# The Question List

Covering 2026-05-30 to 2026-08-28
364 threads read this period, 16 of them new. Threads are counted per distinct account. Bots and pull requests are excluded.

## New this week

### 13 people: Can async derived values be fixed to avoid infinite loops and reactivity issues? (new) - svelte

> With compilerOptions.experimental.async enabled, if state is invalidated while an async derived is still pending (so more than one Batch is alive), $derived values lose their …
> - [$derived re-executes once per reading effect (memoization lost) when invalidated while another batch is pending (experimental.async)](https://github.com/sveltejs/svelte/issues/18503), 2026-07-04

> Human note-- the following was created by Fable after many hours of looping Opus 4.8 over the same issue we have had for over a …
> - [Memory leak: a $derived only read in untracked contexts (e.g. a $state initializer) is never disconnected from its dependencies on destroy](https://github.com/sveltejs/svelte/issues/18501), 2026-07-03

> Using `{let foo = $derived()}` can, in at least some cases, result in `foo` being exposed as a _function_ that returns the appropriate value, rather …
> - [SSR compilation of `{let ...= $derived(...)}` can incorrectly use getter instead of actual value](https://github.com/sveltejs/svelte/issues/18490), 2026-06-30

### 6 people: Can the compiler preserve parentheses in logical expressions? (new) - svelte

> When using rune mode in a `.svelte` or `.svelte.ts` file, the `!==` binary operator is incorrectly compiled so that the resulting `$.strict_equals()` call receives the …
> - [\[bug\] Rune mode compiles `!==` to $.strict_equals(a, b, false) — inverted comparison](https://github.com/sveltejs/svelte/issues/18439), 2026-06-18

> The Svelte 5 compiler strips parentheses from boolean expressions, changing operator precedence and producing incorrect runtime behaviour. Source: Compiled output: Which evalu …
> - [Compiler drops parentheses in boolean expressions, changing operator precedence](https://github.com/sveltejs/svelte/issues/18386), 2026-06-05

> When the Svelte compiler re-prints a JavaScript logical expression that contains a **parenthesized `||` (or `&&`) nested on the side that needs the parentheses**, it …
> - [Compiler drops precedence-significant parentheses in logical expressions (`a && (b || c)` → `a && b || c`)](https://github.com/sveltejs/svelte/issues/18380), 2026-06-04

### 5 people: Can declaration tags support multiple declarators and fix binding issues? (new) - svelte

> Consider the following Child component Consider the Main svelte file Throws the following error `Can't find variable: labelKey in \<unknown> in Child.svelte in App.svelte in …
> - [Keyed {#each} blocks fail when using dynamic keys for destructuring](https://github.com/sveltejs/svelte/issues/18519), 2026-07-09

> When you attempt to shadow a known variable in `{const variable = ...}` declaration tag it errors. \[REPL\](https://svelte.dev/playground/hello-world?version=5.56.4#H4sIAAAAAAAA …
> - [Shadowing does not seem to work for declaration tags](https://github.com/sveltejs/svelte/issues/18516), 2026-07-09

> const} is considered legacy and {const} should be used as replacement: * https://svelte.dev/docs/svelte/@const * https://sveltejs.github.io/eslint-plugin-svelte/rules/no-at-co …
> - [{const and {let are not updating as expected](https://github.com/sveltejs/svelte/issues/18514), 2026-07-07

### 5 people: Can the compiler support TypeScript and improve type handling? (new) - svelte

> Svelte 5.35.0+ appears to preserve the TypeScript optional parameter marker (`?`) in generated JavaScript for optional function parameters inside `\<script lang="ts">`. Valid S …
> - [Svelte 5.35+ leaves TypeScript optional parameter marker in compiled JavaScript](https://github.com/sveltejs/svelte/issues/18455), 2026-06-23

> When exporting snippets referencing the type of a value from the regular `\<script>` block, the TypeScript error: is thrown instead of the expected error: > …
> - [TypeScript error thrown when exported snippets reference a type of a script element](https://github.com/sveltejs/svelte/issues/18451), 2026-06-21

> If the vscode setting `js/ts.inlayHints.parameterNames.enabled` is set to `all` and there are attributes before an `{@attach ...}` on an element, the hints gets displayed insi …
> - [JS/TS Inlay hints are broken with {@attach ...}](https://github.com/sveltejs/svelte/issues/18412), 2026-06-10

### 4 people: How can I check progress on a reported issue? (new) - svelte

> Thank you, but I'm closing this for the reasons outlined in https://github.com/sveltejs/svelte/issues/18600#issuecomment-5361606575
> - [source](https://github.com/sveltejs/svelte/pull/18653#issuecomment-5361610314), 2026-08-20

> I really hope this will be merged in a new patch
> - [source](https://github.com/sveltejs/svelte/pull/18614#issuecomment-5315637058), 2026-08-17

> We have also been continually running into the freezing issue, i've tested it against this PR and it seems to disappear completely. @dummdidumm what's the …
> - [source](https://github.com/sveltejs/svelte/pull/18558#issuecomment-5290404913), 2026-08-14

### 4 people: Can I fix whitespace behavior attributed to Bun instead of Svelte? (new) - svelte

> The compiler printer drops a meaningful whitespace text node after an inline element. This changes the rendered text from `Hello bold world` to `Hello boldworld`. …
> - [print() drops whitespace between inline elements](https://github.com/sveltejs/svelte/issues/18683), 2026-08-21

> thank you, can this be expanded to fix text-node whitespace similar issue where result drops all white spaces. svelte version: 5.56.10 INPUT : `"\<p>Hello \< …
> - [source](https://github.com/sveltejs/svelte/pull/18667#issuecomment-5361776112), 2026-08-20

> > This works as designed - if you don't want the whitespace, don't add it. Since this is ultimately a Bun issue it should be …
> - [source](https://github.com/sveltejs/svelte/issues/18659#issuecomment-5333796909), 2026-08-18

### 2 people: Can input defaultvalue be populated during server side rendering? (new) - svelte

> `defaultValue` doesn't work in SSR for inputs too so I guess we should just add CSR for `\<textarea value="hi">\</textarea>` no?
> - [source](https://github.com/sveltejs/svelte/issues/18574#issuecomment-5056327961), 2026-07-23

> I expect `\<input defaultValue="foo" />` to SSR as `\<input value="foo"/>`, but it instead renders as `\<input/>`. This means on pages without JS, the default …
> - [`\<input defaultValue="foo" />` doesn't populate until hydrated](https://github.com/sveltejs/svelte/issues/18446), 2026-06-20

### 2 people: Can svelte transitions work without inline styles under csp? (new) - svelte

> Yes, this is outdated, this will reflect in the upcoming SvelteKit 3 docs. Feel free to use svelte transitions with Svelte 5 + CSP!
> - [source](https://github.com/sveltejs/svelte/issues/18547#issuecomment-5000560105), 2026-07-17

> there is a note in https://svelte.dev/docs/kit/configuration#csp about usage of svelte transitions: > Note that most Svelte transitions work by creating an …
> - [transitions and inline css](https://github.com/sveltejs/svelte/issues/18547), 2026-07-14

### 2 people: Why is regex-based preprocessor parsing unreliable? (new) - svelte

> That's fair. I understand not wanting to have the regex turn into something unmaintainable. I did some basic searching and could not find any related …
> - [source](https://github.com/sveltejs/svelte/pull/18595#issuecomment-5107667564), 2026-07-28

> Previously, we've said that it is components' responsibility to not contain code that's going to confuse the dumb regex-based parsing of preprocessors if the …
> - [source](https://github.com/sveltejs/svelte/pull/18595#issuecomment-5107505558), 2026-07-28

### 2 people: How can I merge the rsvelte compiler with sveltejs/svelte? (new) - svelte

> Thanks! On joining forces — I am for it, and I think there is a concrete shape for it. Merging the two compilers is not …
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5114820616), 2026-07-29

> @MrWaip great work! congrats! I don't know the details, but I wonder if it might make sense to joinj forces with rsvelte, both projects seem …
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5113058055), 2026-07-29

## svelte

25 standing asks

### 13 people: Can async derived values be fixed to avoid infinite loops and reactivity issues? (new)

> With compilerOptions.experimental.async enabled, if state is invalidated while an async derived is still pending (so more than one Batch is alive), $derived values lose their …
> - [$derived re-executes once per reading effect (memoization lost) when invalidated while another batch is pending (experimental.async)](https://github.com/sveltejs/svelte/issues/18503), 2026-07-04

> Human note-- the following was created by Fable after many hours of looping Opus 4.8 over the same issue we have had for over a …
> - [Memory leak: a $derived only read in untracked contexts (e.g. a $state initializer) is never disconnected from its dependencies on destroy](https://github.com/sveltejs/svelte/issues/18501), 2026-07-03

> Using `{let foo = $derived()}` can, in at least some cases, result in `foo` being exposed as a _function_ that returns the appropriate value, rather …
> - [SSR compilation of `{let ...= $derived(...)}` can incorrectly use getter instead of actual value](https://github.com/sveltejs/svelte/issues/18490), 2026-06-30

### 6 people: Can the compiler preserve parentheses in logical expressions? (new)

> When using rune mode in a `.svelte` or `.svelte.ts` file, the `!==` binary operator is incorrectly compiled so that the resulting `$.strict_equals()` call receives the …
> - [\[bug\] Rune mode compiles `!==` to $.strict_equals(a, b, false) — inverted comparison](https://github.com/sveltejs/svelte/issues/18439), 2026-06-18

> The Svelte 5 compiler strips parentheses from boolean expressions, changing operator precedence and producing incorrect runtime behaviour. Source: Compiled output: Which evalu …
> - [Compiler drops parentheses in boolean expressions, changing operator precedence](https://github.com/sveltejs/svelte/issues/18386), 2026-06-05

> When the Svelte compiler re-prints a JavaScript logical expression that contains a **parenthesized `||` (or `&&`) nested on the side that needs the parentheses**, it …
> - [Compiler drops precedence-significant parentheses in logical expressions (`a && (b || c)` → `a && b || c`)](https://github.com/sveltejs/svelte/issues/18380), 2026-06-04

### 5 people: Can declaration tags support multiple declarators and fix binding issues? (new)

> Consider the following Child component Consider the Main svelte file Throws the following error `Can't find variable: labelKey in \<unknown> in Child.svelte in App.svelte in …
> - [Keyed {#each} blocks fail when using dynamic keys for destructuring](https://github.com/sveltejs/svelte/issues/18519), 2026-07-09

> When you attempt to shadow a known variable in `{const variable = ...}` declaration tag it errors. \[REPL\](https://svelte.dev/playground/hello-world?version=5.56.4#H4sIAAAAAAAA …
> - [Shadowing does not seem to work for declaration tags](https://github.com/sveltejs/svelte/issues/18516), 2026-07-09

> const} is considered legacy and {const} should be used as replacement: * https://svelte.dev/docs/svelte/@const * https://sveltejs.github.io/eslint-plugin-svelte/rules/no-at-co …
> - [{const and {let are not updating as expected](https://github.com/sveltejs/svelte/issues/18514), 2026-07-07

### 5 people: Can the compiler support TypeScript and improve type handling? (new)

> Svelte 5.35.0+ appears to preserve the TypeScript optional parameter marker (`?`) in generated JavaScript for optional function parameters inside `\<script lang="ts">`. Valid S …
> - [Svelte 5.35+ leaves TypeScript optional parameter marker in compiled JavaScript](https://github.com/sveltejs/svelte/issues/18455), 2026-06-23

> When exporting snippets referencing the type of a value from the regular `\<script>` block, the TypeScript error: is thrown instead of the expected error: > …
> - [TypeScript error thrown when exported snippets reference a type of a script element](https://github.com/sveltejs/svelte/issues/18451), 2026-06-21

> If the vscode setting `js/ts.inlayHints.parameterNames.enabled` is set to `all` and there are attributes before an `{@attach ...}` on an element, the hints gets displayed insi …
> - [JS/TS Inlay hints are broken with {@attach ...}](https://github.com/sveltejs/svelte/issues/18412), 2026-06-10

### 4 people: How can I check progress on a reported issue? (new)

> Thank you, but I'm closing this for the reasons outlined in https://github.com/sveltejs/svelte/issues/18600#issuecomment-5361606575
> - [source](https://github.com/sveltejs/svelte/pull/18653#issuecomment-5361610314), 2026-08-20

> I really hope this will be merged in a new patch
> - [source](https://github.com/sveltejs/svelte/pull/18614#issuecomment-5315637058), 2026-08-17

> We have also been continually running into the freezing issue, i've tested it against this PR and it seems to disappear completely. @dummdidumm what's the …
> - [source](https://github.com/sveltejs/svelte/pull/18558#issuecomment-5290404913), 2026-08-14

### 4 people: Can I fix whitespace behavior attributed to Bun instead of Svelte? (new)

> The compiler printer drops a meaningful whitespace text node after an inline element. This changes the rendered text from `Hello bold world` to `Hello boldworld`. …
> - [print() drops whitespace between inline elements](https://github.com/sveltejs/svelte/issues/18683), 2026-08-21

> thank you, can this be expanded to fix text-node whitespace similar issue where result drops all white spaces. svelte version: 5.56.10 INPUT : `"\<p>Hello \< …
> - [source](https://github.com/sveltejs/svelte/pull/18667#issuecomment-5361776112), 2026-08-20

> > This works as designed - if you don't want the whitespace, don't add it. Since this is ultimately a Bun issue it should be …
> - [source](https://github.com/sveltejs/svelte/issues/18659#issuecomment-5333796909), 2026-08-18

### 2 people: Can input defaultvalue be populated during server side rendering? (new)

> `defaultValue` doesn't work in SSR for inputs too so I guess we should just add CSR for `\<textarea value="hi">\</textarea>` no?
> - [source](https://github.com/sveltejs/svelte/issues/18574#issuecomment-5056327961), 2026-07-23

> I expect `\<input defaultValue="foo" />` to SSR as `\<input value="foo"/>`, but it instead renders as `\<input/>`. This means on pages without JS, the default …
> - [`\<input defaultValue="foo" />` doesn't populate until hydrated](https://github.com/sveltejs/svelte/issues/18446), 2026-06-20

### 2 people: Can svelte transitions work without inline styles under csp? (new)

> Yes, this is outdated, this will reflect in the upcoming SvelteKit 3 docs. Feel free to use svelte transitions with Svelte 5 + CSP!
> - [source](https://github.com/sveltejs/svelte/issues/18547#issuecomment-5000560105), 2026-07-17

> there is a note in https://svelte.dev/docs/kit/configuration#csp about usage of svelte transitions: > Note that most Svelte transitions work by creating an …
> - [transitions and inline css](https://github.com/sveltejs/svelte/issues/18547), 2026-07-14

### 2 people: Why is regex-based preprocessor parsing unreliable? (new)

> That's fair. I understand not wanting to have the regex turn into something unmaintainable. I did some basic searching and could not find any related …
> - [source](https://github.com/sveltejs/svelte/pull/18595#issuecomment-5107667564), 2026-07-28

> Previously, we've said that it is components' responsibility to not contain code that's going to confuse the dumb regex-based parsing of preprocessors if the …
> - [source](https://github.com/sveltejs/svelte/pull/18595#issuecomment-5107505558), 2026-07-28

### 2 people: How can I merge the rsvelte compiler with sveltejs/svelte? (new)

> Thanks! On joining forces — I am for it, and I think there is a concrete shape for it. Merging the two compilers is not …
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5114820616), 2026-07-29

> @MrWaip great work! congrats! I don't know the details, but I wonder if it might make sense to joinj forces with rsvelte, both projects seem …
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5113058055), 2026-07-29

### 2 people: How can I clarify the official status of the rsvelte project? (new)

> Yes, you're right. But as I've understood `rsvelte` is not an official project, no?
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5177234604), 2026-08-04

> > All these projects are cool, but rely heavily on AI written code. I don't think it would be a good idea for the Svelte …
> - [source](https://github.com/sveltejs/svelte/issues/18376#issuecomment-5177186272), 2026-08-04

### 2 people: Why does onDestroy crash on server-side rendering? (new)

> This is expected and documented - onDestroy indeed does run on the server precisely so that you can cleanup things on the server, too. https://svelte.dev/ …
> - [source](https://github.com/sveltejs/svelte/issues/18630#issuecomment-5226587292), 2026-08-08

> When you use onDestroy function inside +page.svelte or +layout.svelte, the callback runs on the server and causes crash. While at the same time, a callback …
> - [onDestroy runs on server and causes crash](https://github.com/sveltejs/svelte/issues/18630), 2026-08-08

### 2 people: Can I fix the print function dropping CSS escape sequences? (new)

> print()` drops CSS escape sequences from selectors, so a stylesheet that `parse()` accepts can come back either invalid or meaning something different. Using the repo's …
> - [\[this was a bot test to see if it opens PRs in response to issues that have a clear reproduction\]](https://github.com/sveltejs/svelte/issues/18671), 2026-08-19

> print()` drops CSS escape sequences from selectors, so a stylesheet that `parse()` accepts can come back either invalid or meaning something different. Using the repo's …
> - [print() drops CSS escape sequences, producing invalid or different selectors](https://github.com/sveltejs/svelte/issues/18664), 2026-08-18

### 1 person: How can I make SvelteKit layout patterns more flexible? (new)

> your compiler(svelte) quite good, but your pattern in svelteKit is really bad, you push someone to use your rules like slug that maybe they …
> - [source](https://github.com/sveltejs/svelte/issues/18632#issuecomment-5231604409), 2026-08-09

### 1 person: Can I make const tag sorting unnecessary for small lists? (new)

> Completely unnecessary, the average number of const tags to sort is ~2, I would be surprised if it's more than 10 anywhere.
> - [source](https://github.com/sveltejs/svelte/issues/18599#issuecomment-5313249890), 2026-08-17

### 1 person: Can hydration mismatches with async derived and bind be fixed? (new)

> Please see the attached minimal example project. Reproduces reliably. It's just 24 lines. Thanks!! \[svelte hydration bug.zip\](https://github.com/user-attachments/files/2868898 …
> - [Hydration mismatch with two components that internally use bind: and svelte:head (getAttribute is not a function / can't access property "nodeType")"](https://github.com/sveltejs/svelte/issues/18404), 2026-06-07

### 1 person: Can the compiler avoid crashing on sibling elements after select? (new)

> throws `can't access property "type", e.object is null` inside esrap. https://svelte.dev/playground/hello-world?version=5.56.3#H4sIAAAAAAAAE23MQQrCMBAF0KuEWSmI7kMV3HkH00VsfiAw …
> - [Any sibiling after a `\<select>` crashes the compiler](https://github.com/sveltejs/svelte/issues/18445), 2026-06-20

### 1 person: Can the compiler warn on invalid declaration usage inside if blocks? (new)

> Hey! I was playing with the new declarations. Not sure if I'm missing something obvious, but the following case doesn't seem to produce any warning …
> - [declaration tags: if with wrong declaration](https://github.com/sveltejs/svelte/issues/18343), 2026-05-31

### 1 person: Can bind:value be fixed to not reset selection on custom select? (new)

> When a `\<select>` uses the customizable-select API (`appearance: base-select`) and contains the author trigger `\<button>\<selectedcontent>\</selectedcontent>\</button>`, adding ` …
> - [`bind:value` on a customizable `\<select>` with a `\<button>`/`\<selectedcontent>` trigger resets the selection to `""`](https://github.com/sveltejs/svelte/issues/18347), 2026-05-31

### 1 person: Can infinite loops with const ending with bare slash be prevented? (new)

> I found this issue while working on https://github.com/sveltejs/svelte/pull/18330. This hang can be particularly problematic when it occurs in the language …
> - [hangs (infinite loop) on a `{const}` / `{let}` whose expression ends with a bare `/` at end of input](https://github.com/sveltejs/svelte/issues/18349), 2026-06-01

### 1 person: Can the documentation clarify that bare const is not reactive like @const? (new)

> The \[declaration-tags docs\](https://svelte.dev/docs/svelte/declaration-tags) say `{@const ...}` is legacy and to use `{const ...}` instead, but don't mention that the two …
> - [Docs: clarify that bare `{const}` is not reactive, unlike the `{@const}` it replaces](https://github.com/sveltejs/svelte/issues/18356), 2026-06-01

### 1 person: Can batch roots error when sibling mutates shared state be fixed? (new)

> Full disclosure: the debugging here was largely done by Claude, reviewed by me (an experienced backend SWE who doesn't know Svelte internals deeply). The bug …
> - [experimental.async: "Batch has scheduled roots" when a sibling mutates shared $state during a suspense-resolution flush](https://github.com/sveltejs/svelte/issues/18363), 2026-06-02

### 1 person: Can $state.eager behavior in if blocks be clarified? (new)

> In \[this playground\](https://svelte.dev/playground/hello-world?version=5.56.1#H4sIAAAAAAAACm2QMW-EMAyF_0rO7QASOriOCJC6dezQremQA1OiBgclprSK-O9Vgk5dbrGlT8_vPTkAqRmhhhc0xorNOjOID …
> - [`$state.eager` - correct behavior or not?](https://github.com/sveltejs/svelte/issues/18364), 2026-06-02

### 1 person: Can false error when using animate with new const be fixed? (new)

> By using the new `{const}` thing, and having an item with the animate directive rigt below it, we get the error: "An element that uses …
> - [New const declaration shows error when using the animate directive inside each block](https://github.com/sveltejs/svelte/issues/18365), 2026-06-02

### 1 person: Can effects executing in wrong batch causing wrong values be fixed? (new)

> The fix I did in revealed a separate bug: An effect scheduled for batch A can be executed by batch B, causing wrong values. This …
> - [Effect scheduled for batch A can be executed by batch B, causing wrong values](https://github.com/sveltejs/svelte/issues/18374), 2026-06-03

## The numbers behind this report

Every line above counts DISTINCT people, never posts. Each is linked to the public page it came from, so you can check any of them in one click.

- New this period: 16
- Read this period: 364
- Analysed: 364
- Maintainer posts excluded: 4

Source: GitHub. 364 threads analysed over 90 days. 4 posts by the project's own maintainers are excluded from every count.