You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We can only implement this in v2 of libcnb. There are breaking changes required for this and I believe 0.8 or 0.9 (I think around extensions). I think this highlights the point that we need to get moving on a 2.0 release, but I was also waiting for 0.10 to be set so that we could include this.
Because we can't do this in 1.x, I'd like to clear out everything with respect to stacks in he 2.x branch, but I'm not sure if that's too aggressive. I'd like to get feedback on that point. Do folks think that we should continue to support stacks into v2? Does that seem necessary for transitional purposes? Or is it sufficient to say stay on 1.x until you're off stacks?
Because we can't do this in 1.x, I'd like to clear out everything with respect to stacks in he 2.x branch, but I'm not sure if that's too aggressive. I'd like to get feedback on that point.
We're not going to be able to be this aggressive. In libcnb v2, we're supporting 0.8 - 0.10, so we still have to support stacks until we drop 0.8 and 0.9. I don't think we're ready on the Paketo side to move away from stacks either, so libpak v2 is going to need to keep supporting them, or it's going to make it really hard to upgrade and we don't want to hold folks up.
My guess is that we can eventually deprecate this in libcnb/libpak v2, then that'll position us to remove it in v3, but that's just a wild guess about what the future will hold.
Describe the Enhancement
Implement this CNB RFC to remove support for stack id buildpack uses API 0.10.
The primary change here is the environment variables that are available to the buildpack as well as the structure of the
buildpack.toml
.There will also need to be a change to the stacks to add the information needed to set
CNB_TARGET_ID
, see the platform spec for more details.Motivation
Having support for the latest API versions is important to support not only our own project use cases, but those of our users as well.
The text was updated successfully, but these errors were encountered: