Playwright testing: guides, comparisons and tools
Playwright is the browser automation library from Microsoft that drives Chromium, Firefox and WebKit through one API, waits for elements to be ready before acting on them, and records a trace of a failed test. It is the engine Dokimore replays recorded tests on, and the first language a recording exports to, so most of what this site publishes about browser testing is written and measured in Playwright.
The guides below were each run before they were published, in Playwright 1.62.1 against Chromium 151, Firefox 153 and WebKit 26.5, with the numbers and the engine differences in the text. Start with the problem you have: a file upload, a download, a login, a failing test you cannot read, or a selector that keeps breaking.
What people search for
The literal queries this hub is written to answer; each leads to one of the pages below.
- playwright file upload
- playwright download file
- how to debug playwright tests
- playwright login test storageState
- playwright vs selenium
- playwright alternatives
- playwright cross browser testing
Do it: executed how-to guides
Each guide carries the code it ran and what the three engines did with it.
How to test file uploads with Playwright
setInputFiles, the file chooser, a file built in the test, and proving the server received the bytes.
How to test file downloads with Playwright
The download event, the race the wait must win, blob downloads, and the WebKit difference we measured.
How to debug a failed Playwright test
What the timeout message says, what a trace holds, and the three signals to capture on every run.
How to test login flows and authenticated pages
Log in once, save storageState, reuse it; the API route; what a stale session looks like.
XPath vs CSS selectors
Measured on 10,000 elements in three engines, and which selectors survive a changed page.
Cross-browser testing with Playwright
What Chromium, Firefox and WebKit report differently on the same page, from one run.
Learn: the concepts behind the code
Short, answer-first pages on the words the guides use.
How to fix flaky tests
What the studies measured and what Playwright ships for the job.
Locator
What a locator is and why a test stores more than one.
Trace file
What a Playwright trace contains.
Storage state
The cookies and storage a signed-in session is made of.
Compare: Playwright and the alternatives
Where Playwright is the right choice, and where a team looks elsewhere.
Playwright alternatives
The case for staying first, then the tools teams move to and why.
Selenium alternatives
The other direction: teams leaving Selenium for Playwright or a recorder.
Migrating Selenium tests
What transfers, what is recorded again.
Prove it: tools and the product
Try a locator by hand, then see the recorder export the same flow as Playwright code.
XPath and CSS selector tester
Paste HTML, see what a selector matches.
Export to Playwright, Selenium and Robot Framework
A recording as code you own.
The test recorder
Record in a real browser, replay on Playwright, see every healed step.
Questions people ask
Is Playwright a test automation tool or a library?
Both, depending on the package. The playwright library automates a browser from your own code; Playwright Test adds a runner with fixtures, retries, parallel workers and reports. Dokimore uses the library to replay recorded tests and exports Playwright Test files you can run in your own pipeline.
Which browsers does Playwright drive?
Chromium, Firefox and WebKit through their own builds, and Chrome and Edge through the installed channel. Playwright 1.62.1, the version these guides ran, ships Chromium 151, Firefox 153 and WebKit 26.5. The cross-browser guide on this page lists what each one reported differently on the same page.
Do I need to know Playwright to use Dokimore?
No. A test is recorded by clicking through a journey in a real browser and replays without code. Knowing Playwright helps when you export the recording as code to keep in your own repository, which is the direction the product supports; existing Playwright files are not imported.
Try it on your own application
The Free plan asks for no card and runs for 14 days after the trial: record a flow, replay it, and read the evidence each step produced.