E2E Tests

Wealthfolio E2E tests use Playwright and run against the web app (not the Tauri desktop app). There are no mocks — both frontend and backend must be running against a fresh database.

Prerequisites

• pnpm installed
• Rust toolchain installed (for the backend server)
• Chrome installed (Playwright uses the system Chrome)

Running E2E Tests

Automated — full suite

The easiest way to run the entire suite. Handles everything automatically: prepares a fresh database, starts the web app, waits for both servers, runs Playwright, and shuts everything down.

pnpm test:e2e

To open the Playwright UI instead:

pnpm test:e2e:ui

Manual — specific tests or debugging

Use this when you want to run a subset of tests or iterate quickly without restarting the server on every run.

Step 1 — Prepare a fresh database

node scripts/prep-e2e.mjs

This creates a new timestamped SQLite database (e.g. db/app-testing-20260411T120000Z.db) and writes its path to .env.web. Run this every time before starting the server — it ensures test isolation.

Step 2 — Start the web app

Option A — watch the terminal output directly:

pnpm run dev:web

Wait until you see Vite's "ready in Xms" and the Rust server binding messages, then move on to Step 3 in a separate terminal.

Option B — redirect output to a log file and use the wait script:

pnpm run dev:web > /tmp/wealthfolio-dev2.log 2>&1 &
./scripts/wait-for-both-servers-to-be-ready.sh

wait-for-both-servers-to-be-ready.sh polls the log file until it detects both "ready in" (Vite) and the Axum server binding on port 8088, then prints the last few lines and exits. The output redirect is required — the script reads from a file, not from a live terminal.

If the web app is already running: Stop it first (Ctrl+C), then re-run prep-e2e.mjs and restart. The running instance is using a stale database — tests assume an empty DB and will silently skip asset creation if data already exists, causing failures for unrelated reasons.

Step 3 — Run specific tests

# Run a specific spec file
npx playwright test e2e/10-symbol-mapping-validation.spec.ts

# Run with browser visible (useful for debugging)
npx playwright test e2e/10-symbol-mapping-validation.spec.ts --headed

# Run all tests
npx playwright test

# Run and open the HTML report afterwards
npx playwright test && npx playwright show-report

Important rules

• Always run prep-e2e.mjs before starting the server. Tests assume an empty database. If you run against an existing database, setup steps may silently skip asset creation and tests may fail for unrelated reasons.
• Do not run E2E tests against the Tauri desktop app. The tests are hardcoded to http://localhost:1420.
• Do not run E2E tests while the Tauri dev server (pnpm tauri dev) is running on the same ports — they conflict.
• Tests run serially (1 worker, serial mode). Do not try to parallelize them.

Test files

Dependency: 10-symbol-mapping-validation.spec.ts requires 01-happy-path.spec.ts to have run first on the same database. Test 0 (setup) calls loginIfNeeded, which expects either a login page or a dashboard — neither exists on a fresh DB until onboarding is complete. Always run spec 01 before spec 10 on a fresh database:

bashCopy

npx playwright test e2e/01-happy-path.spec.ts e2e/10-symbol-mapping-validation.spec.ts

Debugging a failing test

# Run with Playwright inspector (step through actions)
npx playwright test e2e/<spec>.spec.ts --debug

# Show the last HTML report
npx playwright show-report

# Record a trace for a failing test (trace is saved on retry)
# Already configured in playwright.config.ts: trace: "on-first-retry"