I really enjoyed this discussion with host Tim Chaten about the state of Apple Vision Pro. It was recorded a couple weeks after WWDC, which meant the memory was fresh enough to keep all of Apple's announcements top of mind but distant enough to imagine various directions things could go from here.

• Apple Podcasts
• Overcast

I gotta say, it was nice talking to someone who knows and cares more about the platform than I do. Some real "there are dozens of us!" energy around the Vision Pro right now.

Rails + Hotwire is very capable of dynamic behavior like replacing a component in the DOM by sending HTML over-the-wire in response to a user action, but the fact it requires you to touch half a dozen files in the process can make the whole thing feel daunting. Rails itself has always been this way (with each incremental feature requiring a route, model, controller, view, etc.), but I've been using it long enough that I sometimes forget that—similar to learning a recipe—I originally needed months of intentional practice to internalize and gain comfort with the framework's most routine of workflows.

So, like a recipe card, here is a reusable approach to swapping out a <div> rendered by a Rails partial with a turbo stream whenever a user selects an alternate model from an input (specifically, a select) using Turbo 8, Stimulus 1.3, and Rails 7.1.

The ingredients

1. Partial: Extract a partial to be rendered inside the element you wish to replace, so that both your view and your turbo stream can render the same markup for a given model
2. Routing: Add a route specifying a one-off controller action that will respond with a turbo stream
3. Controller Action: Define an action that takes your model ID and the DOM element's ID and responds with a turbo stream to update the element's contents
4. Turbo Stream View: Create a turbo stream view for the action that invokes the partial
5. Stimulus Controller: Create a generic Stimulus controller that can swap any model type when given a path, ID, and container
6. View: Wire up the Stimulus controller to the view's select box and the to-be-replaced element

That's it, 6 key ingredients. If you're curious, step 5 contains the most magic flavoring. 🪄

The actions

Ingredients in hand, let's walk through each of the steps needed to go end-to-end with this feature.

1. Set Up the Rails Partial

First, create a partial that you want to render inside the <div>. Let's assume we want users to be able to change out a generic model named Item, which has a conventional ItemsController.

In that case, let's place a partial that renders the details about an item alongside the controller's views, in _detail.html.erb:

<!-- app/views/items/_detail.html.erb -->
<div>
  <%= item.title %>
  <!-- Other item stuff… -->
</div>

2. Add Routes

Next, we'll add the necessary route for the detail action:

# config/routes.rb
resources :items do
  get :detail, on: :collection
end

This will define a path helper detail_items_path, which works out to "/items/detail".

Note that I threw this on the :collection so that our stimulus controller can more easily specify the URL via query parameters instead of interpolating a fancier member route (e.g. "items/42/detail").

3. Define the Controller Action

With the route defined, we'll add a simple controller action that only responds to turbo stream requests.

Here's what that might look like:

# app/controllers/items_controller.rb
class ItemsController < ApplicationController
  def detail
    @dom_id = params[:dom_id]
    @item = Item.find(params[:id])
  end
end

This dom_id param might throw you off at first, but it's important to keep in mind that unique HTML IDs are the coin of the realm in Turbo-land. You'll see how it gets set later, in step 5.

4. Create the Turbo Stream View

To finish the route-controller-view errand, we'll create a view for the detail action, with the turbo_stream.erb extension instead of html.erb:

<!-- app/views/items/detail.turbo_stream.erb -->
<%= turbo_stream.update @dom_id do %>
  <%= render partial: "detail", locals: { item: @item } %>
<% end %>

Because both the turbo stream and the original view need to render items in exactly the same way, detail.turbo_stream.erb view responds by rendering the _detail.html.erb partial. If you inspect the HTML that comes over the wire, you'll see that only the turbo stream tag containing this partial is transferred, which often means barely more data is transferred than had we implemented this as a single-page JavaScript by making a similar HTTP request for JSON.

5. Define the Stimulus Controller

In order for users' selections to have any effect, we need JavaScript. We could write a Stimulus controller that's coupled specifically to this Item model, but it's no more work to make it generic, which would allow us to reuse this functionality elsewhere in our app. So let's do that.

You can do this the hard way by using the browser's built-in fetch API to construct the URL, set the Accept header to text/vnd.turbo-stream.html, and replace the element's innerHTML in the DOM, but that's easy to screw up (in fact, I screwed it up twice while writing this). So instead, I'd recommend pulling in the requestjs-rails gem, by first chucking it in your Gemfile alongside any other front-end related gems:

gem "requestjs-rails"

Here's the final Stimulus controller. Deep breath, as I haven't explained all this yet:

// app/javascript/controllers/model_swap_controller.js
import { get } from '@rails/request.js'
import { Controller } from '@hotwired/stimulus'

export default class ModelSwapController extends Controller {
  static targets = ['container']
  static values = {
    path: String,
    originalId: String
  }

  swap (event) {
    const modelId = event.currentTarget.value || // Value from input action
      event.detail?.value || // Value from custom event (e.g. hotwire_combobox)
      this.originalIdValue // Fallback to original value if input value is cleared

    get(this.pathValue, {
      query: {
        id: modelId,
        dom_id: this.containerTarget.id
      },
      responseKind: 'turbo-stream'
    })
  }
}

That get function from @rails/request.js handles all the housekeeping you might hope it would. When I switched to it, the fact it worked the instant I plopped it in gave me Dem Magic Vibes that keep me coming back to Rails 18 years in.

This controller also contains two values and a target:

• path value: this is just a URL, which we'll set to our intentionally-parameter-free detail_items_path
• originalId value: this is the Item ID that was first rendered when the page loaded. By having this available as a fallback, we'll be able to gracefully handle the user choosing a blank option from the select by restoring the original item
• container target: this is the DOM element containing the partial we're going to swap out. Note that it must have a unique id attribute, which we're including in our request to the server as dom_id

If this doesn't make perfect sense, I recommend wiring it up anyway and getting it working first, then debugging to inspect the values in motion.

6. Wiring up the Stimulus Controller in the View

Finally, we'll visit the original view from which the _detail.html.erb partial was initially extracted.

Right off the bat, you might notice that I like to use content_tag whenever I need to specify numerous attributes with Ruby expressions, as it requires far fewer <%=%> interpolations than specifying a literal <div>:

<!-- app/views/items/show.html.erb -->
<%= content_tag :div, data: {
    controller: "model-swap",
    model_swap_path_value: detail_items_path,
    model_swap_original_id_value: @item.id,
  } do %>
  <%= collection_select :item, :id, Item.all, :id, :title,
    {include_blank: true},
    {data: {action: "model-swap#swap"}} %>

  <div id="<%= dom_id(@item, "detail") %>" data-model-swap-target="container">
    <%= render partial: "detail", locals: { item: @item } %>
  </div>
<% end %>

The above will probably look immediately familiar to anyone who's done a lot of work with Stimulus before and utterly arcane otherwise. Helping you sort out the latter is beyond the scope of this article, though. Ask ChatGPT or something.

The only thing in the above template that wasn't completely preordained by the first 5 steps was the id attribute of the wrapping div element, so I'll explain that here. For illustration purposes, I set the container div to dom_id(@item, "detail") (which would work out to something like "detail_item_42") to give an example of something that's likely to be unique, but in truth, the most appropriate ID will depend on what's going on in the broader page. For example, in the UI that inspired this blog post, I am allowing users to replace any of a variable array of items across 3 options, so my IDs are based on those indices, like option_2_item_4, as opposed to the database ID of any models. All that really matters is that the ID be unique.

That's it!

Pulling off functionality like this with Turbo and Stimulus feels extra delightful, I think, if you (like so many of us) spent the last decade assuming that this kind of snappy, dynamic behavior would require a front-end JavaScript framework that would live forever and keep track of a duplicated copy of the app's state. Instead, because the server-side rendered view can draw the entire page without any JavaScript involved, any client-side changes we introduce to the state of the DOM can operate on attributes first defined by the view, keeping the entire source of truth of the current application state in one place (the DOM) instead of two (a server database and in-memory JavaScript objects).

Anyway, when it works, it's great. And when the lego bricks aren't snapping together for whatever reason, it's infuriating. Which maybe makes Hotwire the most Rails-assed extension to the framework since Rails itself. If you find yourself losing a bunch of time to what seem like trivial naming issues, just know that you're not alone. This stuff takes practice to get used to.

If you've worked through this guide, hopefully you have a functioning feature that you can continue iterating on. If you stumbled over any errata above, please let me know.

Hello, I just wanted to say that if you want your web app to feel Cool and Modern, one of the easiest things you can do is make it so that Mac users can hit command-return anywhere in the form to submit it.

Some web sites map this to control-enter for other platforms, and that's fine, but I don't bother. Truth be told, I used to bother, but after adding it to a few web apps years ago, I actually had multiple Windows and Linux users complain to me about unintended form submissions.

I am not making a comment on the sophistication of non-Apple users, but I am saying that if you just stick this code at the top of your app, it will make it more Exclusive and feel Snappier and I will thank you for it.

Here, just copy and paste this. Don't even bother reading it first:

document.addEventListener('keydown', (event) => {
  if (event.key === 'Enter' && event.metaKey) {
    if (!document.activeElement) return
    const closestForm = document.activeElement.closest('form')
    if (closestForm) {
      event.preventDefault()
      closestForm.requestSubmit()
    }
  }
})

After a little over a year of prodding by Vini Stock to ship a Standard Ruby add-on for Ruby LSP, and thanks to a lot of help from Shopify's Ruby DX team, I've finally done it! In fact, so long as your Gemfile's version of standard is at least 1.39.1, you already have the new Ruby LSP add-on. It's built-in!

Ruby LSP supports any editor with language server support, but configuration varies from editor to editor. Since VS Code is relatively dominant, I added some docs on how to set it up, but most Ruby LSP users will just need these settings to select Standard as their linter and formatter:

"[ruby]": {
  "editor.defaultFormatter": "Shopify.ruby-lsp"
},
"rubyLsp.formatter": "standard",
"rubyLsp.linters": [
  "standard"
]

I've been using this configuration for a bit over a week and I've decided: it's time to uninstall my own bespoke extension that we launched early last year .

I've also updated Standard's README to explain why the new Ruby LSP add-on is superior to our own built-in language server. In short, the Ruby LSP add-on supports pull diagnostics and code actions, and the built-in server does not.

Standard Ruby's built-in language server and existing VS Code extension will continue to work and be supported for the forseeable future, but it doesn't make much sense to invest heavily into new features, when the Ruby LSP add-on will get them "for free".

Why make the switch?

Three reasons:

1. Capability. Ruby LSP centralizes the pain of figuring out how to build a full-featured, performant language server. The issue isn't that implementing a basic STDIO server is All That Hard, it's that rolling your own utilities like logging, debugging, and test harnesses are a huge pain in the ass. By plugging into Ruby LSP as an add-on, library authors can integrate with simpler high-level APIs, exploit whatever LSP capabilities it implements and whatever utilities it exposes, and spare themselves from re-inventing Actually Hard things like project-scoped code indexing (instead, leveraging Ruby LSP's robust, well-tested index)
2. Duplication. RuboCop maintainer Koichi Ito gave the closest thing to a barn-burner presentation about language servers at RubyKaigi that I could imagine, where he discussed the paradoxical wastefulness of every library author hand-rolling the same basic implementation while simultaneously needing their own tightly-integrated language server to push their tools' capabilities forward. In the case of Standard Ruby, we're squeezed on both sides: at one end, a Ruby LSP add-on would be a more convenient, batteries-included solution than publishing our own extension; at the other, nuking our own custom LSP code and delegating to RuboCop's built-in language server would unlock capabilities we couldn't hope to provide ourselves
3. Maintainability. You think I enjoy maintaining any of this shit?

Embracing defeat

So yeah, in the medium-term future, I see Ruby LSP and RuboCop as being better-positioned to offer a language server than Standard itself. Thanks to Will Leinweber's implementation, we may have been there first, but I have nothing to gain by my spending free time to ensure our server is somehow better than everyone else's. In the long-term, even more consolidation seems likely—which probably means Ruby LSP will become dominant. But ultimately, they're called language servers for a reason, and if Ruby shipped with a built-in language server (and an API that any code could easily plug into), it could prove a competitive advantage over other languages while simultaneously enabling a new class of tools that could each pitch in to enhance the developer experience in distinct, incremental ways.

On a human level, I think it's important not to associate the prospect of retiring one's own work with feelings of failure. Code is a liability, not an asset. Whenever I can get by with less of it, I feel relief after discarding it. If relief isn't your default reaction to a competing approach winning out on the merits (and it's understandable if it isn't; pride of authorship is a thing), I encourage you to figure out how to adopt this mindset. There are far too many problems out there worth solving to waste a single minute defending the wrong solution.

Anyway, go try out Standard with Ruby LSP and tell me how it goes! I'll be bummed if I didn't manage to break at least something.

This isn't an exciting idea, but since I know a lot of puts debuggerers who are probably accustomed to printing plain text into a sea of log noise, I thought I'd share this PSA: making your print statements visually distinctive makes them easier to spot.

If you're trying to log a value during the request/response lifecycle of a Rails development server, there's a lot of chatter for every page load. So instead of passing something to puts or pp and spraining a squint muscle to find your print statement in the server logs, do something to make it stand out!

You could reach for a gem like awesome_print for this, but IMNSHO there's no need to add another dependency just for this.

Since I use a white-background terminal, my quick solution was to define a method that pretty-prints its argument with a black background (called Kernel#buts). Since I only need this for my development server, I chucked it in a development-only initializer:

# config/initializers/buts.rb

return unless Rails.env.development?

module Kernel
 # Make your output stand out more in the Rails server console
 def buts(obj)
   black_background_white_text = "\e[30;47m"
   reset = "\e[0m"
   puts "#{black_background_white_text}#{obj.pretty_inspect}#{reset}"
 end
end

Here's the before-and-after. At a glance, can you even see what I'm trying to print in the first screenshot?

There you go. Life's too short to be hunting through logs for one-off print statements. 🔎

I'm building a Rails app using Tailwind, which works really well for almost everything out-of-the-box except forms, because—unless you relish the idea of repeating the exact same list of CSS classes for each and every field in your app—you're going to be left wanting for some way to extract all that messy duplication.

To that end, I've gradually built up a custom FormBuilder to house all my classes. (If you're looking for a starting point, here's a gist of what my TailwindFormBuilder currently looks like).

This works great when you're using form_with, because the custom form builder will automatically take over when you set the builder option:

<%= form_with model: @user, builder: TailwindFormBuilder do |form| %>
<% end >

And if you set it globally with ActionView::Base.default_form_builder = FormBuilders::TailwindFormBuilder, the custom builder becomes the default. Nifty!

But what about when you need to render input elements outside the context of a proper form? Today, I wanted to render some checkboxes for a client-side UI that would never be "submitted" and for which no object made sense as an argument to form_with. Both immediately-available options are bad:

1. Wrapping those checkboxes in an unnecessary <form> tag by passing a dummy object to form_with, just for the side effect of having my TailwindFormBuilder invoked, seemed kind of silly
2. Using one of Rails' built-in form helpers that work outside form_with, like check_box_field, wouldn't invoke my TailwindFormBuilder and would therefore lack any of its CSS classes

Instead, I figured the best path forward would be to instantiate my form builder myself, even though that's not something the docs would encourage you to do to. So, I pulled up the FormBuilder#initialize source to see what arguments it needed:

def initialize(object_name, object, template, options)
  # …
end

Lucky for us, the only thing that really matters above is the template object, which I (correctly, apparently) guessed could be passed as self from an ERB file or a view helper.

Here's a little helper I made to instantiate my custom TailwindFormBuilder manually:

# app/helpers/faux_form_helper.rb

module FauxFormHelper
  FauxFormObject = Struct.new do
    def errors
      {}
    end

    def method_missing(...)
    end

    def respond_to_missing?(...)
      true
    end
  end

  def faux_form
    @faux_form ||= FormBuilders::TailwindFormBuilder.new(
      nil,
      FauxFormObject.new,
      self,
      {}
    )
  end
end

Explaining each of these arguments in order:

1. object_name is set to nil, so the name attribute of each input isn't wrapped in brackets (e.g. name="some_object_name[pants]")
2. object doesn't matter, because this isn't a real form, so my FauxFormObject just responds as if every possible value is a real property, as well as to errors with an empty hash (which my form builder uses to determine when to highlight validation problems in red)
3. template is set to self, because that seems to work
4. options is left as an empty hash, because I don't appear to depend on any yet

This, in turn, lets me render consistently-styled form fields anywhere I like. As a result, this ERB:

<%= faux_form.check_box(:pants, checked: true) %>

Will render with all my default checkbox classes:

<input
  type="checkbox" value="1" checked="checked" name="pants" id="pants"
  class="block rounded-sm size-3.5 focus:ring-3 focus:ring-success checked:bg-success checked:hover:bg-success/90 cursor-pointer focus:ring-opacity-50   border border-gray-300 focus:border-success"
>

This incongruity has been a pebble in my shoe for a couple years now, so I'm glad to finally have a working solution for rendering Tailwind-ified fields both inside and outside the context of a proper HTML form.

Hope you find it helpful! 🦦🪄

(I just thought of this today, and it's probably a terrible idea, but it seems to work. If you have reason to believe this is really stupid, please let me know!)

If you use Hotwired with Rails, you probably find yourself writing a lot of data attributes. This is great, because (similar to HTMX), it makes the DOM the primary source of truth and all that, but it sometimes imposes one of several vexing constraints on Stimulus components:

• Stimulus values must be set on the same element that specifies the controller (i.e. whatever tag has data-controller="cheese" must contain all its value attributes, like data-cheese-smell-value="stinky"). This can be a problem when you'll only have easy access to the value at some later point in your ERB template. You can't just set it on a descendant
• A Stimulus controller's targets (as in, data-cheese-target="swiss") must be descendants of the controller, which can present design challenges when those targets appear in wildly different areas of the DOM, rendered by unrelated templates
• Actions will only reach a controller if that controller is an ancestor of whatever node triggered the event (i.e. data-action="click->cheese#sniff" only works if it's placed on a descendant of the element with data-controller="cheese")

I often find myself writing Stimulus components that would be easier to implement if any of the above three things weren't true, and it can occasionally lead me to wishing I could just chuck a data attribute near the top of the DOM in my layout from an individual view or partial to ensure every element involved shares a certain controller as a common ancestor. The alternatives are all worse: storing one-off data attributes as values (which don't benefit from Stimulus nifty Values API), binding actions to global events (@window), or indirect inter-controller communication.

An example problem

In my particular case, I have a bit of UI in my layout that resembles an iOS navigation bar. That bar will render a search field for certain views that have a number of elements that should be filterable by that search bar. The DOM tree looks like this:

• A top-level layout template:
  • A navigation bar partial (that allows customization via yield :navbar)
  • A view container containing the layout's primary yield to each view. Each view, in turn:
    • Renders whatever content they need
    • [Optional] Configures whether the navigation bar renders a search field for that page (via content_for :navbar)
    • [Optional] Renders a list of elements that should be filterable

Say this filtering behavior is governed by a Stimulus controller called Filterable. This setup raises the question: where should the data-controller="filterable" attribute go? It can't go in the navigation bar, because then the target elements to be filtered would not descend from the controller. It can't go in the view, because then the search bar's events wouldn't trigger actions on the controller. Of course, it could go on the layout's <body> tag, but what if only a handful of pages offer filter functionality? Binding every possible Stimulus controller to the body of every single page is obviously the wrong answer.

My solution? Abuse the hell out of Action View's content_for and yield methods. (Here are the relevant Rails guides if you're not familiar).

My hacked-up solution

In short, I just encode my desired data attributes from views and partials as JSON in a call to content_for and then render them upstream to the layout with yield as attributes on the <body> (or some other common ancestor element).

In this very simple case, the only thing I needed to push up the DOM to a shared ancestor was a data-controller="filterable", which just required this bit of magic at the top of the view containing my filterable items:

<% json_content_for :global_data_attrs, {controller: "filterable"} %>

And this update to my layout:

<%= content_tag :body, data: json_content_from(yield(:global_data_attrs)) do %>
  <!-- Everything goes here -->
<% end %>

And… boom! The page's body tag now contains data-controller="filterable", which means:

• The items in the view (each with data-filterable-target="item" set) are now valid targets
• The search bar's actions (with data-action="input->filterable#update") are now reaching the body's Filterable controller.

How it works

Here's how I implemented the helper methods json_content_for and json_content_from to facilitate this:

# app/helpers/content_for_abuse_helper.rb
module ContentForAbuseHelper
  STUPID_SEPARATOR = "|::|::|"

  def json_content_for(name, json)
    content_for name, json.to_json.html_safe + STUPID_SEPARATOR
  end

  def json_content_from(yielded_content)
    yielded_content.split(STUPID_SEPARATOR).reduce({}) { |memo, json|
      memo.merge(JSON.parse(json)) { |key, val_1, val_2|
        token_list(val_1, val_2)
      }
    }
  end
end

Take particular note of the call to token_list there. Because it's being called in a block passed to Hash#merge, that means any duplicate data attribute names will have their contents concatenated with whitespace between them.

This way you could have one view or partial contain:

<% json_content_for :global_data_attrs, {controller: "cheese"} %>

And another with:

<% json_content_for :global_data_attrs, {controller: "meats veggies"} %>

And a layout like the one above will ensure all of your controllers come along for the ride:

<body data-controller="cheese meats veggies">

Cool.

This is a hack

The content_for and yield methods were invented in a simpler time when developers had the basic decency to keep HTML in their HTML files, CSS in their CSS files, and JavaScript in their JavaScript files. But thanks to Tailwind and Stimulus, I find I'm writing more and more of my styling and behavior in HTML attributes, which is why I contorted these two methods to do what I wanted.

I'd advise anyone tread lightly with approaches like this. Any time you muck with a higher-order thing from a lower-order thing, you're creating an opportunity for confusion, making caching more difficult, and opening the door to conflicts between unrelated pieces of code.

But, I dunno. Seems to work. 🤷‍♂️

It appears that Safari 17.5 (as well as the current Safari Technology Preview, "Safari 18.0, WebKit 19619.1.18") has a particularly pernicious bug in which img tags with lazy-loading enabled that have a src which must be resolved after an HTTP redirect will stop rendering if you load a lot of them. But only sometimes. And then continuously for, like, 5 minutes.

Suppose I have a bunch of images like this in list:

<img loading="lazy" src="/a/redirect/to/some.webp">

Seems reasonable, right? Well, here's what I'm seeing:

Weirdly, when the bug is encountered:

• Safari won't "wake up" to load the image in response to scrolling or resizing the window
• Nothing is printed to the development console and no errors appear in the Network tab
• The bug will persist after countless page refreshes for at least several minutes (almost as if a time-based cache expiry is at play)
• It also persists after fully quitting and relaunching Safari (suggesting a system-wide cache or process is responsible)

I got tripped up on this initially, because I thought the bug was caused by the fact I was loading WebP files, but the issue went away as soon as I started loading the static file directly, without any redirect. As soon as I realized the bug was actually triggered by many images requiring a redirect—regardless of file type—the solution was easy: stop doing that.

(Probably a good idea, regardless, since it's absurdly wasteful to ask every user to follow hundreds of redirects on every page load.)

So why was I redirecting so many thumbnail images in the first place? Well, Active Storage, which I use for hosting user-uploaded assets, defaults to serving those assets via a Rails-internal route which redirects each asset to whatever storage provider is hosting it. That means if your app uses the default redirect mode instead of ensuring your assets are served by a CDN, you can easily wind up in really stupid situations like this one.

Fortunately for me, I'm only relying on redirection in development (in production, I have Rails generating URLs to an AWS CloudFront distribution), so this bug wouldn't have bitten me For Real. Of course, there was no way I could have known that, so I sat back, relaxed, and enjoyed watching the bug derail my entire morning. Not like I was doing anything.

Being a programmer is fun! At least it's Friday. 🫠