daily-fix: fix red main: test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py] — make the four issue1689_user_slot entryp
Highlight text on any card (body, plan, or an event) to anchor a comment, or leave a whole-task comment here. Mention @claude to summon a reply.
No comments yet.
Overview / Motivation
Auto-filed by the watcher urgent-park router (#1681,
autonomous_session_watch.urgent_wf_park_pass) from an URGENT
(urgency: main-red) parked workflow-fix candidate raised on task
#1840. The named test is red on origin/main NOW — every
intervening session's Step 9c gate re-classifies it until this fix lands.
Goal
fix tests/test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py] red on origin/main: make the four issue1689_user_slot entrypoints satisfy the os_exit contract (call os._exit with flush-before and rc type coercion per the test's pins), or update the test's param list if the entrypoints were deliberately re-shaped — NOTE a live #1689 session (uncommitted shared-root edits on exactly these scripts) may be about to land this; the router's red-verification + the spawned session's clarifier should check #1689's branch state first
Workflow gap
- Bug observed (emitter's claim, candidate block): tests/test_issue1689_os_exit_shutdown_bypass.py is red on pristine origin/main — 12 failing params (test_entrypoint_calls_os_exit / test_entrypoint_flushes_before_os_exit / test_entrypoint_os_exit_rc_type_coerced x issue1689_user_slot_{capture,fits,render,gen_a1}.py): the committed issue1689 user-slot entrypoints do not satisfy the os_exit shutdown-bypass contract the test pins
- Failing node (router-verified):
tests/test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py] - Confidence (emitter): medium
- verified-at-filing:
uv run pytest tests/test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py] -q-> rc=1 at main @ 05ffc4f18f (2026-07-31T00:13:09Z) (FAILED — red confirmed)
Proposed change (candidate diff sketch — refine in planning)
(see the verbatim candidate block under ## Provenance — the router
forwards it unmodified and never synthesizes fields)
Scope / surfaces
- Primary target:
scripts/issue1689_user_slot_capture.py, scripts/issue1689_user_slot_fits.py, scripts/issue1689_user_slot_render.py, scripts/issue1689_user_slot_gen_a1.py - Failing node:
tests/test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py]
Constraints / invariants
- NON-workflow-surface fix (
wf_fix: falsedeclared by the parking session — the /daily route-2 analogue); scope stays on the named target, nevertasks/state.
Provenance
- fingerprint: e5f17affb1b8
- routed-by: autonomous_session_watch urgent-wf-park-router (#1681)
target_file: scripts/issue1689_user_slot_capture.py, scripts/issue1689_user_slot_fits.py, scripts/issue1689_user_slot_render.py, scripts/issue1689_user_slot_gen_a1.py bug_observed: tests/test_issue1689_os_exit_shutdown_bypass.py is red on pristine origin/main — 12 failing params (test_entrypoint_calls_os_exit / test_entrypoint_flushes_before_os_exit / test_entrypoint_os_exit_rc_type_coerced x issue1689_user_slot_{capture,fits,render,gen_a1}.py): the committed issue1689 user-slot entrypoints do not satisfy the os_exit shutdown-bypass contract the test pins why_workflow_gap: a live red on origin/main taxes every intervening session's Step 9c gate with a re-classification round (the #1643/#1713 fleet-cost class); the fix itself is experiment-code (out of wf-fix scope) but the red needs mechanical routing rather than per-session re-discovery proposed_change: make the four issue1689_user_slot entrypoints satisfy the os_exit contract (call os._exit with flush-before and rc type coercion per the test's pins), or update the test's param list if the entrypoints were deliberately re-shaped — NOTE a live #1689 session (uncommitted shared-root edits on exactly these scripts) may be about to land this; the router's red-verification + the spawned session's clarifier should check #1689's branch state first diff_sketch: |
- per failing param: add the flush + os._exit(int(rc)) tail the test pins
- (or) tests/test_issue1689_os_exit_shutdown_bypass.py: drop/adjust params if the
- entrypoint shape legitimately changed on the #1689 branch confidence: medium related_task: #1840 urgency: main-red failing_test: tests/test_issue1689_os_exit_shutdown_bypass.py::test_entrypoint_calls_os_exit[issue1689_user_slot_capture.py] wf_fix: false