agileflow 3.4.3 → 4.0.0-alpha.10

package/content/plugins/testing/skills/agileflow-test-writer/references/test-patterns.md

@@ -0,0 +1,420 @@
+# Test Patterns Reference
+
+**Load this when:** choosing test structure for a specific layer, deciding on mocking strategies, picking framework-specific patterns, or structuring a new test file from scratch.
+
+---
+
+## Test layer patterns
+
+### Unit tests
+
+Unit tests verify a single function, class, or module in complete isolation. All dependencies are replaced with fakes, mocks, or stubs.
+
+**When to write unit tests:**
+
+- Pure functions with deterministic input/output
+- Business logic / domain rules
+- Utility functions
+- State machines and reducers
+- Data transformation / mapping functions
+
+**Unit test structure (Vitest / Jest):**
+
+```js
+import { describe, it, expect, beforeEach, vi } from "vitest";
+import { PricingEngine } from "./pricing-engine";
+
+describe("PricingEngine", () => {
+  let engine;
+
+  beforeEach(() => {
+    engine = new PricingEngine({ taxRate: 0.1 });
+  });
+
+  describe("calculateTotal", () => {
+    it("applies tax to the base price", () => {
+      const result = engine.calculateTotal({ basePrice: 100 });
+      expect(result.total).toBe(110);
+    });
+
+    it("applies a percentage discount before tax", () => {
+      const result = engine.calculateTotal({ basePrice: 100, discountPct: 20 });
+      expect(result.total).toBe(88); // (100 * 0.8) * 1.1
+    });
+
+    it("throws when basePrice is negative", () => {
+      expect(() => engine.calculateTotal({ basePrice: -1 })).toThrow(
+        "basePrice must be non-negative",
+      );
+    });
+
+    it("returns zero tax when basePrice is zero", () => {
+      const result = engine.calculateTotal({ basePrice: 0 });
+      expect(result.tax).toBe(0);
+    });
+  });
+});
+```
+
+**Unit test structure (pytest):**
+
+```python
+import pytest
+from pricing_engine import PricingEngine
+
+@pytest.fixture
+def engine():
+    return PricingEngine(tax_rate=0.1)
+
+class TestCalculateTotal:
+    def test_applies_tax_to_base_price(self, engine):
+        result = engine.calculate_total(base_price=100)
+        assert result.total == pytest.approx(110.0)
+
+    def test_applies_discount_before_tax(self, engine):
+        result = engine.calculate_total(base_price=100, discount_pct=20)
+        assert result.total == pytest.approx(88.0)
+
+    def test_raises_on_negative_price(self, engine):
+        with pytest.raises(ValueError, match="non-negative"):
+            engine.calculate_total(base_price=-1)
+
+    @pytest.mark.parametrize("base_price", [0, 0.01, 1, 999999])
+    def test_does_not_raise_on_valid_prices(self, engine, base_price):
+        engine.calculate_total(base_price=base_price)  # should not raise
+```
+
+**Unit test structure (Go):**
+
+```go
+func TestCalculateTotal(t *testing.T) {
+    engine := NewPricingEngine(0.1)
+
+    tests := []struct {
+        name      string
+        basePrice float64
+        discount  float64
+        want      float64
+        wantErr   bool
+    }{
+        {"applies tax to base price", 100, 0, 110.0, false},
+        {"applies discount before tax", 100, 20, 88.0, false},
+        {"rejects negative price", -1, 0, 0, true},
+        {"handles zero price", 0, 0, 0.0, false},
+    }
+
+    for _, tt := range tests {
+        t.Run(tt.name, func(t *testing.T) {
+            got, err := engine.CalculateTotal(tt.basePrice, tt.discount)
+            if (err != nil) != tt.wantErr {
+                t.Fatalf("wantErr=%v, got err=%v", tt.wantErr, err)
+            }
+            if got != tt.want {
+                t.Errorf("got %v, want %v", got, tt.want)
+            }
+        })
+    }
+}
+```
+
+---
+
+### Integration tests
+
+Integration tests verify that two or more real components work together correctly. Use a real database (test instance or in-memory), real module dependencies, but no external network calls.
+
+**When to write integration tests:**
+
+- Repository / data access layer (real DB queries)
+- Service + repository wired together
+- API route + business logic
+- Module A calling Module B with real data
+
+**Integration test structure (Vitest + real SQLite):**
+
+```js
+import { describe, it, expect, beforeAll, afterAll, beforeEach } from "vitest";
+import { UserRepository } from "./user-repository";
+import { createTestDatabase, seedUsers, clearUsers } from "../test-helpers/db";
+
+describe("UserRepository", () => {
+  let db;
+  let repo;
+
+  beforeAll(async () => {
+    db = await createTestDatabase(); // in-memory SQLite
+    repo = new UserRepository(db);
+  });
+
+  afterAll(() => db.close());
+  beforeEach(() => clearUsers(db));
+
+  it("creates a user and retrieves it by id", async () => {
+    const created = await repo.create({
+      email: "alice@example.com",
+      name: "Alice",
+    });
+    const found = await repo.findById(created.id);
+    expect(found.email).toBe("alice@example.com");
+  });
+
+  it("returns null when user does not exist", async () => {
+    const found = await repo.findById("non-existent-id");
+    expect(found).toBeNull();
+  });
+
+  it("lists all users ordered by created_at desc", async () => {
+    await seedUsers(db, [
+      { email: "a@example.com" },
+      { email: "b@example.com" },
+    ]);
+    const users = await repo.findAll();
+    expect(users).toHaveLength(2);
+    expect(users[0].created_at >= users[1].created_at).toBe(true);
+  });
+});
+```
+
+**Integration test structure (pytest + SQLAlchemy):**
+
+```python
+import pytest
+from sqlalchemy import create_engine
+from sqlalchemy.orm import sessionmaker
+from models import Base, User
+from repositories import UserRepository
+
+@pytest.fixture(scope="function")
+def db_session():
+    engine = create_engine("sqlite:///:memory:")
+    Base.metadata.create_all(engine)
+    Session = sessionmaker(bind=engine)
+    session = Session()
+    yield session
+    session.close()
+    Base.metadata.drop_all(engine)
+
+@pytest.fixture
+def user_repo(db_session):
+    return UserRepository(db_session)
+
+def test_create_and_retrieve_user(user_repo):
+    created = user_repo.create(email="alice@example.com", name="Alice")
+    found = user_repo.find_by_id(created.id)
+    assert found.email == "alice@example.com"
+```
+
+---
+
+### API / HTTP integration tests
+
+Test HTTP endpoints with a real application instance (no real external dependencies).
+
+**Express + Supertest (Jest/Vitest):**
+
+```js
+import request from "supertest";
+import { app } from "../app";
+import { createTestDatabase } from "../test-helpers/db";
+
+describe("POST /api/users", () => {
+  it("creates a user and returns 201", async () => {
+    const res = await request(app)
+      .post("/api/users")
+      .send({ email: "new@example.com", name: "New User" });
+
+    expect(res.status).toBe(201);
+    expect(res.body.id).toBeDefined();
+    expect(res.body.email).toBe("new@example.com");
+  });
+
+  it("returns 400 when email is missing", async () => {
+    const res = await request(app)
+      .post("/api/users")
+      .send({ name: "No Email" });
+
+    expect(res.status).toBe(400);
+    expect(res.body.errors).toContain("email is required");
+  });
+
+  it("returns 409 when email already exists", async () => {
+    await request(app)
+      .post("/api/users")
+      .send({ email: "dup@example.com", name: "A" });
+    const res = await request(app)
+      .post("/api/users")
+      .send({ email: "dup@example.com", name: "B" });
+    expect(res.status).toBe(409);
+  });
+});
+```
+
+---
+
+## Mocking patterns
+
+### Mock at the boundary — not inside the unit
+
+```
+Your module → [MOCK HERE] → External dependency (DB, HTTP, FS, Time)
+```
+
+Do NOT mock:
+
+- The module under test
+- Helper utilities that are pure functions
+- The production database schema (use test DB instead)
+
+### Vitest / Jest mock patterns
+
+**Mock a module import:**
+
+```js
+vi.mock("./email-service", () => ({
+  sendEmail: vi.fn().mockResolvedValue({ messageId: "msg-123" }),
+}));
+```
+
+**Mock with implementation factory:**
+
+```js
+const mockSend = vi.fn();
+vi.mock("./email-service", () => ({ sendEmail: mockSend }));
+
+it("sends a welcome email on registration", async () => {
+  mockSend.mockResolvedValueOnce({ messageId: "abc" });
+  await registerUser({ email: "x@example.com" });
+  expect(mockSend).toHaveBeenCalledWith({
+    to: "x@example.com",
+    template: "welcome",
+  });
+});
+```
+
+**Spy on real module (partial mock):**
+
+```js
+import * as emailService from "./email-service";
+const spy = vi
+  .spyOn(emailService, "sendEmail")
+  .mockResolvedValue({ messageId: "spy-id" });
+```
+
+**Mock Date.now for time-sensitive tests:**
+
+```js
+vi.setSystemTime(new Date("2025-01-01T00:00:00Z"));
+// ... test ...
+vi.useRealTimers();
+```
+
+### pytest mock patterns
+
+```python
+from unittest.mock import patch, MagicMock
+
+def test_sends_welcome_email(mocker):
+    mock_send = mocker.patch('services.email_service.send_email')
+    mock_send.return_value = {'message_id': 'abc'}
+
+    register_user(email='x@example.com')
+
+    mock_send.assert_called_once_with(
+        to='x@example.com',
+        template='welcome'
+    )
+```
+
+### Fake in-memory implementations
+
+Prefer fakes over mocks for stable interfaces:
+
+```js
+// Fake in-memory user repository (stable interface, no brittle call assertions)
+class InMemoryUserRepository {
+  constructor() {
+    this.users = new Map();
+  }
+  async findById(id) {
+    return this.users.get(id) ?? null;
+  }
+  async create(data) {
+    const user = { id: crypto.randomUUID(), ...data };
+    this.users.set(user.id, user);
+    return user;
+  }
+  async delete(id) {
+    this.users.delete(id);
+  }
+}
+```
+
+---
+
+## Async test patterns
+
+**Always await async operations — never fire-and-forget in tests:**
+
+```js
+// Bad — test may pass before rejection
+it("resolves correctly", () => {
+  fetchUser(1).then((u) => expect(u.id).toBe(1));
+});
+
+// Good
+it("resolves correctly", async () => {
+  const user = await fetchUser(1);
+  expect(user.id).toBe(1);
+});
+
+// Good — testing rejection
+it("rejects when user not found", async () => {
+  await expect(fetchUser(999)).rejects.toThrow("User not found");
+});
+```
+
+---
+
+## Test data patterns
+
+### Factory functions (preferred over hard-coded objects)
+
+```js
+function createUser(overrides = {}) {
+  return {
+    id: "user-1",
+    email: "test@example.com",
+    name: "Test User",
+    role: "member",
+    createdAt: new Date("2025-01-01"),
+    ...overrides,
+  };
+}
+
+// Usage — only specify what matters for the test
+const admin = createUser({ role: "admin" });
+const unverified = createUser({ emailVerified: false });
+```
+
+### Fixtures vs factories
+
+| Pattern              | When to use                                                  |
+| -------------------- | ------------------------------------------------------------ |
+| **Factory function** | Test data that varies per test; prefer this by default       |
+| **Fixture file**     | Large, complex, stable data (e.g., a full JSON API response) |
+| **Seeding helper**   | Integration tests that need data in a real DB                |
+| **Inline object**    | Simple tests with one or two fields — factory is overkill    |
+
+---
+
+## Common anti-patterns to avoid
+
+| Anti-pattern                                               | Problem                                         | Fix                                                         |
+| ---------------------------------------------------------- | ----------------------------------------------- | ----------------------------------------------------------- |
+| `expect(result).toBeTruthy()`                              | Passes on `1`, `"false"`, `{}` — proves nothing | Use `toBe(true)`, `toEqual(expected)`, or specific matchers |
+| `expect(mockFn).toHaveBeenCalled()` without argument check | Doesn't verify correct arguments were passed    | Use `toHaveBeenCalledWith(...)`                             |
+| Tests that depend on execution order                       | Brittle — fails when test runner parallelises   | Use `beforeEach` to reset state; never share mutable state  |
+| `setTimeout(fn, 100)` in tests                             | Flaky — timing-dependent                        | Use `vi.useFakeTimers()` / `jest.useFakeTimers()`           |
+| Asserting on full response body                            | Brittle to schema changes                       | Assert on specific fields that matter for the test          |
+| Empty `catch` blocks                                       | Hides failures                                  | Always `throw` or `fail()` in error branches                |
+| Test file with 0 `expect` calls                            | Doesn't assert anything                         | Vitest/Jest won't catch this — audit tests manually         |

package/content/plugins/testing/skills/agileflow-test-writer/workflows/add-coverage.md

@@ -0,0 +1,154 @@
+# Workflow: Add Coverage to Existing Code
+
+**Triggers:** "increase test coverage", "add tests to this file", "this module has no tests", user mentions low coverage percentages
+
+**Goal:** Analyse an existing file or module, identify the highest-value untested paths, and add tests that meaningfully improve confidence — not just line counts.
+
+---
+
+## Inputs needed
+
+| Input                              | Required | How to get it                                                   |
+| ---------------------------------- | -------- | --------------------------------------------------------------- |
+| Target file or module              | Yes      | User specifies; or ask                                          |
+| Current coverage report            | No       | Run `npm test -- --coverage` or `pytest --cov` and paste output |
+| Test framework                     | Yes      | Detect from config                                              |
+| Existing test file for this module | No       | Search for `*.test.*` or `*_test.*` adjacent to the file        |
+
+---
+
+## Steps
+
+### Step 1: Read the target file
+
+Read the source file completely. Build a mental model of:
+
+- Every exported function / class / method
+- Every branch (if/else, switch, ternary, nullish coalescing)
+- Every error path (throw, reject, catch, return null/undefined)
+- Every async operation
+
+### Step 2: Find (or analyse) what's already tested
+
+If an existing test file exists, read it. Map each test to the branches it covers. Identify the gaps.
+
+If no test file exists, every branch is untested.
+
+If a coverage report is available, read it to find:
+
+- Uncovered lines (marked with `|` in text reports)
+- Uncovered branches (marked with `U` in branch summary)
+
+### Step 3: Rank uncovered paths by value
+
+Not all untested code is equally worth testing. Rank by:
+
+| Priority | What to test                                                   | Rationale                         |
+| -------- | -------------------------------------------------------------- | --------------------------------- |
+| P0       | Error paths that could cause data loss or security holes       | Silent failure is dangerous       |
+| P0       | All branches in auth / permission checks                       | Wrong result = security breach    |
+| P1       | Branches in core business logic (pricing, routing, validation) | Bugs here cost real money         |
+| P1       | Edge cases on public API inputs                                | API contracts matter              |
+| P2       | Happy paths not yet covered                                    | Baseline confidence               |
+| P3       | Simple getters / trivial branches                              | Low ROI; skip if time-constrained |
+
+Present the prioritised list to the user if it's long (10+ items).
+
+### Step 4: Choose the right test layer
+
+For each gap:
+
+- If the gap is in a pure function → unit test
+- If the gap is in a database query / ORM call → integration test (with test DB)
+- If the gap is in an HTTP handler → integration test (with supertest / httptest)
+- If the gap is in a UI component → unit test with a render utility (React Testing Library, Vue Test Utils)
+
+### Step 5: Write the tests
+
+For each P0 and P1 gap:
+
+1. Check if a factory or fixture exists for the required test data
+2. Write the test in the existing test file (or create one next to the source file)
+3. Follow the Arrange → Act → Assert pattern
+4. Use specific assertions — not `toBeTruthy`
+
+### Step 6: Estimate coverage improvement
+
+After writing tests, estimate:
+
+- Which lines / branches are now covered
+- Expected new coverage % (rough estimate)
+- What is still uncovered and why (e.g. infrastructure code, trivial getters excluded intentionally)
+
+### Step 7: Present the result
+
+```xml
+<invoke name="AskUserQuestion">
+<parameter name="questions">[{
+  "question": "Coverage analysis complete for {filename}. Found {N} uncovered paths. Added {M} tests targeting the {K} highest-value gaps.",
+  "header": "Next step",
+  "multiSelect": false,
+  "options": [
+    {"label": "Write the updated test file now (Recommended)", "description": "I'll update {test-file-path} with the {M} new tests"},
+    {"label": "Also cover the remaining {R} lower-priority paths", "description": "Full coverage pass — will add {R} more tests for trivial branches and getters"},
+    {"label": "Review the test plan first", "description": "I'll show you each test I'm planning before writing anything"},
+    {"label": "Run coverage to confirm the improvement", "description": "I'll write the file then run the test suite to show the before/after coverage delta"}
+  ]
+}]</parameter>
+</invoke>
+```
+
+---
+
+## Coverage gap analysis template
+
+When presenting gaps to the user, use this format:
+
+```
+Coverage gaps in src/services/user-service.ts
+──────────────────────────────────────────────
+Estimated current: ~45% branch coverage
+
+P0 gaps (write these first):
+  ✗ Line 34-38: deleteUser() — no test for the DB error path (throws CascadeError)
+  ✗ Line 67: checkPermission() — only 'admin' role tested; 'viewer' and 'editor' untested
+
+P1 gaps:
+  ✗ Line 102: updateEmail() — happy path exists but no test for duplicate email (409 case)
+  ✗ Line 118-125: sendVerificationEmail() — called but never asserted on
+
+P2 gaps:
+  ✗ Line 77: getUser() — null return when user not found has no test
+  ✗ Line 89: listUsers() — empty result set not tested
+
+Skip (low ROI):
+  ✗ Line 12: constructor — getter only, no logic
+  ✗ Line 200-202: formatName() — trivial string concat
+
+Estimated coverage after writing P0 + P1: ~78% branch coverage
+```
+
+---
+
+## Iterative coverage improvement (brownfield strategy)
+
+For large legacy files with very low coverage, don't try to cover everything at once. Use the ratchet approach:
+
+1. **Today:** Write tests for all P0 and P1 gaps in the files you're currently touching
+2. **Add a coverage gate:** Set the CI threshold to current coverage − 0% (no regression allowed)
+3. **Each PR:** Add tests for new code + one cleanup pass on P1 gaps in touched files
+4. **Never go backwards:** If coverage drops, it fails CI
+
+This turns coverage improvement into an automatic side effect of regular development.
+
+---
+
+## When coverage is the wrong goal
+
+Flag to the user when you detect:
+
+- Tests that only check `expect(result).toBeDefined()` — this is coverage inflation, not real testing
+- Tests that call the function but assert nothing useful
+- 100% coverage on a file that has no assertions (this is possible!)
+
+In these cases, improving assertion quality matters more than adding new tests.