CodeWiki-Plus/.github/workflows/ci.yml at develop · mambo-wang/CodeWiki-Plus · 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
98
99
100
101
name: CI

on:
  pull_request:
    branches: [develop]
  push:
    branches: [develop]

permissions:
  contents: read

jobs:
  test:
    name: Test
    runs-on: ubuntu-latest
    timeout-minutes: 20
    steps:
      - uses: actions/checkout@v5
        with:
          fetch-depth: 0

      - uses: astral-sh/setup-uv@v7
        with:
          enable-cache: true

      - name: Set git config for tests
        run: git config --global user.email "ci@codewiki.test" && git config --global user.name "ci"

      - name: Install dependencies
        run: uv sync --frozen

      - name: Run tests
        run: uv run pytest --cov=codewiki --cov-report=term-missing -p no:cacheprovider tests/ -q

      # The MCP handler contract lives in this standalone script (126 checks over
      # fake-repo fixtures); pytest never collects it because its entrypoint is
      # main(), not test_* functions.  Run it so handler regressions are visible,
      # but non-blocking: 6 of its assertions have drifted behind deliberate
      # product changes (ADR-0013 stable-requires-reason, batch_ingest results →
      # report_file, doc_writer drops `components` for component_count,
      # lint check count 16→22, raw/sources OKF conformance errors, get_prompt
      # output_dir-only addressing).  Triage them, then flip to blocking.
      # tests/smoke_test_mcp.py stays out of CI entirely: it passes this repo's
      # real path to analyze_repo, whose output_dir is now derived from repo_path
      # (analysis.py:57-64), so running it writes into the live repowiki/.
      - name: Run MCP handler regression suite
        continue-on-error: true
        run: uv run python tests/okf_regression_test.py

  lint:
    name: Lint
    runs-on: ubuntu-latest
    timeout-minutes: 10
    steps:
      - uses: actions/checkout@v5
        with:
          fetch-depth: 0

      - uses: astral-sh/setup-uv@v7
        with:
          enable-cache: true

      - name: Install dependencies
        run: uv sync --frozen

      - name: Collect changed Python files
        env:
          PR_BASE_SHA: ${{ github.event.pull_request.base.sha }}
          PR_HEAD_SHA: ${{ github.event.pull_request.head.sha }}
          PUSH_BASE_SHA: ${{ github.event.before }}
        run: |
          if [[ "$GITHUB_EVENT_NAME" == "pull_request" ]]; then
            diff_range="${PR_BASE_SHA}...${PR_HEAD_SHA}"
          else
            diff_range="${PUSH_BASE_SHA}..${GITHUB_SHA}"
          fi

          git diff --name-only --diff-filter=ACMR -z "$diff_range" -- '*.py' \
            > "$RUNNER_TEMP/changed-python-files"

      - name: Ruff check changed files
        run: |
          xargs -0 -r uv run ruff check --output-format=github \
            < "$RUNNER_TEMP/changed-python-files"

      - name: Ruff format check changed files
        run: |
          xargs -0 -r uv run ruff format --check \
            < "$RUNNER_TEMP/changed-python-files"

      # Phase-in mypy: only check changed Python files so existing type debt
      # does not block every PR. Full-project baseline has 519 errors plus 137
      # notes across 106 files (see .qoder/mypy-baseline.txt); tighten once the
      # baseline is cleaned up.
      # continue-on-error: changed files may still carry legacy errors, so
      # report first and flip to blocking after the debt is burned down.
      - name: Mypy type check changed files
        continue-on-error: true
        run: |
          xargs -0 -r uv run python -m mypy --follow-imports=silent \
            < "$RUNNER_TEMP/changed-python-files"