Brett Porter wrote:
Sorry for the delayed reply.
Better late than never :-)
I would say not 2.0.x, but I think it makes sense for 2.1.x.
Sounds good.
I believe we could drop the "stripped" syntax already and handle it with errors when used.
Not sure whether we should really produce an error/exception. I would just have the expression pass through like any other undefined property, e.g. null or the literal expression, whatever Maven currently does.
This one seems a bit trickier - though it is clearly inconsistent I could see how someone might have come to depend on it. Perhaps we can detect when both are set, different, and it is used, and introduce a warning that they may be using unsupported ordering in a future release?
The warning is a nice idea to ease migration but about what Maven versions are we talking here exactly? E.g. fix 2.1.x to be consistent and output the warning in 2.0.x or leave the entire 2.x line as is and output the warning?
Did these already have corresponding JIRAs for tracking?
The latter one is MNG-1992, I haven't come across a JIRA for the first issue with the non-prefixed lookup so far (it's quite an edge case and not easily noticed unless a plugin looks at the effective model, say by inspecting other plugin configs).
Benjamin --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
