feat: WP-0001 + WP-0002 complete — LEVEL1 core + service interfaces

WP-0001 (Foundation & LEVEL1 Core): - manifest model (FR-100), MD→DOCX builder (FR-200), DOCX→MD importer (FR-300/400), template family registry (FR-600), drift detector (FR-700), CLI wiring, pre-commit config, CI skeleton, regression harness WP-0002 (Service Interfaces & Workflow Orchestration): - REST service via FastAPI (FR-900): /health, /version, /capabilities, /templates, /styles, /validate, /build, /import, /compare, /templates/register, /workflows/{name}, /evidence/{run_id} - Evidence & report store (FR-1400): JSON-backed, per-run, retrievable through all interfaces, classification (pass/warnings/failed) - Composite workflow orchestration (FR-1300): single-file-roundtrip, multi-file-roundtrip, release-regression, family-switch-build - MCP server via FastMCP (FR-1000): all tools + resources - CLI additions: `markidocx serve`, `markidocx workflow`, `markidocx mcp` - Interface parity tests: CLI / REST / MCP produce equivalent results 135 tests passing, ruff + mypy clean. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-16 07:46:31 +00:00
parent 42789cad1e
commit 1f3dddf7d6
30 changed files with 4158 additions and 26 deletions
--- a/tests/test_evidence.py
+++ b/tests/test_evidence.py
@@ -0,0 +1,212 @@
+"""Tests for T03 — Evidence & report storage (FR-1400)."""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+
+import pytest
+
+from markidocx.evidence import EvidenceStore, ReportContext, RunReport
+
+
+@pytest.fixture()
+def store(tmp_path: Path) -> EvidenceStore:
+    return EvidenceStore(base_dir=tmp_path / "evidence")
+
+
+# ---------------------------------------------------------------------------
+# EvidenceStore basics (FR-1401–FR-1404)
+# ---------------------------------------------------------------------------
+
+
+def test_new_run_id_is_unique(store: EvidenceStore) -> None:
+    ids = {store.new_run_id() for _ in range(10)}
+    assert len(ids) == 10
+
+
+def test_save_and_get_report(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    data = {"status": "ok", "project": "Test", "warnings": [], "errors": []}
+    store.save_report(run_id, "validation", data)
+    report = store.get_report(run_id, "validation")
+    assert report is not None
+    assert report.run_id == run_id
+    assert report.report_type == "validation"
+    assert report.data["status"] == "ok"
+
+
+def test_get_missing_report_returns_none(store: EvidenceStore) -> None:
+    assert store.get_report("nonexistent-id", "validation") is None
+
+
+def test_save_report_writes_json_file(store: EvidenceStore, tmp_path: Path) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    path = tmp_path / "evidence" / run_id / "build.json"
+    assert path.exists()
+    content = json.loads(path.read_text())
+    assert content["report_type"] == "build"
+
+
+def test_save_all_report_types(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    for rtype in ("validation", "build", "import", "drift"):
+        store.save_report(run_id, rtype, {"status": "ok", "warnings": [], "errors": []})
+    reports = store.list_reports(run_id)
+    types = {r.report_type for r in reports}
+    assert types == {"validation", "build", "import", "drift"}
+
+
+# ---------------------------------------------------------------------------
+# List and retrieve (FR-1409)
+# ---------------------------------------------------------------------------
+
+
+def test_list_runs_empty(store: EvidenceStore) -> None:
+    assert store.list_runs() == []
+
+
+def test_list_runs_after_save(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "validation", {"status": "ok", "warnings": [], "errors": []})
+    assert run_id in store.list_runs()
+
+
+def test_list_reports_empty_for_unknown_run(store: EvidenceStore) -> None:
+    assert store.list_reports("no-such-run") == []
+
+
+def test_list_reports_returns_all(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "validation", {"status": "ok", "warnings": [], "errors": []})
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    reports = store.list_reports(run_id)
+    assert len(reports) == 2
+
+
+# ---------------------------------------------------------------------------
+# Traceability fields (FR-1410)
+# ---------------------------------------------------------------------------
+
+
+def test_report_context_stored_and_retrieved(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    ctx = ReportContext(project="TestDoc", family="article", feature_level="level1", workflow="single-file-roundtrip")
+    store.save_report(run_id, "validation", {"status": "ok", "warnings": [], "errors": []}, context=ctx)
+    report = store.get_report(run_id, "validation")
+    assert report is not None
+    assert report.context.project == "TestDoc"
+    assert report.context.family == "article"
+    assert report.context.feature_level == "level1"
+    assert report.context.workflow == "single-file-roundtrip"
+
+
+def test_report_has_timestamp(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    report = store.get_report(run_id, "build")
+    assert report is not None
+    assert report.created_at  # non-empty ISO timestamp
+
+
+# ---------------------------------------------------------------------------
+# EvidenceSet assembly (FR-1406–FR-1408)
+# ---------------------------------------------------------------------------
+
+
+def test_assemble_set(store: EvidenceStore) -> None:
+    run1 = store.new_run_id()
+    run2 = store.new_run_id()
+    store.save_report(run1, "build", {"status": "ok", "warnings": [], "errors": []})
+    store.save_report(run2, "import", {"status": "ok", "warnings": [], "errors": []})
+    ev_set = store.assemble_set([run1, run2])
+    assert len(ev_set.reports) == 2
+    assert set(ev_set.run_ids) == {run1, run2}
+
+
+def test_evidence_set_classification_pass(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    ev_set = store.assemble_set([run_id])
+    assert ev_set.classification == "pass"
+
+
+def test_evidence_set_classification_pass_with_warnings(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": ["some warning"], "errors": []})
+    ev_set = store.assemble_set([run_id])
+    assert ev_set.classification == "pass-with-warnings"
+
+
+def test_evidence_set_classification_failed(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "error", "warnings": [], "errors": ["oops"]})
+    ev_set = store.assemble_set([run_id])
+    assert ev_set.classification == "failed"
+
+
+def test_evidence_set_composition(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "validation", {"status": "ok", "warnings": [], "errors": []})
+    ev_set = store.assemble_set([run_id])
+    comp = ev_set.composition
+    assert len(comp) == 1
+    assert comp[0]["run_id"] == run_id
+    assert comp[0]["type"] == "validation"
+
+
+def test_evidence_set_summary_keys(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    ev_set = store.assemble_set([run_id])
+    summary = ev_set.summary()
+    assert "classification" in summary
+    assert "run_count" in summary
+    assert "report_count" in summary
+    assert "complete" in summary
+    assert "warnings_count" in summary
+    assert "composition" in summary
+
+
+# ---------------------------------------------------------------------------
+# Human-readable and machine-readable output (FR-1411, FR-1412)
+# ---------------------------------------------------------------------------
+
+
+def test_to_markdown(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    md = store.to_markdown(run_id)
+    assert run_id in md
+    assert "Build" in md
+
+
+def test_to_json(store: EvidenceStore) -> None:
+    run_id = store.new_run_id()
+    store.save_report(run_id, "build", {"status": "ok", "warnings": [], "errors": []})
+    raw = store.to_json(run_id)
+    parsed = json.loads(raw)
+    assert parsed["run_id"] == run_id
+    assert len(parsed["reports"]) == 1
+
+
+# ---------------------------------------------------------------------------
+# RunReport round-trip serialisation
+# ---------------------------------------------------------------------------
+
+
+def test_run_report_to_from_dict() -> None:
+    ctx = ReportContext(project="P", family="article")
+    r = RunReport(
+        run_id="abc",
+        report_type="build",
+        data={"status": "ok", "warnings": [], "errors": []},
+        created_at="2026-01-01T00:00:00+00:00",
+        context=ctx,
+    )
+    d = r.to_dict()
+    r2 = RunReport.from_dict(d)
+    assert r2.run_id == "abc"
+    assert r2.context.project == "P"
+    assert r2.context.family == "article"