David Widmann

Are there any news regarding this issue? I would be very interested in arm/v7 support and I noticed that a PR (https://github.com/TomBursch/kitchenowl-backend/pull/13) already exists :slightly_smiling_face:

The depwarns are annoying but they don't cause errors (at least not in my CI setup). The actual problem for me other that it broke tests - from all green...

A simple example (inspired by test/linearity.jl in MTK): ```julia julia> using ModelingToolkit julia> # Define some variables @parameters t σ ρ β 4-element Vector{Num}: t σ ρ β julia> @variables...

AFAICT the DelayDiffEq test failures are caused by this PR.

I like `teststatistic` most, I think. I agree that `statistic` seems a bit too general, and I prefer not shortening the function name (but I'd be fine with `teststat` as...

I always assumed the solutions support the Tables interface? Doesn't `DataFrame(x)` support anything that supports the Tables interface?

This seems too complicated to me. Isn't it possible to just convert apply `float` to parsed quantities as suggested in #65 but otherwise not modify the user input?

To me it seems the main problem is that the current implementation of the `u_str` macro always returns a `Quantity`. If no values are given, I think it should just...

> These conversions are to make the return value of uparse type stable It's a macro, so the string is replaced with whatever expression you want to replace it with....

I still think that it would be preferable to change the macro and parsing in more fundamental ways. But in any case, I don't think adding a new numeric type...