test: extract shared coverage audit fixture + review E2E

- Extract billing.ts fixture into coverage-audit-fixture.ts (DRY)
- Refactor ship-coverage-audit E2E to use shared fixture
- Add review-coverage-audit E2E for Step 4.75
- Update touchfiles: both E2Es depend on shared fixture

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-03-20 07:57:28 -07:00
parent f48b70a240
commit c286a09793
3 changed files with 148 additions and 62 deletions
+76
View File
@@ -0,0 +1,76 @@
/**
* Shared fixture for test coverage audit E2E tests.
*
* Creates a Node.js project with billing source code that has intentional
* test coverage gaps: processPayment has happy-path-only tests,
* refundPayment has no tests at all.
*
* Used by: ship-coverage-audit E2E, review-coverage-audit E2E
*/
import * as fs from 'fs';
import * as path from 'path';
import { spawnSync } from 'child_process';
export function createCoverageAuditFixture(dir: string): void {
// Create a Node.js project WITH test framework but coverage gaps
fs.writeFileSync(path.join(dir, 'package.json'), JSON.stringify({
name: 'test-coverage-app',
version: '1.0.0',
type: 'module',
scripts: { test: 'echo "no tests yet"' },
devDependencies: { vitest: '^1.0.0' },
}, null, 2));
// Create vitest config
fs.writeFileSync(path.join(dir, 'vitest.config.ts'),
`import { defineConfig } from 'vitest/config';\nexport default defineConfig({ test: {} });\n`);
fs.writeFileSync(path.join(dir, 'VERSION'), '0.1.0.0\n');
fs.writeFileSync(path.join(dir, 'CHANGELOG.md'), '# Changelog\n');
// Create source file with multiple code paths
fs.mkdirSync(path.join(dir, 'src'), { recursive: true });
fs.writeFileSync(path.join(dir, 'src', 'billing.ts'), `
export function processPayment(amount: number, currency: string) {
if (amount <= 0) throw new Error('Invalid amount');
if (currency !== 'USD' && currency !== 'EUR') throw new Error('Unsupported currency');
return { status: 'success', amount, currency };
}
export function refundPayment(paymentId: string, reason: string) {
if (!paymentId) throw new Error('Payment ID required');
if (!reason) throw new Error('Reason required');
return { status: 'refunded', paymentId, reason };
}
`);
// Create a test directory with ONE test (partial coverage)
fs.mkdirSync(path.join(dir, 'test'), { recursive: true });
fs.writeFileSync(path.join(dir, 'test', 'billing.test.ts'), `
import { describe, test, expect } from 'vitest';
import { processPayment } from '../src/billing';
describe('processPayment', () => {
test('processes valid payment', () => {
const result = processPayment(100, 'USD');
expect(result.status).toBe('success');
});
// GAP: no test for invalid amount
// GAP: no test for unsupported currency
// GAP: refundPayment not tested at all
});
`);
// Init git repo with main branch
const run = (cmd: string, args: string[]) =>
spawnSync(cmd, args, { cwd: dir, stdio: 'pipe', timeout: 5000 });
run('git', ['init', '-b', 'main']);
run('git', ['config', 'user.email', 'test@test.com']);
run('git', ['config', 'user.name', 'Test']);
run('git', ['add', '.']);
run('git', ['commit', '-m', 'initial commit']);
// Create feature branch
run('git', ['checkout', '-b', 'feature/billing']);
}
+3 -2
View File
@@ -83,8 +83,9 @@ export const E2E_TOUCHFILES: Record<string, string[]> = {
// QA bootstrap
'qa-bootstrap': ['qa/**', 'browse/src/**', 'ship/**'],
// Ship coverage audit
'ship-coverage-audit': ['ship/**'],
// Coverage audit (shared fixture)
'ship-coverage-audit': ['ship/**', 'test/fixtures/coverage-audit-fixture.ts'],
'review-coverage-audit': ['review/**', 'test/fixtures/coverage-audit-fixture.ts'],
// Design
'design-consultation-core': ['design-consultation/**'],
+69 -60
View File
@@ -2727,66 +2727,9 @@ describeIfSelected('Test Coverage Audit E2E', ['ship-coverage-audit'], () => {
copyDirSync(path.join(ROOT, 'ship'), path.join(coverageDir, 'ship'));
copyDirSync(path.join(ROOT, 'review'), path.join(coverageDir, 'review'));
// Create a Node.js project WITH test framework but coverage gaps
fs.writeFileSync(path.join(coverageDir, 'package.json'), JSON.stringify({
name: 'test-coverage-app',
version: '1.0.0',
type: 'module',
scripts: { test: 'echo "no tests yet"' },
devDependencies: { vitest: '^1.0.0' },
}, null, 2));
// Create vitest config
fs.writeFileSync(path.join(coverageDir, 'vitest.config.ts'),
`import { defineConfig } from 'vitest/config';\nexport default defineConfig({ test: {} });\n`);
fs.writeFileSync(path.join(coverageDir, 'VERSION'), '0.1.0.0\n');
fs.writeFileSync(path.join(coverageDir, 'CHANGELOG.md'), '# Changelog\n');
// Create source file with multiple code paths
fs.mkdirSync(path.join(coverageDir, 'src'), { recursive: true });
fs.writeFileSync(path.join(coverageDir, 'src', 'billing.ts'), `
export function processPayment(amount: number, currency: string) {
if (amount <= 0) throw new Error('Invalid amount');
if (currency !== 'USD' && currency !== 'EUR') throw new Error('Unsupported currency');
return { status: 'success', amount, currency };
}
export function refundPayment(paymentId: string, reason: string) {
if (!paymentId) throw new Error('Payment ID required');
if (!reason) throw new Error('Reason required');
return { status: 'refunded', paymentId, reason };
}
`);
// Create a test directory with ONE test (partial coverage)
fs.mkdirSync(path.join(coverageDir, 'test'), { recursive: true });
fs.writeFileSync(path.join(coverageDir, 'test', 'billing.test.ts'), `
import { describe, test, expect } from 'vitest';
import { processPayment } from '../src/billing';
describe('processPayment', () => {
test('processes valid payment', () => {
const result = processPayment(100, 'USD');
expect(result.status).toBe('success');
});
// GAP: no test for invalid amount
// GAP: no test for unsupported currency
// GAP: refundPayment not tested at all
});
`);
// Init git repo with main branch
const run = (cmd: string, args: string[]) =>
spawnSync(cmd, args, { cwd: coverageDir, stdio: 'pipe', timeout: 5000 });
run('git', ['init', '-b', 'main']);
run('git', ['config', 'user.email', 'test@test.com']);
run('git', ['config', 'user.name', 'Test']);
run('git', ['add', '.']);
run('git', ['commit', '-m', 'initial commit']);
// Create feature branch
run('git', ['checkout', '-b', 'feature/billing']);
// Use shared fixture for billing project with coverage gaps
const { createCoverageAuditFixture } = require('./fixtures/coverage-audit-fixture');
createCoverageAuditFixture(coverageDir);
});
afterAll(() => {
@@ -2841,6 +2784,72 @@ Output the diagram directly.`,
}, 180_000);
});
// --- Review Coverage Audit E2E ---
describeIfSelected('Review Coverage Audit E2E', ['review-coverage-audit'], () => {
let reviewCoverageDir: string;
beforeAll(() => {
reviewCoverageDir = fs.mkdtempSync(path.join(os.tmpdir(), 'skill-e2e-review-coverage-'));
// Copy review skill files
copyDirSync(path.join(ROOT, 'review'), path.join(reviewCoverageDir, 'review'));
// Use shared fixture for billing project with coverage gaps
const { createCoverageAuditFixture } = require('./fixtures/coverage-audit-fixture');
createCoverageAuditFixture(reviewCoverageDir);
});
afterAll(() => {
try { fs.rmSync(reviewCoverageDir, { recursive: true, force: true }); } catch {}
});
test('/review Step 4.75 produces coverage diagram', async () => {
const result = await runSkillTest({
prompt: `Read the file review/SKILL.md for the review workflow instructions.
You are on the feature/billing branch. The base branch is main.
This is a test project — there is no remote, no PR to create.
ONLY run Step 4.75 (Test Coverage Diagram) from the review workflow.
Skip all other steps (scope drift, checklist, design review, fix-first, etc.).
The source code is in ${reviewCoverageDir}/src/billing.ts.
Existing tests are in ${reviewCoverageDir}/test/billing.test.ts.
Produce the ASCII coverage diagram showing which code paths are tested and which have gaps.
Output the diagram directly.`,
workingDirectory: reviewCoverageDir,
maxTurns: 15,
allowedTools: ['Bash', 'Read', 'Write', 'Edit', 'Glob', 'Grep'],
timeout: 120_000,
testName: 'review-coverage-audit',
runId,
});
logCost('/review coverage audit', result);
recordE2E('/review Step 4.75 coverage audit', 'Review Coverage Audit E2E', result, {
passed: result.exitReason === 'success',
});
expect(result.exitReason).toBe('success');
// Check output contains coverage diagram elements
const output = result.output || '';
const hasGap = output.includes('GAP') || output.includes('gap') || output.includes('NO TEST');
const hasTested = output.includes('TESTED') || output.includes('tested') || output.includes('✓');
const hasCoverage = output.includes('COVERAGE') || output.includes('coverage') || output.includes('paths tested');
console.log(`Output has GAP markers: ${hasGap}`);
console.log(`Output has TESTED markers: ${hasTested}`);
console.log(`Output has coverage summary: ${hasCoverage}`);
// At minimum, the agent should have read the source and test files
const readCalls = result.toolCalls.filter(tc => tc.tool === 'Read');
expect(readCalls.length).toBeGreaterThan(0);
}, 180_000);
});
// --- Codex skill E2E ---
describeIfSelected('Codex skill E2E', ['codex-review'], () => {