forge-works/codecov.yml at main · adamatdevops/forge-works · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
# Codecov configuration for forge-works
# Reference: https://docs.codecov.com/docs/codecovyml-reference
#
# First-pass policy (advisory-only): post PR comments with diff-coverage
# context, but do NOT fail PRs on threshold breach yet. After ~2 weeks of
# baseline data, tighten by flipping `informational: true` → false on the
# patch + project status entries and promote the Codecov Check Run to
# Required in branch protection.
#
# Adopted per research/github_apps/evaluation_list.md (Decision Record: Codecov).
# Companion: ci.yml jobs `test-backend` + `test-frontend` upload coverage
# under flags `backend` and `frontend` respectively.

codecov:
  # Don't post the coverage comment until all CI is done — avoids racing
  # the comment with in-flight test runs that may still update coverage.
  notify:
    wait_for_ci: true
  require_ci_to_pass: true

coverage:
  precision: 2
  round: down
  range: "60...90" # red <60, yellow 60-89, green ≥90

  status:
    # Project-level: overall repo coverage trend.
    project:
      default:
        target: auto # compare against the PR's base branch
        threshold: 2% # allow up to 2% drop before flagging
        if_ci_failed: error
        only_pulls: false
        informational: true # ADVISORY-ONLY for first rollout

    # Patch-level: coverage of new lines in the PR diff.
    patch:
      default:
        target: 60% # new code should be ≥60% covered
        threshold: 5%
        if_ci_failed: error
        informational: true # ADVISORY-ONLY for first rollout

comment:
  layout: "header, diff, flags, files, footer"
  behavior: default
  require_changes: true # only post if coverage actually changed
  show_carryforward_flags: true

flag_management:
  default_rules:
    # If a flag's coverage isn't uploaded in a given run (e.g., PR only
    # touches backend → frontend test job is skipped), use the previous
    # value rather than treating it as 0%.
    carryforward: true
  individual_flags:
    - name: backend
      paths:
        - src/backend/app/
      carryforward: true
    - name: frontend
      paths:
        - src/frontend/src/
      carryforward: true

# Paths excluded from coverage measurement entirely (won't count toward
# either numerator or denominator).
ignore:
  # Test code itself
  - "**/tests/**"
  - "**/__tests__/**"
  - "**/test_*.py"
  - "**/*.test.ts"
  - "**/*.test.tsx"
  - "src/frontend/e2e/**" # Playwright e2e suite

  # Generated / vendored / cache
  - "**/node_modules/**"
  - "**/.venv/**"
  - "**/venv/**"
  - "**/dist/**"
  - "**/build/**"
  - "**/.next/**"
  - "**/coverage/**"
  - "**/__pycache__/**"

  # Static assets and migrations
  - "src/frontend/public/**"
  - "src/backend/alembic/**"

  # Out of code scope for coverage
  - "infra/**"
  - "docs/**"
  - "research/**"

github_checks:
  annotations: true # inline coverage annotations on PR diff