Jarret Moses

I am attending the biggest #Reactjs conference remotely. You may get a limited free ticket (50% of talks, no workshops) and join me by clicking this badge. #react #reactjs gitnation.com/badges/react-s… #reactsummit

@mattpocockuk What resource(s) would you use to up skill

Pretty sure that soon you'll need an LLM for your app as often as a database. And the same concerns apply. Security, performance, observability. A lot of devs are going to need to upskill.

@mattpocockuk These config files can be per env

@mattpocockuk Store it in an encrypted file, say using KMS, at build time decrypt and generate a module you can import like import conf from ‘myConf’

Your runtime environment variables are not safe. Just this code in an npm package would leak them to a malicious endpoint. So, where should you store your secrets?

@GGBreakingPoint Is there a stream?

Grand Finals is SET LAT vs Tiny Terrors

@mattpocockuk It’s crazy how this exact topic pops into my head frequently and boom @mattpocockuk encapsulates those thoughts perfectly

But this is worth bearing in mind if you want to control the way your types display when they hover. Types are more likely to display computed readouts (though, not always). Interfaces will pretty much always display the name of the interface.

Did you know that types and interfaces handle hovers differently? Let's thread up 🧵

@housecor For this example I wouldn’t do DRY. The constant itself could be incorrect and if your test and code use the same constant you end up with a false positive. So I like to explicitly write out my expected result to ensure a const’s correctness as well

Two ways to write a test: 1. DRY: reference const that’s used in code 2. Repeat: repeat value that’s used in code Which do you prefer, and why?

@NJTRANSIT I think your app is having issues. It’s erroring out trying to purchase bus tickets with error code 69

@HyDrAnml LETS GOOOO HYDRA. Crush em. You’re still the MVP and it’s not even close

Ggs faze won 3-2

@thms_clrk TablePlus is great

@jayphelps You can strongly type but fetching data still takes time. What’s the ideal component look like?

I *really* don't like the extremely common pattern in React apps of propagating null/undefined values throughout all your components while data is still loading (or missing). The reality is much worse than this, but a contrived example:

@YenSauce Solid start

@Bruhooke That was rough 🙁

DFRD lol so pitiful ggs

@Subliners Not great boys 😞

Staying alive. 1 round at a time #StandClear | #NYSL

@Bruhooke My first time at a live event. I am really disappointed by how much this affected the game. Can’t imagine the mental toll it takes on them.

HAHAHA HERE we go AGAIN great game & pcs league 🤣

@CesarSkyz That match has an asterisk in my books. Just get the dub tomorrow

@kenwheeler @ReactSummit Was a great presentation

When I was a little kid I used to go to the Liberty Science Center. Then I took my kids there. Almost got thrown out for ripping a juul lol. Now, thanks to @ReactSummit , I got to speak here with my slides on the god damn planetarium. Pretty special stuff.

@housecor I like this. Something recently I worked on was generating all our fixtures based on our swagger api definitions. All typed and simple to use. Very fun stuff and totally based on our api. If the api changes the fixtures change.

Problem: Mock datasets tend to be bloated, unreliable, and repetitive. So, it's often unclear what mock data can safely be changed, reused, or added. Solution: 1. Create *strategic*, static mock data, from scratch (it often doesn't require very much data to test an app). 2. Document what each piece of mock data is for. I use the TypeScript type below to enforce it. Benefits: ✅ The mock data is static, so automated tests are reliable and fast. ✅ Avoids needless bloat. A small, focused, purpose built dataset is easier to maintain. ✅ Documents the unique purpose of each piece of mock data. ✅ Automated tests and devtools can access specific pieces of mock data based on description, which improves test readability and documents intent. ✅ People are less likely to change critical data and break tests, because they understand the purpose of each piece of data.

@jessephelps @housecor That’s a good one. I like WET - Write everything twice

You know DRY - Don’t Repeat Yourself. Let me introduce MOIST - Moderate Overlap Is Sometimes Tolerable.

@ryanflorence Can’t someone change a tailwind class and it breaks everything?

First, someone will change .previous__button in some other UI and break this one and nobody will know until it’s in production, and maybe not for a while.

After 26 years of CSS experience, I'll take the left every single time.