breaking: treat special elements that never have content as void elements

[jrmajor]

Since we already know that self-closing tags are not cool, we don't want to write <svelte:options />, and instead have to do <svelte:options></svelte:options>. These elements can never have content anyway, so with these changes, we will be able to just write <svelte:options>

<svelte:options>

<!-- no longer valid -->
<svelte:options></svelte:options>
<svelte:options>this was ignored anyway</svelte:options>
<!-- still valid -->
<svelte:options />
<!-- invalid before -->
<svelte:options>

<svelte:window>, <svelte:document>, <svelte:body>

If there was content inside, compiler would throw invalid-element-content error. Now it will throw invalid-void-content.

<!-- no longer valid -->
<svelte:window></svelte:window>
<!-- still valid -->
<svelte:window />
<!-- invalid before -->
<svelte:window>

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
• Prefix your PR title with feat:, fix:, chore:, or docs:.
• This message body should clearly illustrate what problems it solves.
• Ideally, include a test that fails without this PR but passes with it.

Tests and linting

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

[changeset-bot]

🦋 Changeset detected

Latest commit: 54ab321

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

This PR includes changesets to release 1 package

Name	Type
svelte	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

[Rich-Harris]

I definitely get where your head's at but I don't think we should make this change. Void elements are a concept specific to HTML, and extending it to <svelte:...> components adds learning curve and introduces tricky questions:

• if we treat these elements as HTML, then <svelte:component this={Foo} /> would cease to be valid, unless we make things inconsistent
• if we decide that <svelte:component this={Foo} /> should be invalid for the sake of consistency, then what about <Foo />?
• if Foo.svelte doesn't contain {@render someprop()} then it too could be considered 'void' — should it be treated as such?

Much better to keep the quirks of HTML confined, in my view. But if none of those arguments are persuasive, then consider the behaviour of most IDEs:

Screen.Recording.2024-04-19.at.11.36.24.AM.mov

[dummdidumm]

I stand by my point in #11224 (comment) and also think we should not do this (neither the other PR)

[jrmajor]

I thought we do treat svelte: elements like HTML. If we don't, then it all makes sense :)

Closing this and #11244. Thanks for the reply!