Is the npm package version published? I don’t see it on npmjs.com and `npm install -g stimulus-language-server` still installs the old version, and when I specify @1.1.0 npm says the version doesn’t exist 🤔
I just released a new version of the Stimulus LSP that is now fully powered by the power and intelligence of the Herb toolchain! 🌿
I'm so happy that this finally came full circle. These features are initially what prompted me to start working on Herb!
🚀 We just shipped v.0.231!
Git graph is finally here: Open it from the git panel or the command palette (git graph: open).
Glad that there is a way to turn this off 🙌
This is part of my March update. I also got accepted into Anthropic's Claude for Open Source program, and welcomed a new GitHub sponsor! Both of which I deeply appreciate 🙏
More RDoc improvements (RBS signatures, coverage) are coming too.
st0012.dev/updates/marc...
First Ruby LSP beta got released with partial migration to Rubydex 👀
github.com/Shopify/ruby...
Indexing our Rails application (+ dependencies) currently takes 12s with Ruby LSP's internal indexer.
Rubydex does it in 2s 💯 And I think it's even doing more work.
I joined RubyCentral to release a postmortem, and today I'm delivering my report on what happened. The hope is to provide more transparency and closure, 194 days since the incident on September 18.
I've named the incident "RubyGems Fracture." Read my report. #ruby
rubycentral.org/news/rubygem...
I just released Herb v0.9.3 🌿
Another big release with a lot of improvements and more stabilization work in the parser and the engine.
Also, a lot of new linter features and a ton of new linter rules, thanks to @markokajzer.bsky.social!
The only thing I’m missing after moving from Twitter to Bluesky are @tailwindcss.com news.
I used to get them from @adamwathan.com, but it appears he’s not active here (gentle nudge).
Otherwise I’m perfectly content 🙂
> We recognize that this creates continued uncertainty.
There was never any risk that Andre would jeopardize RubyGems.org infrastructure. He maintained it and has too much trust built up.
Yeah, we already deprecated it in the team. But it's impossible to actually remove it without a consistent effort of eradication.
When a ticket requires changes on a page, not many people will take time to first replace Bootstrap on that page. So, it needs to be a separate effort.
Zeta2 is here. 30% better acceptance rate than Zeta1. 200x more training data, LSP-powered context, faster predictions, open weights. Try it now in Zed.
We didn't just improve the model. We rebuilt the entire data pipeline behind it: zed.dev/blog/zeta2
Here is an excerpt of some lint offenses👇
This will allow us to gradually narrow down our Bootstrap usage until we're ready to remove it completely.
I don't know how we'd do it without Herb, thanks @marcoroth.dev!
As part of our migration from Slim to HTML+ERB at work, we're also replacing our legacy Bootstrap 3 usage with Tailwind/modern CSS.
To make it official, I vibe-coded a custom Herb linter rule that disallows Bootstrap 3 usage – gist.github.com/janko/e0e106...
Like, we actually have tooling now 🙌
I finally made it, I'm running 3 Claude Code instances in parallel with cctop by @st0012.dev to monitor them.
github.com/st0012/cctop
The context switching is killing me. But I'm officially an AI developer.
Apologies, it seemed to me like it wasn't working in Zed, that new symbols weren't being picked up by go-to-defintion and workspace symbol search. But either I used it wrong or it got fixed. All good!
@vinistock.bsky.social Do you think Rubydex's speed will allow Ruby LSP to incrementally update indexed constants/methods/etc as ruby files change after language server initialization?
Not sure if there is an open issue already, but it would be great to have symbols always up-to-date.
I’d be curious to hear your thoughts. I also wondered whether Stimulus could be different after learning web components.
The thing that drives me crazy the most is the controller name repetition. Also would like if Stimulus outlets were more polished.
Yesss ✨
Herb 🤝 Stimulus LSP
Herb v0.9 is here! 🌿
This release focused on stability, compatibility, and deepening Herb's understanding of HTML+ERB templates.
Featuring Action View helper support, Prism AST nodes, new linter rules, a faster linter CLI, new language server features, and engine bug fixes. 🚀
What do you like about it over npm?
Seems to me like npm has improved a lot since Yarn was first created.
I also can't believe we're still using Yarn v1 in Rails when there is v4.
I get that Yarn has grown significantly in complexity starting from Berry, but v1 is now effectively unmaintained.
I'm considering switching to npm, so that we use the defaults. I think that would've avoided this problem.
After trying to upgrade my Rails app to Vite 8, I've grown new appreciation Bundler/RubyGems allowing only one gem version to be active in a project.
I had to put vite-plugin-ruby package to "resolutions" field in package.json to force it to upgrade, since it was a dependency of vite-plugin-rails.
⚡️ Vite 8.0 is here!
The most significant architectural change since Vite 2.
⏬ Powered by @rolldown.rs bringing faster production builds and more consistency
🛤️ New features such as tsconfig paths and emitDecoratorMetadata support
vite.dev/blog/announc...
I'm pretty sure this can be simplified. For instance, why not use Dry::Params::Nil and Dry::Coercible::String instead? Or use `Dry::Params::Integer.optional` to handle nils?
Some tips and tricks for working with #ClaudeCode batsov.com/articles/202...
Wait, do they not support not supporting it, or they don’t support supporting it?
In other words, does the @supports block not get gracefully skipped when @scope is not supported, or does it get incorrectly skipped when @scope is supported?
