Ted Driggs

Why isn’t that signature what we want in most cases?

> would you consider supporting this in the project? Not at this time; `derive_builder` has widespread-enough usage that I am trying to keep it as lean and focused as possible....

> This allows for partials at the top-level, but any nested structs must be completely actualized. I'm not understanding this piece. Referring back to this example from the original post:...

Given the lack of activity on this over the past 3 months, I'm closing as "Not Planned" - if there is more to discuss, we can pick it back up.

I'm not sure when I'll have time to review this, but I have it on my to-do list.

What does syn say the type of the `print_foo` declaration is?

This just burned me as well. I agree that the callback should be optional but regardless silent failure isn't the right behavior.

Pulling this thread further - it looks like several methods promise to only return one of the variants, e.g. `SafetyContext::enable` can only return the `ConditionalNoEffect` variant. In those cases, should...

> I will say ahead that I would like to keep the amount of code to a minimum, so I think specifically stuff like pulling the ConditionalNoEffectError into its own...

I understand the desire and some of the use-cases for this, but I don't think this is possible under `darling`'s current architecture. `!my_feature` isn't a valid [`syn::Meta`]. `darling` _recursively_ transforms...