Unlock parent POM?

[Cantor, Scott]

On 6/9/15, 12:15 AM, "Brent Putman" <putmanb at georgetown.edu> wrote:


>
>I think when we bump a project trunk to a new minor/major, at that point 
>it *can* use the parent TRUNK-SNAPSHOT (i.e. it becomes eligible).  But 
>IMHO it doesn't have to and shouldn't necessarily be an automatic part of 
>making a new minor.  If we don't change anything in the parent that 
>requires a new parent version, then there's no reason to change it.

I agree with that. I meant in this specific instance, not automatically.

>So personally I think it's like everything else re: versioning:  if you 
>make a change that requires a dependency version bump, that person, at 
>that time, should update things to be consistent.  If someone has changed 
>the parent and wants those changes to be
> in the new minors, then the time to change the dependencies' parent is 
>when the initial parent changes are made.

Well, I guess my question is, absent an unusual case, isn't it a 
presumption that a depdendency change there implies the unlock?

Or coming at it from the other perspective, I think if we *don't* take 
that view, we probably should also lock the testbed(s) to the tagged 
parent POM. They should be consistent.

But in this specific case, we bumped Spring, so I think we want to unlock 
the parent?

-- Scott