use free variable for native fetch

[Rich-Harris]

i have ABSOLUTELY NO IDEA how this passed CI in the first place, but apparently #5041 broke a bunch of stuff somehow? my browser suddenly started complaining that you can't call fetch if it's a property of something that isn't window. seems like a weird constraint but who am i to judge.

no changeset because #5041 wasn't released yet, so no need

Please don't delete this checklist! Before submitting the PR, please make sure you do the following:

• It's really useful if your PR references an issue where it is discussed ahead of time. In many cases, features are absent for a reason. For large changes, please create an RFC: https://github.com/sveltejs/rfcs
• This message body should clearly illustrate what problems it solves.
• Ideally, include a test that fails without this PR but passes with it.

Tests

• Run the tests with pnpm test and lint the project with pnpm lint and pnpm check

Changesets

• If your PR makes a change that should be noted in one or more packages' changelogs, generate a changeset by running pnpm changeset and following the prompts. All changesets should be patch until SvelteKit 1.0

[changeset-bot]

⚠️ No Changeset found

Latest commit: 77f5d5f

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

This PR includes no changesets

When changesets are added to this PR, you'll see the packages that this PR includes changesets for and the associated semver types

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[mrkishi]

Huh. Are the failures on Safari/Firefox? We were fooled by green checks pretty hard.

[Rich-Harris]

No, just running pnpm test locally. Truly weird

[Conduitry]

Somewhat recently, there was a change to have Playwright use whatever version of Chrome it could find on your system, rather than using separate downloaded binaries. I wonder whether that could be related to the difference in behavior in the tests?

I'd be in favor of reverting that change anyway. Requiring the user have Chrome installed on their system makes local dev less portable, and not using a specific version makes tests less reproducible.

[sartoshi-foot-dao]

My latest local builds using '@sveltejs/kit': [email protected] and '@sveltejs/adapter-vercel': 1.0.0-next.52 failing for perhaps the same reason:

> Using @sveltejs/adapter-vercel
✘ [ERROR] Could not resolve "@sveltejs/kit/install-fetch"

    .svelte-kit/vercel-tmp/serverless.js:1:29:
      1 │ import { installFetch } from '@sveltejs/kit/install-fetch';
        ╵                              ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

  The path "./install-fetch" is not exported by package "@sveltejs/kit":

    node_modules/@sveltejs/kit/package.json:47:13:
      47 │   "exports": {
         ╵              ^

  You can mark the path "@sveltejs/kit/install-fetch" as external to exclude it from the bundle,
  which will remove this error.

> Build failed with 1 error:
.svelte-kit/vercel-tmp/serverless.js:1:29: ERROR: Could not resolve "@sveltejs/kit/install-fetch"

Hope this helps.

Edit: perhaps I need to update adapter-vercel to .54

Edit2: That did indeed fix it. Sorry for the noise. Thank you for fixing things so quickly.