ComputerGuy
@computer-guy.bsky.social
Definitely, I'm working on a project that would very much benefit from this. Currently I'm just waiting for Import Bytes to be standardized to use that.
September 12, 2025 at 7:21 PM
Definitely, I'm working on a project that would very much benefit from this. Currently I'm just waiting for Import Bytes to be standardized to use that.
Interesting! So sources have a reactions array, while effects don't know their dependencies? Or is it the other way around?
August 29, 2025 at 7:45 PM
Interesting! So sources have a reactions array, while effects don't know their dependencies? Or is it the other way around?
This is really cool! Looking at the source code, I can definitely see how Svelte influenced it 😄
You said in your other post that it doesn't use signals, what does it use? The code looks very signal-esque...
You said in your other post that it doesn't use signals, what does it use? The code looks very signal-esque...
I've had so many folks ask now, that I had to go for it.
You asked me to build that ideal Inferno X framework that I described ages ago, so I went ahead and did just that. I built something that actually works and, well, it works well. Oh, and doesn't use signals!
You asked me to build that ideal Inferno X framework that I described ages ago, so I went ahead and did just that. I built something that actually works and, well, it works well. Oh, and doesn't use signals!
August 29, 2025 at 3:49 AM
This is really cool! Looking at the source code, I can definitely see how Svelte influenced it 😄
You said in your other post that it doesn't use signals, what does it use? The code looks very signal-esque...
You said in your other post that it doesn't use signals, what does it use? The code looks very signal-esque...
That's why migration scripts exist. A while ago, I wrote a component in Svelte 1 and tried to migrate it to Svelte 5 using migration scripts provided by Svelte; the only diff between what I expected and what resulted was state was declared via $props, bc state and props were the same before v3.
June 26, 2025 at 6:06 PM
That's why migration scripts exist. A while ago, I wrote a component in Svelte 1 and tried to migrate it to Svelte 5 using migration scripts provided by Svelte; the only diff between what I expected and what resulted was state was declared via $props, bc state and props were the same before v3.
You could try using a Portal: github.com/romkor/svelt...
GitHub - romkor/svelte-portal: Svelte component for rendering outside the DOM of parent component
Svelte component for rendering outside the DOM of parent component - romkor/svelte-portal
github.com
June 23, 2025 at 6:29 PM
You could try using a Portal: github.com/romkor/svelt...
I'm not saying React is bad by any means. All I'm stating is how different React really is from pretty much every other framework, and showing how much it has changed over the past decade, both in its features and goals.
June 11, 2025 at 8:46 PM
I'm not saying React is bad by any means. All I'm stating is how different React really is from pretty much every other framework, and showing how much it has changed over the past decade, both in its features and goals.
This is also why Signals don't really work in React and React-adjacent frameworks. Fine-grained reactivity doesn't make sense when the component tree gets recreated on every state change.
June 11, 2025 at 8:46 PM
This is also why Signals don't really work in React and React-adjacent frameworks. Fine-grained reactivity doesn't make sense when the component tree gets recreated on every state change.
React uses functions in a more mathematical sense rather than a software sense— functions represent the state of your application at any point in time, and regular optimizations don't account for this. The internal mechanisms of hooks would be infeasible in almost any other framework.
June 11, 2025 at 8:46 PM
React uses functions in a more mathematical sense rather than a software sense— functions represent the state of your application at any point in time, and regular optimizations don't account for this. The internal mechanisms of hooks would be infeasible in almost any other framework.
Not to mention that the reason the React Compiler needs to exist is simply because of all the new concepts and abstractions React provides. Regular optimizations wouldn't have the same effect on React code since it doesn't make use of JS like most frameworks.
June 11, 2025 at 8:46 PM
Not to mention that the reason the React Compiler needs to exist is simply because of all the new concepts and abstractions React provides. Regular optimizations wouldn't have the same effect on React code since it doesn't make use of JS like most frameworks.
Things like Hooks and Suspense have their own vocabulary and concepts. Additionally, React tries pretty hard to make it feel like you're using the DOM, with concepts like JSX, when in actuality you are using a thick layer of abstraction between what you're writing and what appears in the DOM.
June 11, 2025 at 8:46 PM
Things like Hooks and Suspense have their own vocabulary and concepts. Additionally, React tries pretty hard to make it feel like you're using the DOM, with concepts like JSX, when in actuality you are using a thick layer of abstraction between what you're writing and what appears in the DOM.
React originally tried to reduce the vocabulary needed to understand how to use it, but I feel that it has been moving away from that over the past decade. IIRC Pete Hunt mentioned all the vocabulary AngularJS had created/redefined to describe its concepts and implied that React would reduce this.
June 11, 2025 at 8:46 PM
React originally tried to reduce the vocabulary needed to understand how to use it, but I feel that it has been moving away from that over the past decade. IIRC Pete Hunt mentioned all the vocabulary AngularJS had created/redefined to describe its concepts and implied that React would reduce this.
Would this function ever rerun? Also, how do you differentiate between derived values and sources? (Not the exactly the same thing here, but I don't really know what else to call them)
May 29, 2025 at 8:12 PM
Would this function ever rerun? Also, how do you differentiate between derived values and sources? (Not the exactly the same thing here, but I don't really know what else to call them)
Didn't you make something that can do that for you?
github.com/Rich-Harris/...
github.com/Rich-Harris/...
GitHub - Rich-Harris/dts-buddy: WORK IN PROGRESS DO NOT USE
WORK IN PROGRESS DO NOT USE. Contribute to Rich-Harris/dts-buddy development by creating an account on GitHub.
github.com
May 20, 2025 at 10:44 PM
Didn't you make something that can do that for you?
github.com/Rich-Harris/...
github.com/Rich-Harris/...
I agree, not to mention the large number of devs who simply choose to avoid using AI. For example, I mostly rely on human-made tools like docs and Stack Overflow instead of AI. Declaring React as the "last framework" just because a tool used by some devs supports it the most is irrational.
May 20, 2025 at 3:21 AM
I agree, not to mention the large number of devs who simply choose to avoid using AI. For example, I mostly rely on human-made tools like docs and Stack Overflow instead of AI. Declaring React as the "last framework" just because a tool used by some devs supports it the most is irrational.
Maybe try adding "safari>=12" to your targets in your vite-legacy config? (According to MDN, Safari 13.1 added support)
I'm not entirely sure what sort of polyfills vite-legacy does though; if it just adds runtime polyfills then it won't work and you'd have to add something else (eg Babel or SWC)
I'm not entirely sure what sort of polyfills vite-legacy does though; if it just adds runtime polyfills then it won't work and you'd have to add something else (eg Babel or SWC)
May 18, 2025 at 6:15 PM
Maybe try adding "safari>=12" to your targets in your vite-legacy config? (According to MDN, Safari 13.1 added support)
I'm not entirely sure what sort of polyfills vite-legacy does though; if it just adds runtime polyfills then it won't work and you'd have to add something else (eg Babel or SWC)
I'm not entirely sure what sort of polyfills vite-legacy does though; if it just adds runtime polyfills then it won't work and you'd have to add something else (eg Babel or SWC)
If it's something about an unexpected token, the newish syntax features I can think of that Svelte uses are nullish coalescence and optional chaining. You could try finding a polyfill for that (probably using Babel or sth similar)
But more context is definitely needed here to be sure.
But more context is definitely needed here to be sure.
May 18, 2025 at 6:21 AM
If it's something about an unexpected token, the newish syntax features I can think of that Svelte uses are nullish coalescence and optional chaining. You could try finding a polyfill for that (probably using Babel or sth similar)
But more context is definitely needed here to be sure.
But more context is definitely needed here to be sure.