[workers-playground] Fix Vite 8 / rolldown crash: TypeError: createRenderer is not a function

[petebacondarwin]

Fixes #[insert GH or internal issue link(s)].

@cloudflare/style-provider ships a hybrid ESM+CJS package: its es/ directory contains files that are nominally ESM but internally use require(). Vite 8's new bundler (rolldown) mishandles this pattern and generates an anonymous, unreachable module initializer, leaving createRenderer as undefined at runtime:

Uncaught TypeError: createRenderer is not a function
    at getRenderer (index-B9Sbpvoo.js:42950:28)

Rolldown wraps the module in a lazy __esmMin thunk but discards the return value (assigns it to no variable), so init_createRenderer() is never called and the createRenderer variable stays undefined.

The fix aliases @cloudflare/style-provider to its CJS build (lib/index.js) in vite.config.ts. Rolldown handles plain CJS correctly via its interop layer, avoiding the broken initializer entirely.

Also replaces base: '/playground' with assetsDir: 'playground/assets': using base causes Vite to write assets to dist/assets/ but reference them as /playground/assets/ in the HTML, which breaks after Wrangler uploads them. Keeping assets under dist/playground/assets/ means the paths in the HTML match their actual deployed location.

---

• Tests
  • Tests included/updated
  • Automated tests not possible - manual testing has been completed as follows: ran pnpm vite build && pnpm vite preview and confirmed the TypeError is gone
  • Additional testing not necessary because:
• Public documentation
  • Cloudflare docs PR(s):
  • Documentation not necessary because: internal build config change only

A picture of a cute animal (not mandatory, but encouraged)

---

[changeset-bot]

🦋 Changeset detected

Latest commit: 5e5489f

The changes in this PR will be included in the next version bump.

Not sure what this means? Click here to learn what changesets are.

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

[workers-devprod]

Codeowners approval required for this PR:

• ✅ @cloudflare/wrangler

Show detailed file reviewers

[ask-bonk]

I see a minor double-space on line 32 ("would cause"), but that's purely cosmetic in a comment — not actionable per my guidelines. The logic and approach are sound.

The change is clean and well-reasoned. Both fixes address real issues:

1. The CJS alias for @cloudflare/style-provider is a valid workaround for a Vite 8/rolldown bug.
2. The base → assetsDir switch correctly fixes the asset path mismatch.

LGTM

github run

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no potential bugs to report.

View in Devin Review to see 2 additional findings.

[pkg-pr-new]

create-cloudflare

npm i https://pkg.pr.new/create-cloudflare@13105

@cloudflare/kv-asset-handler

npm i https://pkg.pr.new/@cloudflare/kv-asset-handler@13105

miniflare

npm i https://pkg.pr.new/miniflare@13105

@cloudflare/pages-shared

npm i https://pkg.pr.new/@cloudflare/pages-shared@13105

@cloudflare/unenv-preset

npm i https://pkg.pr.new/@cloudflare/unenv-preset@13105

@cloudflare/vite-plugin

npm i https://pkg.pr.new/@cloudflare/vite-plugin@13105

@cloudflare/vitest-pool-workers

npm i https://pkg.pr.new/@cloudflare/vitest-pool-workers@13105

@cloudflare/workers-editor-shared

npm i https://pkg.pr.new/@cloudflare/workers-editor-shared@13105

wrangler

npm i https://pkg.pr.new/wrangler@13105

commit: 5e5489f

[github-actions]

✅ All changesets look good

[jamesopstad]

This feels like it should be fixed in either @cloudflare/style-provider or Rolldown, depending on where you believe the bug to be.

[penalosa]

I agree that this should be fixed upstream, but this seems fine as a stopgap

[petebacondarwin]

__esmMin

I think the bug is reported here rolldown/rolldown#8777
and a fix was attempted but closed here rolldown/rolldown#8922

But meanwhile...

[jamesopstad]

__esmMin

I think the bug is reported here rolldown/rolldown#8777 and a fix was attempted but closed here rolldown/rolldown#8922

But meanwhile...

That's fine but in these cases we should reference the upstream issue in the code with a reminder to remove the workaround when it's no longer needed.

[petebacondarwin]

__esmMin

I think the bug is reported here rolldown/rolldown#8777 and a fix was attempted but closed here rolldown/rolldown#8922
But meanwhile...

That's fine but in these cases we should reference the upstream issue in the code with a reminder to remove the workaround when it's no longer needed.

Yes, you're right. Sorry. I'll track this and hopefully it will be released while I am still holding it in memory 😄