Does trim() order matter in chaining?

[JoshuaHintze]

I've been arguing with Code Rabbit and the AI on zod's documentation. My question is

Does this

z.string().min(1).trim(), or this
z.string().trim().min(1)

give the same result?

Below is code rabbit's output

Test Results:

z.string().min(1).trim() with input "   " (3 spaces): PASSES ✓ → returns ""
z.string().trim().min(1) with input "   " (3 spaces): FAILS ✗ → "Too small: expected string to have >=1 characters"
The order does matter. With .min(1).trim(), the min check runs on the untrimmed string first (3 characters passes), then trim executes. With .trim().min(1), trim runs first (producing an empty string), then the min check fails.

Documentation:

Official Zod docs: https://zod.dev/
Strings documentation: https://zod.dev/?id=strings
According to the web search results, .trim() is a transformation that runs after validation checks, which explains why order matters.

For your PR: If you want whitespace-only strings (like "   ") to be accepted and trimmed to "", the current order is correct. If you want them rejected, swap to .trim().min(1). Based on the PR title mentioning "improved validation," I assumed rejecting whitespace-only input was the goal, but if the current behavior works for your use case, no change is needed.

My local testing shows it doesn't matter but wanted to hear from others.

Thanks!