Description

For a setup like the following:

const router = sentryCreateHashRouter(
  [
    {
      path: '/',
      element: (
        <Layout/>
      ),
      children: [
        { index: true, element: <BaseRouteRedirect /> },
        { path: '/a', element: <PageA /> },
        { path: '/b/view', element: <PageB /> },
        { path: '*', element: <NoMatch /> },
      ],
    },
  ]
);

When the index route, / is hit, a redirect happens to either page a or page b. This is effectively a redirect guard.

As a consequence, there is no pageloads, instead only navigations, which means that web vital values are not properly calculated for these routes.

We need to figure out a way to make sure that the SDK is aware of this pattern. Essentially redirects should be treated as part of the pageload, with just the name being updated. This needs the code to be handled a little differently.

Relevant slack thread internally: https://sentry.slack.com/archives/C4E31MGD8/p1731320946423889

┆Issue is synchronized with this Jira Improvement by Unito