mirror of
https://github.com/semver/semver.git
synced 2026-07-11 06:53:03 -05:00
Should rule #9 be upgraded to a SHOULD #227
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/semver#227
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 @BurtHarris on GitHub (Jun 13, 2016).
Currently, rule 9 indicates that use of pre-release suffixes is optional, by using the word MAY.
In real-world implementations, like npm, consumers often depend on publishers following rule 9. It seems to me like the rule should be updated to use the word SHOULD in place of the first occurance of "MAY".
@haacked commented on GitHub (Jun 13, 2016):
Hmm, I can see how that is a bit confusing. The way I read it, it's "MAY" because not every version is a pre-release version. So a pre-release version MAY be denoted.
@BurtHarris commented on GitHub (Jun 17, 2016):
OK, good point. So perhaps rule nine should begin with "Any pre-release version SHOULD be denoted by appending a hyphen..."
@FichteFoll commented on GitHub (Jun 17, 2016):
The way you phrased it requires "MUST" to be used.
@crazedsanity commented on GitHub (Jun 17, 2016):
I agree that pre-releases must be distinguished from regular/non-pre-release versions. Having "MAY" there seems indicative of the possibility that pre-releases might not be distinguishable as such. So rule 9 would then change to:
Just wanted to (hopefully) clarify the wording to ensure we're all on the same page.
@BurtHarris commented on GitHub (Jun 17, 2016):
I'd be happy enough with either version.
@tdpsk commented on GitHub (Jun 19, 2016):
I agree that in practice, pre-release tags should be mandatory because users rely on them in the context of package repositories. Therefore I have created a corresponding pull request in #316, feel free to comment and discuss.
@FichteFoll commented on GitHub (Jun 19, 2016):
This could be misunderstood, but the wording in the PR is clear.