It turns out that LLMs are really good at writing bash too. even perl! maybe we should rethink some of these lost bits because we no longer need to worry about the arcane parts.
It turns out that LLMs are really good at writing bash too. even perl! maybe we should rethink some of these lost bits because we no longer need to worry about the arcane parts.
Define "really good?" I don't think I've ever had them produce a script that I didn't need to correct in some way. They can write it, that's true, but we still need to be able to read it.
Fair. I'm comparing it to the output I see generated in more mainstream, day to day, languages. I'm not going to bother to get into an argument on if the frontier models can generate amazing code in general.
What I've found is I can get get the frontier models to generate bash scripts, perl one liners, etc that do exactly what I need at roughly the same quality as any other code it generates.
100% They're super useful and I often use them to generate bash scripts of, but that's exactly how I know how important it is to check their work!
I'm a shell scripter at heart, the arcane stuff has always been a delight for me, so I've driven them to do some pretty complex stuff where previously I would have "copped out" and used python. I'd say the general adage of them being at the level of a very talent junior holds true.
I agree with you, instead of working on creating a bash script if we just sit as a senior developer ,check and correct what is needed is the way forward, it kind a remind me Pair programming when one do the scripting and other corrects it
[dead]
Bash scripts requires "set -eu" at minimum, when written by LLM or human.
It's the only language terrible enough to make the default behavior ignore undefined variables, commands, and execution errors, and happily continue executing whatever was produced by me smashing my hands on the keyboard, until the end of the file, while returning an exit code of 0, claiming complete success.
“Only language” eh? I only have a passing familiarity or very distant memory so could be wrong but I’d say these are the same:
Perl (needs use strict)
Ancient VB/VBA/VB script
Original PHP (no idea about modern)
PowerShell
Old Windows/DOS batch
Javascript
DataFlex
> Bash scripts requires "set -eu" at minimum, when written by LLM or human.
No. It does not. Just write good scripts and catch errors and handle them. That's the same more less with every language. And bash can be written in a way that it is sane and readable and maintainable. Just because you have seen a lot of junk in the language does not make the language bad per se. Sure there are a lot of languages "features" which are more then questionable but I want to rise again the point that it has its place and can be used in a good way.
Ah yes, the Feynman Algorithm for shell scripting.
I have seen them write massively more complex bash scripts than any normal person would ever attempt, but the failure rate is absurdly high compared to when they write sane typed languages.
An LLM taught me how to use the shell colon. Shell scripts are short enough that it's been worth my time to edit the AI's work.
I have some absolutely amazing bash scripts that I would never have contemplated making myself in bash. And they'd only have been somewhat better in python. Claude for the win!