https://bz.apache.org/bugzilla/show_bug.cgi?id=62534
--- Comment #4 from Jaikiran Pai <jaiki...@apache.org> --- >> Maybe we should provide support for nested <arg>s so peopke can use whatever >> flag they need and don't have to wait for us to provide the implementation >> and a new release? For a task like this one, where we wrap a command/tool which has too many permutations/combinations, plus no real specification to follow, I think it's better to support a generic <arg>s element for it. As for validation of these specified args, in context of this and similar (keytool based) tasks, I am not too sure we doing that validation, by exposing specific attribute names to the task, adds much value. What I mean is, ultimately we end up doing almost the same set of validations that keytool would have done. Plus, recent bug/enhancement reports show that keytool has too many permutations of arguments that behave differently in context of what the command is, adding to the complexity of such validations. -- You are receiving this mail because: You are the assignee for the bug.
