# Playwright locator generator: HTML to getByRole

> Paste an element's HTML and get its Playwright locators, best first: getByRole with the accessible name, getByLabel, getByPlaceholder, getByText, getByTestId, in JavaScript or Python. Plus an XPath to CSS converter. Free, in your browser.

Canonical: https://dejo.run/tools/playwright-locator-generator

> For AI agents: the index of this site is https://dejo.run/llms.txt, and every page is also served as markdown (add .md to its address, or send Accept: text/markdown).

**The short answer:** paste an element's HTML and get its Playwright locators, best first: `getByRole` with the element's accessible name, then label, placeholder, text, test id and CSS, in JavaScript or Python. A second tab turns an XPath into a CSS selector.


It runs in your browser; the HTML you paste is not sent anywhere. Each result also shows the matching [Dejo](https://dejo.run/) step where there is one, since Dejo's steps use the same idea: `button "Sign in"` finds the element the way a person would.

## Which Playwright locator to use

Playwright's documentation recommends "prioritizing user-facing attributes and explicit contracts", and says "CSS and XPath are not recommended as the DOM can often change leading to non resilient tests." Its built-in locators, in the order the generator tries them:

| Locator | Finds an element by |
|---|---|
| `getByRole` | its role (button, link, textbox, heading...) and its accessible name |
| `getByLabel` | the text of the label tied to a form control |
| `getByPlaceholder` | an input's placeholder |
| `getByAltText` | an image's text alternative |
| `getByTitle` | its title attribute |
| `getByTestId` | its `data-testid` attribute |
| `getByText` | its text |

## How the role and name are found

The role is the element's own `role` attribute, or the one HTML gives it: a link with `href` is a `link`, a button is a `button`, `<input type="email">` is a `textbox`, `<select>` is a `combobox`, `<h2>` is a `heading`. The name comes from `aria-labelledby`, then `aria-label`, then a tied `<label>` for form controls, `alt` for images, or the element's own text for buttons, links and headings. That is the same order browsers use for these elements, which is what `getByRole` matches against.

## When a locator stops working

A locator that worked yesterday can miss today when the site renames a button or wraps it in a new element. Dejo keeps up to nine ways to find each element and tries them in order, so one change rarely breaks a saved step. See [what happens when a website changes](https://dejo.run/docs/when-the-site-changes).

## Sources

- Playwright locators: [playwright.dev/docs/locators](https://playwright.dev/docs/locators)

## FAQ

## FAQ

### What is the best Playwright locator?

Playwright recommends user-facing locators first: getByRole, then getByLabel, getByPlaceholder, getByText, getByAltText, getByTitle and getByTestId. Its docs say CSS and XPath are not recommended, because the DOM can often change and the test breaks with it.

### How do I get the HTML of an element?

In Chrome DevTools, right click the element in the Elements panel, then Copy, then Copy element. Paste that here. To keep a label with its input, copy both, label first.

### Can every XPath be turned into CSS?

No. CSS has no text(), no parent or sibling axes, and no or/and between conditions. The converter turns the common shapes (tags, attributes, contains, starts-with, positions) into CSS and says when an XPath has no CSS form, so you can use getByText or getByRole instead.

## Related

- [Dejo: browser automation MCP that remembers websites](https://dejo.run/)
- [Dejo MCP server setup: install in a minute](https://dejo.run/docs)
