dice.camp is one of the many independent Mastodon servers you can use to participate in the fediverse.
A Mastodon server for RPG folks to hang out and talk. Not owned by a billionaire.

Administered by:

Server stats:

1.8K
active users

#svelte

1 post1 participant0 posts today
Replied in thread

@itamarst one of the nice things is that I was touching this just now because of a weird bug¹ with Svelte and a state management library I was using, that was only affecting integration testing but started affecting the app itself when I added a schema validation library². So I removed the state management library and now I can also get on writing those tests!

¹) I submitted a bug report like 3 weeks ago but no response so far
²) dependency hell is my least favorite part of working in js/ts. I remember this from when I was learning React

🚨 Nuevo en @civio
2ª pieza de nuestra serie sobre #algoritmos en la #administraciónPública. Y nuestro tema más surrealista en mucho tiempo:

🤖 Hablamos de Veripol, la hasta hace poco ⭐️ IA estrella de la Policía Nacional ⭐️ para detectar denuncias falsas.

Y, para que entiendas mejor cómo funcionaba, hemos diseñado un pequeño juego interactivo, que te reta a adivinar el criterio de veracidad/falsedad de Veripol 👇

👩🏻‍💻👨🏻‍💻 Mano a mano con Ter, Adrián y @dcabo #dataviz #svelte

civio.es/transparencia/2025/03

🔥 Cool Svelte 5 feature: Function bindings!

You can now transform values before displaying/saving them in inputs. Perfect for unit conversions, formatting, etc.

Example: USD input that automatically stores EUR in memory:
```svelte
<input bind:value={
() => usd2eur(value),
(v) => value = eur2usd(v)
}>
```

The input shows USD but the state stays in EUR.

Complete example here : svelte.dev/playground/93ce69fd

svelte.devbind-func-example • Playground • Svelte
Continued thread

Another anecdote. We had an internal CRM app at work, built in 2021 with the backend / API / frontend architecture.

Nest on the backend. TypeORM broke several times, preventing smooth upgrade. Nest was stable enough though.

The frontend was built using #Svelte and Sapper. OK, Sapper was a beta at the time. Replaced with SvelteKit in 2022. We tried to migrate. Twice. Resources were limited: the app was useful to us, but not paying our salaries. Migration would taken *weeks.*

I think if I started #Svelte for the first time, today, I'd just go back to Vue or something and I'd never have become a fan.

I _understand_ that this is a 'skill issue' and that it's all my fault for not yet fully understanding the new way of doing things.

But dude, it was dead simple before. The learning curve was nearly a flat line. Shit 'just worked'. You're telling me performance is way better now with Svelte 5, but it never felt like an issue previously.