Bug Fixes
-
reject a lone '@' as a reference name, like Git does
@is shorthand forHEADin revision syntax, so Git refuses a reference
name consisting solely of it - seecheck_or_sanitize_refname()inrefs.c,
whichcheck_refname_format()delegates to. It substitutes-when
sanitizing, which is done here too.The visible effect was in refspec destinations:
git push origin "HEAD:@"
fails withfatal: invalid refspec, whilegix_refspec::parse("HEAD:@", Push)
returnedOk(dst = "@"). Sources were already correct, asparse()rewrites a
bare@source toHEAD, but destinations reachreference::name_partial().@remains valid as a component, sorefs/heads/@and a tag named@are
unaffected.Pins that
@is rejected byname()andname_partial()while staying valid
inside a path (refs/heads/@,@/x,@@), and that inputs collapsing to a
lone@during sanitization are replaced too. The tag expectation changes
becausemktests!sanitizes as a reference name - a tag named@is still
valid, asrefs/tags/@is a legal ref.
Commit Statistics
- 7 commits contributed to the release over the course of 30 calendar days.
- 30 days passed between releases.
- 1 commit was understood as conventional.
- 0 issues like '(#ID)' were seen in commit messages
Commit Details
view details
- Uncategorized
- Update manifests prior to release (ebe9095)
- Merge pull request #2905 from GitoxideLabs/various-improvements (f3bbfad)
- Adapt to changes in
gix-testtools(0cbe539) - Merge pull request #2886 from ameyypawar/validate-at-refname (12240ab)
- Review (1384133)
- Reject a lone '@' as a reference name, like Git does (679f624)
- Merge pull request #2812 from GitoxideLabs/report-july (ae8845a)