How to Test PDF Generation and Downloads

Short answer

PDF invoices, statements, and legal documents are high-value artifacts—clicking "Download" without reading content misses wrong totals and missing clauses. Use Playwright page.waitForEvent('download') registered before click, save to disk, parse text with pdf-parse, assert seeded order ids and amounts—not today's date string snapshots.

Part of Testing Guides by integrations.

Who this is for

Invoicing, fintech, healthcare, legal, and HR apps generating PDFs client-side (jsPDF), server-side (Puppeteer, wkhtmltopdf), or via third-party document APIs.

Why testing PDF downloads matters

• Billing disputes — PDF total ≠ charged amount
• Legal exposure — missing signature block or wrong jurisdiction text
• Silent failures — download event never fires; user sees spinner
• Dynamic content flake — snapshot includes Generated on {today}

Complexity map

Scenario	Edge case	Why tests break	Approach
Download event	Listener after click	Lost event	waitForEvent before click
Dynamic dates	Daily snapshot fail	Flake	Assert order id not date
Async generation	Job queue	Click too early	Poll probe then download
Pop-up window	New tab PDF	Wrong page	expect popup or direct URL
Large PDF	Parse timeout	CI fail	Increase timeout; assert key page
Password PDF	Encrypted	Parse fails	Test unencrypted test env
Multi-language	Font missing	Garbled text	Assert ASCII identifiers
Content-Disposition	Inline vs attach	No download event	goto URL with request
Zero-byte file	500 masked as PDF	False pass	Assert file size > 0
Template version	v2 clause missing	Compliance	Assert clause substring

Playwright download pattern

From Playwright downloads docs:

import fs from 'fs';
import path from 'path';
import pdf from 'pdf-parse';

test('invoice PDF contains order total', async ({ page }) => {
  await page.goto(`/orders/${orderId}/invoice`);

  const downloadPromise = page.waitForEvent('download');
  await page.getByRole('button', { name: /download pdf/i }).click();
  const download = await downloadPromise;

  const filePath = path.join(test.info().outputDir, 'invoice.pdf');
  await download.saveAs(filePath);
  expect(fs.statSync(filePath).size).toBeGreaterThan(0);

  const buffer = fs.readFileSync(filePath);
  const { text } = await pdf(buffer);

  expect(text).toContain('ORDER-98765');
  expect(text).toContain('$1,234.56');
  expect(text).not.toContain('{{'); // template leak
});

Critical: register waitForEvent('download') before the click that triggers download.

Async PDF generation

When PDF builds in background job:

await page.getByRole('button', { name: /generate invoice/i }).click();
await expect.poll(() => probeInvoiceStatus(orderId)).toBe('ready');

const downloadPromise = page.waitForEvent('download');
await page.getByRole('link', { name: /download/i }).click();
// ... parse as above

Assert stable fields only

Assert	Avoid
Order id, SKU, totals	"Generated March 15, 2026" unless seeded
Legal clause id string	Full page layout pixel diff every PR
Customer name from seed	Random footer timestamps

Direct URL download (no click)

const response = await page.request.get(`/api/invoices/${orderId}/pdf`);
expect(response.headers()['content-type']).toContain('application/pdf');
const buffer = await response.body();
const { text } = await pdf(buffer);
expect(text).toContain('TOTAL');

Faster for CI when UI click is not the subject under test.

Probe + PDF cross-check

const probe = await request.get(`/api/test/probe-invoice/${orderId}`).then(r => r.json());
// ... parse PDF ...
expect(text).toContain(String(probe.total_cents / 100));

Probe is authoritative when PDF generation bug suspected.

Anti-patterns

Anti-pattern	Why it fails	Better approach
Click then waitForEvent	Race lost	Register promise first
Assert download started only	Empty PDF	pdf-parse content
Snapshot PDF bytes	Dynamic gen time	Text substring asserts
Skip async job wait	Corrupt file	Poll probe ready
Visual diff every PR	Font/OS variance	Text asserts; visual nightly

Example scenario

Situation: User downloads invoice PDF for seeded order with known line items.

Expected outcome: PDF text includes order id, line SKU, and total matching probe invoice row.

Why UI-only automation breaks: Download succeeds but PDF shows wrong tax line—probe and PDF disagree.

1. Arrange: Seed order ORDER-98765 total $1234.56 via test API.
2. Act: Open invoice page; waitForEvent download before click Download PDF.
3. Assert: pdf-parse contains ORDER-98765 and amount; matches probe.total_cents.

TestChimp workflow: Track document_type in TrueCoverage when credit_note PDF added without test.

Same Arrange/Act/Assert pattern as expired-coupon checkout.

Connect scenarios to your QA workflow

Capture business rules in markdown test plans and enforce them with seed routes and probe Assert. Link SmartTests with scenario annotations for requirement traceability. Use /testchimp test on PRs; /testchimp explore on SmartTest paths for non-functional gaps (ExploreChimp).

Related scenarios

• Legal e-signatures — signed PDF workflows
• Fintech apps — statements
• File uploads — inbound PDF upload
• Webhooks async — async PDF job

External references

• Playwright downloads
• pdf-parse on npm
• Playwright download path API

Frequently asked questions

How do I verify PDF content in Playwright?

Capture download with waitForEvent before click, saveAs to test output, parse with pdf-parse, assert text contains order id and totals. Cross-check probe invoice row when available.

Why does my download test flake?

Usually waitForEvent registered after click—register promise first. Or PDF not ready—poll probe job status before download.

How do I avoid failing on today's date in PDF?

Assert stable seeded identifiers (order id, SKU, amounts)—not generation timestamp strings unless you freeze clock in test env.

Can I test PDF without clicking Download?

Yes—page.request.get on PDF URL with application/pdf content-type and pdf-parse body. Useful when UI is out of scope.

Large PDFs timeout in CI—increase what?

pdf-parse timeout and test timeout; assert first-page key strings instead of parsing 200-page docs in default PR job.

Should I visual-diff PDFs?

Reserve for release or legal layout-critical docs. Default CI uses text asserts on financial fields.

New credit_note document type in prod—how to add coverage?

TrueCoverage document_type slice flags gap. Add seed + download spec via /testchimp evolve linked to markdown scenario.

Apply these patterns in your repo

Run `/testchimp init` to connect TestChimp to your repo, then `/testchimp test` on PRs to turn these patterns into maintained SmartTests. Use `/testchimp evolve` when you want to expand coverage as your app grows.

Start free on TestChimp · Book a demo

Related pages

• How to Test E-Signature and Document Workflows
• How Startups Test Fintech Web Apps
• How to Test File Upload Flows (CSV, Images, Docs)
• How to Test Stripe Webhooks in CI
• How to Test Ecommerce Checkout Flows Without Flaky Tests
• AI Testing Tool for Startups