Closed pixelzoom closed 1 year ago
In Slack#DM, @AgustinVallejo asked me:
I'm not sure at all how https://github.com/phetsims/keplers-laws/issues/171 would be happening. After examining the call stack, it seems like for that reported value to be a NaN, things should've broken way before reaching that point. Not sure what you'd recommend for that case? I'm also not sure how I would replicate it
I'm not sure how to proceed here. I don't know why the error message includes JSON that looks like it's from NumberIO. I recommend that you consult with @zepumph, since he's more familiar with validation and its PhET-iO aspects.
This is just validation code working as desired. If you look at the call stack, you set a property that says "no NaN values please" to NaN. One of these calls:
I wasn't able to reproduce in 3 minutes of fuzzing. @AgustinVallejo let me know if you want to discuss more.
@zepumph that's what I figured, but to me it's weird that it happened in line 247 since for any of these values to become NaN, assertions should've triggered way before. And I haven't been able to reproduce either.
I didn't see any problems in the spot check when fuzzing on the debug version. If there is another way to test this, let me know.
Nope, not that I know of. Should I close then?
If there's nothing else for us to look out for, sure.