fix: prerender errors during development

[teemingc]

We've been tracking the current page's prerender option during SSR so we might as well enable some page prerendering behaviours during development to match what they'll get during the build.

---

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. Changesets that add features should be minor and those that fix bugs should be patch. Please prefix changeset messages with feat:, fix:, or chore:.

Edits

• Please ensure that 'Allow edits from maintainers' is checked. PRs without this option may be closed.

[pkg-svelte-dev]

Install the latest version of @sveltejs/kit from 5de4b97:

pnpm add https://pkg.svelte.dev/@sveltejs/kit/c/5de4b9729f2a268b7b545c5845dc15c7d32e702f

Open in pkg.svelte.dev: https://pkg.svelte.dev/repos/kit/pr/16507

[changeset-bot]

🦋 Changeset detected

Latest commit: 5de4b97

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

This PR includes changesets to release 1 package

Name	Type
@sveltejs/kit	Patch

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

[svelte-docs-bot]

Preview: https://svelte-dev-git-preview-kit-16507-svelte.vercel.app/

[Nic-Polumeyv]

state.prerendering || state.prerender_default === true appears nine times now, could be one helper:

/** @param {import('types').SSRState} state */
export function renders_prerendered_output(state) {
	return !!state.prerendering || (DEV && state.prerender_default === true);
}

The DEV guard is optional. Prerendered routes are filtered out of the server manifest at build time so these gates can't fire in production anyway, but that invariant lives in the adapter layer, and this makes the changeset's "during development" true by construction.

[Rich-Harris]

is there any value in doing something like this immediately after this block, rather than adding prerender_default?

if (DEV && prerender) {
  state.prerendering = {
    dependencies: new Map(),
    remote_responses: new Map()
  };
}

It's a bit of a kludge but it would mean we didn't need to worry about forgetting the

|| state.prerender_default === true

part of

if (state.prerendering || state.prerender_default === true) {...}

Or does that not work for whatever reason?

[Nic-Polumeyv]

I checked. event_state is created earlier in respond with prerendering: state.prerendering, so the request store would keep undefined unless that becomes a getter (or gets the same assignment). #16508 makes it a bit less cumbersome.

[Rich-Harris]

That's true. But I think that itself is a problem — we have SSRState and EventState that overlap on prerender, and many functions that take both as arguments.

This was never intentional, just an artifact of organic software development — they should probably be unified into a single object. There are lots of adjacent tidy-ups that would be beneficial. But those would all be slightly disruptive changes that we should hold off on until the PR queue is emptier (specifically of things like #16464 and #15574), and as such I think we should hit pause on both this and #16508.

[teemingc]

is there any value in doing something like this immediately after this block, rather than adding prerender_default?

if (DEV && prerender) {
  state.prerendering = {
    dependencies: new Map(),
    remote_responses: new Map()
  };
}

Only downside I can think of is extra memory usage (if any?? maybe it's miniscule). Otherwise, would be nice to make the code simpler

[Nic-Polumeyv]

With #16605 the desync I described is gone.