mirror of
https://github.com/proffesor-for-testing/agentic-qe.git
synced 2026-09-19 08:45:47 +08:00
Dragan Spiridonov
4e513b431d
- Delete dead MCP server.ts (911 lines, never used in production — entry.ts imports MCPProtocolServer). Add legacy aliases in index.ts for backwards compat. - Fix ESLint: rename .eslintrc.js to .eslintrc.cjs (CommonJS in ESM project) - Fix hardcoded version '3.0.0' in protocol-server.ts and http-server.ts — now reads dynamically from package.json - Fix vitest process hang: add worker-level afterAll force-exit (setup-worker.ts) and global teardown safety net. Native modules (better-sqlite3, hnswlib-node) held open handles preventing fork worker exit. - Add test gate to npm-publish.yml — 194 unit test files must pass before publish - Disable benchmark.yml release trigger (benchmarks/suite.ts doesn't exist) - Remove continue-on-error from test steps in optimized-ci.yml (was masking failures) - Add getToolDefinitions() to MCPProtocolServer for test introspection - Update e2e and integration tests to use production server API Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
77 lines
2.8 KiB
TypeScript
77 lines
2.8 KiB
TypeScript
/**
|
||
* Agentic QE v3 - Vitest Configuration
|
||
*/
|
||
|
||
import { defineConfig } from 'vitest/config';
|
||
import path from 'path';
|
||
import os from 'os';
|
||
|
||
export default defineConfig({
|
||
test: {
|
||
globals: true,
|
||
environment: 'node',
|
||
env: {
|
||
// ISOLATION: Use temp directory so tests never write to production memory.db.
|
||
// Each test process (via forks pool) gets its own DB in the OS temp dir.
|
||
// This prevents test runs from inflating .agentic-qe/memory.db.
|
||
AQE_PROJECT_ROOT: path.join(os.tmpdir(), `aqe-vitest-${process.pid}`),
|
||
},
|
||
reporters: ['default', 'junit'],
|
||
outputFile: {
|
||
junit: 'junit.xml',
|
||
},
|
||
globalSetup: ['tests/global-teardown.ts'],
|
||
setupFiles: ['tests/setup-worker.ts'],
|
||
teardownTimeout: 5000,
|
||
include: ['tests/**/*.test.ts'],
|
||
benchmark: {
|
||
include: ['tests/**/*.bench.ts'],
|
||
},
|
||
exclude: [
|
||
'node_modules',
|
||
'dist',
|
||
'**/_archived/**',
|
||
'**/browser/**',
|
||
'**/*.e2e.test.ts',
|
||
'**/vibium/**',
|
||
'**/browser-swarm-coordinator.test.ts',
|
||
],
|
||
coverage: {
|
||
provider: 'v8',
|
||
reporter: ['text', 'json', 'html', 'junit'],
|
||
include: ['src/**/*.ts'],
|
||
exclude: ['src/**/*.d.ts', 'src/**/index.ts'],
|
||
},
|
||
testTimeout: 10000,
|
||
hookTimeout: 15000, // Prevent beforeEach/afterEach hangs (e.g., fleet init)
|
||
// OOM & Segfault Prevention for DevPod/Codespaces (Vitest 4 format)
|
||
// Process isolation via forks prevents HNSW native module segfaults.
|
||
// fileParallelism is OFF in DevPod to prevent concurrent fleet inits from
|
||
// allocating multiple HNSW indices + Queen + Kernel stacks simultaneously.
|
||
// CI uses sequential execution too since test files share global fleet state.
|
||
pool: 'forks',
|
||
// OOM Prevention (Issue #294):
|
||
// DevPod has 16GB total. VS Code (~2GB) + Claude (~5GB) = 7GB overhead.
|
||
// Each fleet init allocates ~200-400MB (13 domain plugins + HNSW native memory).
|
||
// With maxForks=2 + main process: 3 × 1GB heap + native = ~4-5GB for tests.
|
||
// Setting maxForks=1 keeps peak at ~2-3GB, leaving headroom for the environment.
|
||
fileParallelism: false,
|
||
maxForks: 1,
|
||
minForks: 1,
|
||
isolate: true, // Full process isolation prevents native module conflicts
|
||
// Fail fast on OOM-prone environments
|
||
bail: process.env.CI ? 5 : 3,
|
||
},
|
||
resolve: {
|
||
alias: {
|
||
'@shared': path.resolve(__dirname, './src/shared'),
|
||
'@kernel': path.resolve(__dirname, './src/kernel'),
|
||
'@domains': path.resolve(__dirname, './src/domains'),
|
||
'@coordination': path.resolve(__dirname, './src/coordination'),
|
||
'@adapters': path.resolve(__dirname, './src/adapters'),
|
||
'@integrations': path.resolve(__dirname, './src/integrations'),
|
||
'@': path.resolve(__dirname, './src'),
|
||
},
|
||
},
|
||
});
|