Repository navigation
test: raise the security package line budget for the git publish check - #31625
Conversation
#30774 added 20 lines to argv_floor_git without raising the package budget, so main is 34,753 lines against 34,733 and test_the_package_stays_within_its_line_budget fails on every PR. Raise the budget by those 20. No code change.
Design Review (Opus 5.5) — ✅ PASSDesign-level review of Design-Verdict: PASS This is a one-constant test-budget raise that matches the 20 lines #30774 added. It changes no runtime code, takes nothing away, and adds no new user-facing surface. Notes
[DESIGN-REVIEWED] 589b119 Description read[DESCRIPTION-READ] 40a0ec6a4714668ec278699dccc472536b05e29c3a49848aa01b3e4d161fc525 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 |
UX Review (Opus 5.5) — ⏭️ skippedRevision |
First Principles Review (Opus 5.5) — ⏭️ skippedRevision |
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: |
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: |
|
VERDICT: APPROVE (MKT-APP head, at 589b119) Reasons, read at this head:
This unblocks every PR whose Backend shard 2 and Coverage Gate fail on this test, #31601 included. It still needs a human maintainer's approval. |
The backend shards failed on test_the_package_stays_within_its_line_budget, a security-package budget that main raised in #31625 after this branch's merge ref was built. A re-run reuses the old merge ref, so this empty commit gives CI a fresh one. No content change.
Two main-side reds reached this branch's merge refs: the security-package line budget (fixed on main by #31625) and the pull_request_target allowlist missing sensitive-change-review.yml (fixed by #31803). Neither touches files in this PR. Re-runs reuse the old merge ref, so this empty commit gives CI a fresh one. No content change.
Problem / Motivation
Goal: Main's backend test shards pass again: the security package line budget covers the git publish check that merged in #30774.
test/test_security_gate_liveness.py::test_the_package_stays_within_its_line_budgetfails on main since #30774 (package grew to 34753 lines, budget 34,733). #30774 added a net 20 lines tosecurity/argv_floor_git.pywithout raising the budget, so every PR's Backend Tests shard is red from it.Why it matters
Every open PR shows a red backend shard it did not cause, and readiness stays red for all of them.
Not a goal
What changed (motivation → approach → change)
_PACKAGE_LINE_BUDGETgoes from 34,733 to 34,753, with a note in the test's existing style naming what #30774 added. Test-only.Backwards compatibility
Compatible: test-only; no runtime code changes.
Removes nothing: the budget only grows by the 20 lines #30774 added.
Tests
test/test_security_gate_liveness.py: 10 passed on fresh origin/main8095bc3e70plus this change (one process, no xdist); the budget test fails there without it.Proof: not provable -- test-only change
Manual verification
N/A — unit coverage sufficient: the change is one constant in a test.
Related Issues
Follow-up to #30774 and #31469.
Sibling sweep
Swept: git grep -n "_LINE_BUDGET = " -- test/test_security_gate_liveness.py
Siblings: none -- the package budget is the only total-size budget in that test
Pattern harvest
Not generalizable: a security change merged without its budget raise; the same budget raced once before (#31469).
Checklist
feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)