test: prove desktop discovery preserves imported and unmeasured cases

Author: waleedlatif1

.github/CI.md

@@ -47,4 +47,4 @@ The separate `desktop-e2e` workflow retains its desktop-specific path filters. I
 
 The initial weights are median test durations in seconds from runs [39734295953](https://github.com/simstudioai/sim/actions/runs/39734295953), [39833640524](https://github.com/simstudioai/sim/actions/runs/39833640524), and [39836596108](https://github.com/simstudioai/sim/actions/runs/39836596108). Refresh them from complete `desktop-smoke-reports-*` artifacts when the workload changes: sum `tests[].results[].duration` by `spec.file` in each run, take the median across runs, and divide milliseconds by 1,000. The weights steer scheduling only; Playwright discovery owns coverage. Before adopting new weights, compare repeated paired runs and the complete case/status multiset, including skipped cases and retries.
 
-Each shard uploads its coverage/timing report, the complete discovery report, and its deterministic shard plan. `bun run test:scripts scripts/duration-shards.test.ts` checks partition coverage, unknown files, deterministic ordering, and balancing without launching Electron.
+Each shard uploads its coverage/timing report, the complete discovery report, and its deterministic shard plan. `bun run test:scripts scripts/duration-shards.test.ts scripts/desktop-e2e.test.ts` checks partition coverage, unknown files, deterministic ordering, and balancing. The CLI regression executes tiny browser-free Playwright fixtures to prove that imported test factories and paths containing regex characters are selected correctly; it launches no Electron or browser.

scripts/desktop-e2e.test.ts

@@ -0,0 +1,65 @@
+import { spawnSync } from 'node:child_process'
+import { mkdirSync, mkdtempSync, readFileSync, rmSync, symlinkSync, writeFileSync } from 'node:fs'
+import { tmpdir } from 'node:os'
+import { join, resolve } from 'node:path'
+import { expect, it } from 'vitest'
+
+it('executes imported test factories and unmeasured files exactly once across desktop shards', () => {
+  const root = resolve(import.meta.dirname, '..')
+  const fixture = mkdtempSync(join(tmpdir(), 'desktop-shard-'))
+  try {
+    symlinkSync(join(root, 'node_modules'), join(fixture, 'node_modules'), 'dir')
+    writeFileSync(join(fixture, 'package.json'), '{"type":"module"}')
+    writeFileSync(
+      join(fixture, 'playwright.config.ts'),
+      "export default { testDir: './tests', workers: 1, retries: 0 }"
+    )
+    mkdirSync(join(fixture, 'tests/nested'), { recursive: true })
+    mkdirSync(join(fixture, 'tests/support'), { recursive: true })
+    writeFileSync(
+      join(fixture, 'tests/support/factory.ts'),
+      "import { test } from '@playwright/test'; export function register() { test('factory', () => {}); }"
+    )
+    writeFileSync(
+      join(fixture, 'tests/nested/a[1].spec.ts'),
+      "import { register } from '../support/factory'; register();"
+    )
+    for (const file of ['brand-new.test.ts', 'ordinary.spec.ts']) {
+      writeFileSync(
+        join(fixture, 'tests', file),
+        `import { test } from '@playwright/test'; test('${file}', () => {});`
+      )
+    }
+    const cases: { title: string; status: string }[] = []
+    for (const index of [1, 2, 3]) {
+      const reportPath = join(fixture, `${index}.json`)
+      const result = spawnSync('bun', ['run', join(root, 'scripts/desktop-e2e.ts'), `${index}/3`], {
+        cwd: fixture,
+        env: { ...process.env, PLAYWRIGHT_JSON_OUTPUT_FILE: reportPath },
+        encoding: 'utf8',
+        timeout: 10_000,
+      })
+      expect(result.status, result.stdout + result.stderr).toBe(0)
+      const report = JSON.parse(readFileSync(reportPath, 'utf8')) as {
+        suites: { specs: { title: string; tests: { results: { status: string }[] }[] }[] }[]
+      }
+      for (const suite of report.suites) {
+        for (const spec of suite.specs) {
+          for (const test of spec.tests) {
+            for (const run of test.results) cases.push({ title: spec.title, status: run.status })
+          }
+        }
+      }
+    }
+    expect(cases).toEqual(
+      expect.arrayContaining([
+        { title: 'factory', status: 'passed' },
+        { title: 'brand-new.test.ts', status: 'passed' },
+        { title: 'ordinary.spec.ts', status: 'passed' },
+      ])
+    )
+    expect(cases).toHaveLength(3)
+  } finally {
+    rmSync(fixture, { recursive: true, force: true })
+  }
+})