feat: support ?inline query on svelte style virtual modules
#1024
+46
−4
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
dominikg
reviewed
Nov 22, 2024
dominikg
reviewed
Nov 25, 2024
| optimizeDeps: { | ||
| // eagerly include these, otherwise vite optimizer might interfere with restarting while the test is running | ||
| include: ['svelte-i18n', 'e2e-test-dep-svelte-api-only'] | ||
| } | ||
| }; | ||
|
|
||
| export default config; | ||
| /** | ||
| * Workaround until https://github.com/sveltejs/kit/pull/13007 is merged |
There was a problem hiding this comment.
lets merge the kit PR first then, really not a fan of extra code in tests to work around issues in the setup being tested. users don't have the same luxury so this is just hiding an issue.
would rather merge this with a failed test than hiding it if we can't get the kit PR released in time
dominikg
reviewed
Nov 25, 2024
| @@ -162,4 +162,11 @@ describe.runIf(!isBuild)('ssrLoadModule', () => { | |||
| const result = await ssrLoadDummy('?raw&svelte&type=style'); | |||
| await expect(result).toMatchFileSnapshot(snapshotFilename('ssr-style')); | |||
| }); | |||
| test('?inline&svelte&type=style&lang.css', async () => { | |||
| // Preload Dummy.svelte first so its CSS is processed in the module graph, otherwise loading | |||
There was a problem hiding this comment.
wait, wouldn't this also affect real use if the inline style is explicitly asked for without the .svelte being processed first? The solution here would be to compile the file if cache is empty instead of just bailing.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
In Vite 6, importing CSS in SSR now no longer return a default export of the transformed CSS file, you'd need to use
?inlineinstead. This was already pushed for in Vite 5 but we forgot to remove support in SSR.That means frameworks who do CSS crawling to prevent FOUC will now have to use
?inlineto get the transformed CSS, which most already did. However, there's a quirk where v-p-s always strips out the?inlinehere:vite-plugin-svelte/packages/vite-plugin-svelte/src/index.js
Lines 127 to 136 in da54670
This was fine for Vite 5 as mentioned above, Vite still supported default exports of CSS files in SSR, but with Vite 6 it is now empty. For v-p-s to properly support
?inline, it needs to leave the query as is, similar to?rawhandling.In practice, this isn't a breaking change, however due to some unintended reliance of the old behaviour,
the SvelteKit tests here will fail due to HMR(added a workaround here for now). I sent a PR upstream to fix this sveltejs/kit#13007 and if we want to play it safe for now, we can merge this only in the next major.With this merged, it should unblock Vite 6 support for CSS FOUC for both SvelteKit and Astro.