Add a FuzzyEq trait

[cmyr]

We should have a floating-point aware equality comparison. There are crates out there that provide this logic, and I think it's tricky enough that it's worth a dependency if somebody has done it well.

I suspect the best way to do this would be via some sort of FuzzyEq trait. the float-cmp crate looks like a likely candidate, but I would ideally like something that could, say, infer a good epsilon value for you based on the input values, although maybe this is a bit optimistic.

[richard-uk1]

I agree about float-cmp. I have a fork where I started implementing FuzzyEq for the kurbo types. The idea of the float-cmp trait is that it allows you to choose between ULPS and a fixed epsilon. IIUC, ULPS is an integer that represents the number of fp numbers between the two comparison numbers, which means that the absolute range of accepted numbers scales with the size of the (largest) input. I think this gives you what you want @cmyr?

The downside of the comprehensiveness of the FuzzyEq trait is that it is more work to implement, but IMO this is worth it, because being able to test functions for accuracy is so important.

[cmyr]

Yea, that sounds about right. I'd be curious to see what the API actually looks like, if you have a branch?

[richard-uk1]

We talked at the xilem office hours. We decided that the best way forward is to experiment with using float-cmp in tests, but not expose it in the public API. We can then get a feel for how well it works before we commit to it.

[ctrlcctrlv]

Why not expose it with an option like IOF does?

[richard-uk1]

Doing it in tests doesn't preclude adding it in future.

…

On Wed, 22 Mar 2023, 18:40 Fredrick Brennan, ***@***.***> wrote: Why not expose it with an option like IOF does? — Reply to this email directly, view it on GitHub <#50 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAKT4XSEWDSEUSTCJBURTFLW5NBS3ANCNFSM4IG5KHTQ> . You are receiving this because you commented.Message ID: ***@***.***>