Manfred Moser a écrit : > I totally agree with William here. I also think that any standard that is > not enforced easily is pointless. +1 I measure it every time I take time to run the Checkstyle report and fix every little inconsistency we made over time
> Thats why I implemented the checkstyle > verification to be part of the default build in the android maven plugin > (and other projects). I find this tremendously helpful. It removes any > doubts on the potentially committer side. Its gotta build.. yes, I like it, but last time we discussed it, some other ones didn't like it: they felt more tracked than helped notice that we probably set the bar too high in our Checkstyle rules: even when I take time to fix errors reported by Checkstyle, I don't try to fix everything, but only my favorite ones we probably should define a lighter configuration to check (and fail) on each build Regards, Hervé > > manfred > > William Ferguson wrote on 23.05.2014 18:10: > >> parameters and local variables are not considered final: there is nothing > >> wrong > > > > with modifying their value, and nothing good to get by forcing them to be > > > >> final > > > > But there *is* value is defining them to be final. > > > > If you as the programmer know that a value should never change, then by > > denoting it as final you can enforce that it is not inadvertently modified > > by someone else. If they deem that the restriction needs to be loosened > > because requirements have changed then they can do so. > > > > It is just like specifying an assertion. It is about code correctness and > > less failures. > > --------------------------------------------------------------------- > To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org > For additional commands, e-mail: dev-h...@maven.apache.org --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org For additional commands, e-mail: dev-h...@maven.apache.org
