mirror of
https://github.com/semver/semver.git
synced 2026-07-11 03:53:53 -05:00
[GH-ISSUE #80] When your Public API is in my Public API #2775
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @EddieGarmon on GitHub (Mar 15, 2013).
Original GitHub issue: https://github.com/semver/semver/issues/80
The current specification is great when you completely control all of the Public API. If instead, you also export parts of another Public API (consume or produce), you should also take into account any version changes of that consumed API when updating.
I propose the following for discussion:
Use the greater of the API changes in both yours and the exported API, as your next API change reason.
Using the exact same mechanic as currently defined:
An example:
@colin-haber commented on GitHub (Mar 17, 2013):
Honestly, I've run into this problem as SemVer prepares to update to RC2. The way I see it, as long as the API you expose doesn't change, then there's no reason to update the version for your library. If you're implementing or using external APIs, however, including a reference to their version in your library somewhere would be nice.
As RC2 will be creating backwards-incompatible changes, (mostly regarding build metadata,) I'll be bumping the major version number on my SemVer lib. I think we're saying the same thing, in which case, I don't see why anything additional needs to be added to the spec.
@jeffhandley commented on GitHub (Apr 30, 2013):
If you have a dependency whose version has made a major change, but you are only making a minor change, I think it's perfectly fine to only make a minor bump to your own version number. The consumer of the dependency would identify that the dependency had a major version change, and it may or may not affect them.
I don't think additional notes are needed in the spec for this, but we could put this into a FAQ that is an addendum to the spec itself.
@haacked commented on GitHub (May 6, 2013):
I'm going to close this issue for now. If anyone wants to take a crack at making an addendum to the FAQ, please submit a PR with the proposed change and I'd be happy to look at it.