LPT: ChatGPT is incredible for generating and evaluating regex
I have to use a ton of regex in my new job (plz save me), and I use ChatGPT for all of it. My job would be 10x harder if it wasn’t for ChatGPT. It provides extremely detailed examples and warns you of situations where the regex may not perform as expected. Seriously, try it out.
The new Code Interpreter plugin that went live for this week for Plus users can actually execute Python code on a sandboxed environment. This allows you to add “Write and execute tests for the regex” to the end of your prompt.
I often need to deal with half a dozen different programming languages in any day/week and the context switching can be difficult at times. When you’ve spent all day switching between JavaScript, Python, and YAML and suddenly need to draft some Regex, tools like ChatGPT can help immensely at reducing the mental burden of switching gears.
The syntax of regular regexes is the same across languages though. It’s just the regex library which is different, but so is every other library between languages.
I have yet to see a regex that is so complicated that I would need some help. I expect programmers to know how to use regexes but it seems that it’s not the case. And when it becomes too big, you always can write verbose regexes with comments, it’s even easier. If someone could show me something too difficult for a human being (excluding the regex to validate emails), I’m interested.
It’s not that I’m incapable of evaluating regex, but rather the mental burden of evaluating complex regex statements and determining their purpose can be time-consuming. Why take 20 minutes to understand some regex when ChatGPT can do it in 20 seconds?
A coworker once defined regex as a write-only language and he definitely had a point. I love regex but it can be time consuming figuring out exactly what a complex regex expression is doing.
This this this!!! I know this is a post from the place that shall not be named, but it just showcases the issues with ChatGPT (this is from when GPT4 was just released)
Up to now that usually was faster than trying to get chatGPT to generate something worthwhile. However, if you define some test cases first, the combination of both will even get the sales guy there eventually.
That depends on the regex flavor. Some of them have full support for variable length lookbehinds, for example JavaScript and third-party regex module for Python.
I’ve tried it and found it wanting at regex and excel formulas, but I’m glad to hear it’s working for you! Are you using 4? I haven’t tried that one and I hear it’s better.
I typically try 3.5 first and switch to 4 if the results aren’t great. 3.5 typically handles basic use cases quite well, for example, writing regex that detects jira ticket naming nomenclature. For more complex things, I go to 4.
It sometimes gets things wrong, but I’ve also found that just saying “that didn’t work” gets it to reevaluate for more complex situations
it helps if you hold ChatGPTs hand and walk it through what you need. For example if you have a regex with 3 requirements, ask it to write a regex for the first requirement, then ask it to modify the previous output to add another requirement, and so on. that way you can sort of “audit” it as it generates the correct regex.
So I was trying to write a regex for use with my ChatGPT discord bot. I wanted to trim off any final partial sentence at the end. I went around and around with it for a couple of hours because look ahead and look behind are just not something I do often enough.
It kept writing more and more complicated regex that didn’t work. The final solution, while not exactly perfect - it won’t keep a quote at the end of a sentence, and honorifics like Mr. and Dr. throw it - it wasn’t nearly as complicated as ChatGPT was making it. It still never did give me anything working - I just fucked around on regex101 until I got it right. As usual but having wasted 90 minutes or so.
I’ve found that you need to be very careful when asking it to modify things it produced directly without making significant changes to the regex it provides. Once I get to the 3rd or 4th iteration of asking it to modify previous responses I’ve found the likelihood that it starts hallucinating to increase dramatically. The best solution I’ve found to this is to put your entire request in a single prompt that walks it through all requirements step-by-step.
Add comment