The exact mount the score was measured on. Scored result.
Tabreach the triggerEntershould open with focus insideTabrepeatedly: should cycle inside the dialog onlyEscshould close and return focus to the trigger
The same mount source, editable, with Radix UI resolved at the exact scored version. Example styling ships as polish.css; delete its import for the bare mount. Runs on CodeSandbox's bundler; the measurement belongs to the untouched mount.
import*asDialogfrom"@radix-ui/react-dialog";import{TEXT}from"./harness-kit";/**
* Radix UI modal dialog, mounted exactly as the Radix documentation describes.
*
* Nothing here manages focus, adds ARIA attributes, or handles keys. That is the
* entire point, we are measuring what the library does unaided. If you find
* yourself adding an onKeyDown to make an assertion pass, you have found a
* defect and are in the process of hiding it.
*
* The aria-labels on the two text inputs are the one exception, and they are
* fine: those inputs are harness furniture, not library components, and the
* spec makes no assertion about them. They exist only to give the focus trap
* something to traverse.
*/exportfunctionDialogHarness(){return(<><buttondata-testid="hr-before"type="button">
Before
</button><Dialog.Root><Dialog.Triggerdata-testid="hr-trigger">{TEXT.dialogTrigger}</Dialog.Trigger><Dialog.Portal><Dialog.Overlay/><Dialog.Contentdata-testid="hr-dialog"><Dialog.Titledata-testid="hr-title">{TEXT.dialogTitle}</Dialog.Title><Dialog.Description>
A fixed harness for automated accessibility conformance testing.
</Dialog.Description><inputdata-testid="hr-field-1"type="text"aria-label="Field one"/><inputdata-testid="hr-field-2"type="text"aria-label="Field two"/><Dialog.Closedata-testid="hr-close">{TEXT.close}</Dialog.Close></Dialog.Content></Dialog.Portal></Dialog.Root><buttondata-testid="hr-outside-content"type="button">{TEXT.outsideContent}</button><buttondata-testid="hr-after"type="button">
After
</button></>);}
Self-run companion check, approximated in-browser from ARIA attributes and visibility. Official scores are measured by the runner with replayable Playwright traces.