Realy cool idea. Vite is amazing tool. How long might it take to get this into production?

Realy cool idea. Vite is amazing tool. How long might it take to get this into production?

Thanks, This branch is simply a playground to see if it would be easy to refactor the codebase to use Vite instead of Parcel. I don't know much about Vite, and so if you'd like to help please let me know!

Would love this as well. Run into segmentation faults many times

Have you considered looking into Bun as a compiler too? It has recently introduced React HMR and is really fast in many ways.

I already use it to compile many functions and main content scripts that Plasmo just copy-pastes into the output. I do this since at some point in the past Plasmo began lagging a lot, slow compiles, and out-dated changes (one of those stale open issues with no clear reason as to why).

That being said, thousand times thank you for doing this! Maybe we could also draw inspiration fromhttps://github.com/wxt-dev/wxt. Recently read their entire documentation and there are a lot of nice things. Or maybe we could join forces and merge Plasmo into WXT (easier migration) and help build one universal tool to hide the pain away that browsers provide.

My two cents.

Hi@flexchar, thanks for your comment.

Have you considered looking into Bun as a compiler too?

I have not, but it's definitely an option. Will look into it

I do this since at some point in the past Plasmo began lagging a lot

Is your project quite large? Either way, I think moving from Parcel to Vite will hopefully improve performance enough for it not to matter. We will see.

That being said, thousand times thank you for doing this!

You are welcome. Please do note that this is simply a playground as I've mentioned before - I have no prior experience with Vite, and so it's all a learning experience for me. I'm also a relatively new contributor to Plasmo and so am learning the codebase too.

Help is welcomed!