A note to self I might have already noted... myself.

angular.module('MyApp')['_invokeQueue'].forEach(function(value){ 
    console.log(value[1] + ": " + value[2][0]);
});

Sauce.

So look, I'm at a company where much of the app is still in AngularJS. The downside is that this reminds me of a post I'd shared before that said:

If your list of things to develop is really a list of things that you won’t have to do in a more functional environment, none of which will make you more employable elsewhere… it’s time to walk away.

That's not wrong. More experience in AngularJS isn't making me more employable at this point unless I want to work in the small subset of positions that plan to continue using it well past its dead-on date. Which, again, was January of this year.

That admission aside, as I've mentioned before, there is, however, a practical advantage: Most of the questions I have when I'm developing have nearly canonical answers. I mean, have you seen StackOverflow's blog on the Next.JS conference? The blog's title is "Goodbye Webpack, hello Turbopack!". With just a little bit of exaggeration, webpack, we barely knew ye.

I was looking to get us to a React stack, and I'd initially liked this "minimal" build tool suite over at 2ality. Okay, okay, it's been over three years, but it's now kaput. Gone. Snowpack is dead. I'd heard a bit about ViteJS already a bit from my RSS sub to Shawn Wildermuth's blog, and sure enough that's what Snowpack says to use now. If I'd learned Snowpack inside and out (not that it was tough to learn, and I've been playing with it on and off for a few years), that'd be nearly wasted time now.

Look, it's fine to need to keep up a little, it's not like .NET 6 doesn't have a decent amount of new stuff (does it though?), and maybe it's just that I've been around long enough to see things change several times, but it doesn't really have to, does it? There are database admins who haven't needed to learn anything exceptional for their whole career. I mean, look, it's not like our buddy Mr. Dave hasn't kept things moving, but it ain't client-side JavaScript. He's had a chance to get good at what he does, to really drill-down and understand rdbms engines.

Okay, yes. I'm a little jealous. 😏

And if I haven't mentioned it before, I'm pretty sure it's because of conferences, YouTube, and Twitter. Everybody who wants to be a bleeding edge type can be, and they all race each other to be the new expert of something nobody else can be yet because it was just released. God made humans, but constantly evolving tech stacks made humans equal[ly clueless].

That said, I do need to write my follow-up to my intro to why we're using transpilation-free Preact. Preact really is a wonderful bridge from a hopelessly out of date client-side app to, well, at least a TypeScript-powered React refactor. Need to develop something new now but can't, for whatever reason, create all new tooling?

If you can limit support to IE 9+ or so, Preact (with hooks) is your stack. We've used Preact with Enzyme to create a tested, cradle-to-grave page with a reasonably complex UI and have it running well right smack in the middle of an AngularJS app. Set down new work in Preact now, start eating from the top with React, and when they meet in the middle, it's a very quick port.

So maybe I won't need to remember that cheat to get all directives, constants, etc out of an AngularJS site for too long. But when I forget again, at least I'll know I've got it sitting here.

Labels: angularjs, enzyme, noteToSelf, obsolescence, preact, vitejs