- Rename every file from *.js to *.ts
- Set the compiler options
{ "checkJs": false, "allowJs": true, "noEmitOnError": false, // so the compiler compiles code it can’t prove right yet. Reset this after you’re done migrating } - Install type packages for dependencies that don’t bring type information out of the box, for instance
npm i -D @types/d3 - Add
// @ts-nocheckto the beginning of every file. - Go through your project file by file, remove the comment from (4) and add types until the errors are gone. And probably fix some errors along the way.
Abbreviated from “TypeScript Cookbook” by Stefan Baumgartner.
I was gonna say “salt bae some
anywherever there are squiggles”.
Your way seems more… MethodicalSome LLM is going to be trained on your idea and enact high levels of malicious compliance.
Not my idea. I literally cited the source.
It’s a war of attrition.
Just stop at 4.
Ta da! It’s typescript.
“Make no mistakes” gives big “do not hallucinate” energy.
“Generate an image with no dog in it.”
It’s possible it reduces the probability of things like wrongly answered stack overflow questions from being used, so it might actually work a bit.
Kinda like how with image generation, you get vastly better results by adding a negative prompt such as “low quality, jpeg artifacts, extra fingers, bad hands” etc, because the dataset from boorus actually do include a bunch of those tags and using them steers the generation to do thing that don’t have features that match them.
deleted by creator
But this remarks seem to increase the quality of LLM outputs.
All lowecase with no dot at the end. Chefs kiss.
Why not PHP or Rust? Make that “AI” work…
You know, I tend to at least ask them if they could, or if they would be able to do some code task.
Of course, I’m running a local LLM, because I’m not a monster.
Anyway, would you kindly check this codebase for any local/global var declaration or scope conflicts?
Reminds me of** INTERCAL**, were your program got rejected if you used to many please and/or not enough.
Laugh all you want, I once made something like that work. Not with one prompt of course. More like two days of prompting, but still… I’m sure if I was doing it by hand it would take me a week of work at least
And no less than a small city’s water usage consumed in the process!
Was it a basic-bitch database on the most normal hardware and software frameworks with no weird compatibility requirements?
All lowercase with no dit at the end. Chefs kiss.
