matthias314

I'm willing to give it a try, although I don't know how much there is to do. Do you think that there are changes outside of `types.jl` required?

The proposed function ``` islongdef(ex) = @capture(ex, function (fcall_ | fcall_) body_ end) ``` seems to work (although I don't fully understand why). Replacing `(fcall_ | fcall_)` by `fcall_` would...

I could give it a try. Do you insist on using `@capture`? For example, I think one could simply say ``` islongdef(ex) = isexpr(ex, :function) ``` Of course, this doesn't...

The optimization level seems to play a role. For Julia 1.10.2 and my Skylake machine I get: `-O2` and `-O3`: ``` Int8: f1 = 537.593 ns, f2 = 340.635 ns...

`map` works for tuples with different lengths by now: JuliaLang/julia#49336

Could one always use `@fastmath` for `sum` and `prod`?. When I tried it out, the fast versions dealt with `NaN` correctly. Rearranging terms should be allowed for `sum` and `prod`...

Thanks for the detailed explanations! I didn't worry about floating point precision because my understanding was (and is) that also the current implementations of `sum` and `prod` add up elements...

I mean that `sum([x1,x2,x3])` boils down to a fixed evaluation order like `(x1+x2)+x3`. If I know the evaluation order, I can cook up a bad input. Changing the evaluation to...

Whatever option is chosen, it would be great if the documentation talked about it. Moreover, in the MWE above the docstring for `f(::Int)` is silently discarded at present. In other...

[I'm moving the discussion out of the code review because it has a wider scope] In my opinion, we should not allow the syntax `f{T}(x::T)` for docstrings if `f` is...