Repository navigation
π ci: repair v5 baseline (openapi, hermetic promotion test, collapsed-summary assert, style ratchet) - #10612
Conversation
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
Changelog: this PR changes code but carries no changelog entry If it is user-visible β a feature, a fix an operator would notice, a This is a reminder, not a gate; it never blocks a merge. |
Generated by changelog-fragment-guard from the PR title (#5675). Edit the wording freely, or replace this file with the no-changelog label if the change is not user-visible. Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Signed-off-by: clubanderson <407614+clubanderson@users.noreply.github.com>
Open PR overlap checkBase branch: Conflicts are checked with every PR merged onto the current tip of
|
There was a problem hiding this comment.
Conclusion: the three fixes look correct to me and this is the PR that should land to restore the v5 baseline. One body claim is inaccurate (low impact), and #10617 conflicts on the same lines.
Read: src/pkg/dashboard/feedback.go:80, src/pkg/hub/feedback.go:378, src/pkg/hub/heartbeat_upgrade_policy_test.go:210-219, and the dashboard/openapi.json entry. The ghcrTagGeneration / stablePromotionFetchRuns hooks are the existing package vars, so stubbing them makes TestStableNextPromotionAtUnknownWhenNothingQueued hermetic as described.
Body vs. code (low): the body says the new regexp was "verified same acceptance set". It is not quite: the original's {0,38} bounded single characters (39-char cap); the replacement (?:-?[A-Za-z0-9]){0,38} bounds 2-char iterations, so e.g. a-a-a-β¦-a at 77 characters now matches (checked with Go's regexp; a 40-char aaaaβ¦ is still rejected). Consequence is just an over-long @mention in a feedback issue β the character class stays tight, so nothing injectable β but the claim should be corrected or the cap enforced outside the regexp as #10617 does (len(strings.TrimSuffix(login, "[bot]")) > 39).
Conflict, not duplicate: #10617 rewrites the same two declarations with a different pattern plus an explicit length check and tests. It also carries ~830 unrelated lines for #10615, so I've recommended there that this PR merge first and #10617 rebase down to the length cap + tests. If a maintainer agrees, nothing further is needed here.
relative links and anchors resolve fails identically on v5; #10608 addresses it, as the body notes.
β hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
hive/awaiting-ci: pushed π Hive Agent: β hive: agent=scanner backend=copilot model=claude-fable-5.1 copilot=1.0.88 |
Generated by changelog-fragment-guard from the PR title (#5675). Edit the wording freely, or replace this file with the no-changelog label if the change is not user-visible. Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Signed-off-by: clubanderson <407614+clubanderson@users.noreply.github.com>
β¦motion test Signed-off-by: Andy Anderson <andy@clubanderson.com>
Signed-off-by: Andy Anderson <andy@clubanderson.com>
β¦chet to 666 Signed-off-by: Andy Anderson <andy@clubanderson.com>
1ae7ce3 to
2237fe9
Compare
There was a problem hiding this comment.
Conclusion: the four repairs do what the body says and look correct to me. Two small things to fix before merge, both verified:
-
Stale changelog fragment.
changelog.d/fixed-10612.mdreads "feedback: drop unsupported regexp lookahead that panicked at init (#10612)" β but this PR contains no regexp change (the body itself says that landed in #10617). This is the guard's title-generated autofix from before the rebase.changelog.d/is the directory the fragment guard reads (changelog-fragment-guard.yml:51,156), so this is the text that will ship in the changelog. Suggest rewriting it to describe what's actually here (openapi entry, hermetic promotion test, collapsed-summary assert, style tokenization) β or move the content ofsrc/changelog.d/fixed-v5-baseline-openapi-and-promotion-test.mdinto it, since that one already says the right thing.- Open question for a maintainer: is
src/changelog.d/still read by anything? It has 91 entries onv5while the guard only watcheschangelog.d/**; if it's legacy, the new fragment there is dead.
- Open question for a maintainer: is
-
gofmt β
src/pkg/dashboard/webstatic/style_ratchet_test.go:31changedrawFontSizes: 667,βrawFontSizes: 666,and lost the column alignment;gofmt -lflags the file. Not CI-blocking (no gofmt gate inv2-ci.yml/v2-tests.ymlthat I could find), just a one-space fix.
What I checked:
heartbeat_upgrade_policy_test.go:213-219βghcrTagGeneration(channel_targeting.go:204) andstablePromotionFetchRuns(stable_promotion.go:145) are both package-levelvar β¦ = funconv5, so the stub/restore pattern compiles and is hermetic. Correct approach; note #10589 removes the lookup fromstableNextPromotionAtinstead, so the two conflict and a maintainer should pick one.agent_collapsed_tiles_test.go:35β the expected string matches what #10616 put invisualSectionSummary.index.html:1413βpadding:1px 16px 1px 6pxβvar(--sp-1) var(--sp-6) var(--sp-1) var(--sp-3)removes the raw-padding hit;rawPaddingbaseline stays 229.dashboard/openapi.jsonβ/api/feedback/statusGET entry added; field names match#10593's status payload as described in the body.
I have not changed anything on this PR.
β hive: agent=reviewer backend=copilot model=claude-fable-5.1 copilot=1.0.88
|
Thank you for your contribution! Your PR has been merged. We'd love to hear how your experience was: share feedback |
Repairs the remaining v5 baseline failures after this morning's red merges (the regexp lookahead panic from #10593 was fixed separately in #10617; this PR is rebased on top of it).
GET /api/feedback/statuswithout documenting it βTestOpenAPISpecCoversEveryRegisteredRoute. Added the entry todashboard/openapi.json.TestStableNextPromotionAtUnknownWhenNothingQueuedhit live GHCR/GitHub β stubbedghcrTagGenerationandstablePromotionFetchRunsso the test is hermetic.repos-sectionβ updatedTestAgentsCollapsedSummaryRendersPerAgentTiles's expected string.padding:1px 16px 1px 6pxβTestStyleRatchet(230 > 229). Tokenized to--sp-*; lowered the font-size baseline 667 β 666 as the ratchet demands.Refs #10639 (v5 branch protection only requires
gate, so these landed red).