Skip to content

Stage 9: 커버리지 게이트 정착 — fail_under 85 발효, CI 래칫 제거 (프로그램 완료)#34

Merged
epruseal merged 1 commit into
mainfrom
test/s9-close
Jul 6, 2026
Merged

Stage 9: 커버리지 게이트 정착 — fail_under 85 발효, CI 래칫 제거 (프로그램 완료)#34
epruseal merged 1 commit into
mainfrom
test/s9-close

Conversation

@epruseal

@epruseal epruseal commented Jul 6, 2026

Copy link
Copy Markdown
Owner

프로그램 Stage 9/9 (마지막). 커버리지 62%→88% 달성으로 목표(80)를 8%p 초과, 게이트를 정착시키고 프로그램을 마감.

변경

  • pyproject fail_under 80→85 (실측 87.95의 -3%p — 후퇴 방지 강화)
  • ci.yml의 --cov-fail-under 래칫 오버라이드 제거 → pyproject 단일 진실(CLI 없이 자동 발효 실증)
  • 무의미 패딩 추가 없음: sub-85 잔여는 재export shim·미배선 스캐폴딩·옵셔널 백엔드 경로

프로그램 최종 fable 검증 PASS

  • 전체 스위트: 1719 passed / 3 skipped / 0 failed, cov 87.95% (게이트 85 통과)
  • 시뮬레이션-CI(fresh HOME·no .env) 0 failed / PG 없는 개발자 시나리오도 1517 passed·에러 0(전부 사유 있는 skip)
  • 프로덕션 엔트리포인트(apps.api.main·opencrab.mcp·TOOLS 16·factory 4종) 전부 정상
  • Stage 5 BREAKING 계약 4백엔드 일관·라이브 호출자 0 재확인, Stage 6 방언 통합 유지(DDL 중복 0, parity 50 green), 활성 xfail 0, 테스트 품질 표본 8파일 — 패딩·버그 고정 0

프로그램 원장 (Stage 0→9)

passed skipped cov 스위트
시작 722 59 62% ~130s
완료 1719 3 88% ~90s

후속 기록 (다음 웨이브 후보)

  • promotion 17%·impact 52%·tools/init 61%는 shim이 아닌 라이브 경로 포함 — 다음 테스트 웨이브 후보
  • tools.py 물리 분할(mock 전략 재설계 동반), execution/identity PG 테스트 스키마 격리, _context.py 서브모듈 리네임, cue 리스트("applicable" vs "because") product decision

… 진실)

- 프로그램 시작 62% → 최종 87.95%, 목표 80을 8%p 초과 달성
- 게이트를 실측 -3%p인 85로 상향(후퇴 방지 강화), pyproject [tool.coverage.report]가 CLI 없이 자동 발효 확인
- sub-85 잔여는 재export shim(_context/_legacy)·미배선 _envelope·옵셔널/레거시 경로뿐 — 무의미 패딩 대신 게이트 정착으로 마감
@epruseal
epruseal merged commit 4250af3 into main Jul 6, 2026
1 check passed
@epruseal
epruseal deleted the test/s9-close branch July 6, 2026 03:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant