Repository navigation
Conversation
UX Review (Opus 5.5) — ⏭️ skippedRevision |
First Principles Review (Opus 5.5) — ✅ PASSPremise-level review of First-Principles-Verdict: PASS Nothing to check. What this change shipsInventory (1 items) — 1 justifiedIntent: keep main's updates-facade manifest test green after #35088 added a new served name. This is a FIX.
[FIRST-PRINCIPLES-REVIEWED] 58f6603 Description read[DESCRIPTION-READ] 3fcc94ce3418f4a63f25681e711c50afcb8a52e889e7d7696a45deab7f1469e8 That sha256 names the PR description this verdict read. Recompute it with |
Design Review (Opus 5.5) — ✅ PASSDesign-level review of Design-Verdict: PASS Nothing blocks: this is one TYPE_CHECKING-only import that adds the missing name to the updates facade's manifest of names served from outside the family, which is what the failing test is about. It changes nothing at runtime and removes nothing. [DESIGN-REVIEWED] 58f6603 Description read[DESCRIPTION-READ] 68d9a20936ab5537554f821b620267dbcecd94ad8dfd465c95f64033520205e8 That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with |
Opus 5.5 Review — ✅ no blocking findingsReviewed Verdict parsed from the review's SHA-scoped output markers for commit False positive or not applicable? A repository writer can comment: |
GPT 6.1 Review — ✅ no blocking findingsGPT 6.1 completed its review of This comment is updated in place on each push. Review detailsNo findings. False positive or not applicable? A repository writer can comment: |
7510b35 to
644e589
Compare
644e589 to
a00c946
Compare
a00c946 to
58f6603
Compare
buluoray
left a comment
There was a problem hiding this comment.
🤖 Kiro Crew Auto-Pipeline [operator: buluoray]:
Approve at 58f6603. Together with #35625 this turns main's backend tests green again.
Main is red since #35088: test_updates_handlers_facade.py::test_every_name_served_from_outside_the_family_is_visible_to_type_checkers fails on Backend Tests (3.12, 6) and (Windows) (4). On current main the failure reads Right contains 1 more item: {'_running_version': ('kiro_crew', '__version__')}, so the facade serves _running_version but does not list it where type checkers can see it.
Checked: I merged this PR and #35625 onto current main (e8283ce) and ran test_updates_handlers_facade.py, test_update_boot_apply.py and test_wait_discipline_ratchet.py: 81 passed. With this PR alone the facade test passes and the two wait-ratchet tests still fail, as the description says; #35625 covers those.
The change is one import line at the end of the block in dashboard/handlers/updates.py that lists, for type checkers, the names the facade serves from other modules. No behaviour change.
|
Superseded by #35625, which merged the same updates facade line and the wait fix into main. Closing; branch kept. |
Pull request was closed
Problem / Motivation
Goal: Keep main's updates-facade test green after the boot-apply change landed.
Since 5638488 (#35088),
src/kiro_crew/dashboard/handlers/updates_step_up.pyimports__version__ as _running_version, but the facade's TYPE_CHECKING list of names served from outside the family does not name it.test/test_updates_handlers_facade.py::test_every_name_served_from_outside_the_family_is_visible_to_type_checkersfails:Right contains 1 more item: {'_running_version': ('kiro_crew', '__version__')}. Seen on the v0.12.2-rc.1 release run (43102437814, Test Release Candidate SHA (4)); main has the same files.Why it matters
The next Nightly and every full test run of main go red on this test. PR CI runs it only when a PR touches the update modules, so it stays hidden until then.
Not a goal
test/test_update_boot_apply.py; test: give the boot apply sigterm test the 30 s wait ceiling #35625 handles it.What changed (motivation → approach → change)
Facade test red → the facade's manifest is missing one served name → add
from kiro_crew import __version__ as _running_version # noqa: F401 # isort: skipat the end of the "names it serves from outside the family" block insrc/kiro_crew/dashboard/handlers/updates.py. Theisort: skipis needed because isort folds two aliases of one name into a single line and drops the secondnoqa, which makes flake8 report F401; release/0.12.0's copy of this line (#35623) has that isort red.Backwards compatibility
Compatible: one TYPE_CHECKING-only import that never runs.
Removes nothing: the facade only gains a type-checking name.
Tests
test/test_updates_handlers_facade.py: the facade manifest now matches the names it serves.Proof: not provable -- other: the change is a TYPE_CHECKING import read only by an AST scan test
Local checks:
isort --check-only,flake8andblack --checkpass on the file.Manual verification
N/A — static manifest change; CI runs the facade test.
Related Issues
Wait-ceiling fix for the same merge: #35625.
Sibling sweep
Swept: git grep -n "import version as _" -- src/kiro_crew/dashboard/handlers/
Siblings: none -- updates_step_up.py is the only updates owner that binds a new alias of version
Pattern harvest
Not generalizable: the facade test already catches this class; it ran only after merge because PR CI skipped it.
Checklist
feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)Contribution License Agreement
Contributions are accepted under the project's Apache-2.0 license; see the Licensing section of CONTRIBUTING.md.