Fix: Infer content type with charset in dev and prod

[bholmesdev]

Context and open questions

Discord comment:

Okay, I hit a snag with our UTF-8 issue: after a triage call, we decided we should assign "charset=UTF-8" to the header of all API endpoints (in dev only) that use the { body } shorthand. Unfortunately, you can't assign a charset without also assigning a content type.
We could use text/plain; charset=utf-8, but this wouldn't seem valid .json endpoints. We could also infer application/json for the .json extension specifically, but this is a slippery slope for other content types we'd need to support
So yeah... not sure what behavior we want here 🤷‍♂️
Oh interesting, sveltekit just says text/plain;charset=UTF-8 for any {body} shorthand, even for .json files in production

Changes

• Resolves 🐛 BUG: Dev server doesn't return UTF-8 compatible responses on non-html-pages #3573
• Infer content type in dev and prod for { body } shorthand using mine package (no, this isn't a new dependency)
• Set text/plain;charset=utf-8 content type when mime can't be inferred

Testing

Add content type test to ssr-api-route.test.js

Docs

N/A

[changeset-bot]

🦋 Changeset detected

Latest commit: d0b0ab3

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

This PR includes changesets to release 9 packages

Name	Type
astro	Patch
@e2e/astro-component	Patch
@e2e/lit-component	Patch
@e2e/preact-component	Patch
@e2e/react-component	Patch
@e2e/solid-component	Patch
@e2e/svelte-component	Patch
@e2e/e2e-tailwindcss	Patch
@e2e/ts-resolution	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

[natemoo-re]

This makes sense to me!

[tony-sull]

Yep this definitely makes sense, and lines up nicely with the UTF-8 default files get during SSG