Antoine Fabri

Meh, let's wait for use cases or requests

Thanks! I will take a look then. There are some design decisions to make though: (1) I could either search the files for function definitions, or assume they contain only...

Let's close it and follow up in #138, seems to have a better scope

json is actually awkwards because serialisation cares about attribute order while the rest of R, including identical() and waldo::compare() do not. And things like subsetting will shuffle the attribute order:...

check out `?constructive_print_mode` with the development version : ``` options(constructive_print_mode = "clipboard") iris[1:2, ] |> construct() ``` I considered having a different function or a different arg but an argument...

I understand, another argument for using the option was that I don't want to maintain `construct_multi_clip()`, `construct_reprex_clip()`... but I wanted the feature for those as well. Maybe we can have...

Also waldo doesn't see those. Ultimately we really need our own waldo, with: * More rigorous use of subsetters, `.subset()` and `.subset2()` * An actual comparison object that we can...

was closed by mistake

Maybe we test if the serialisation is correct, and if it's not we rerun more carefully ?

Since there's a lot of code to update I'd like to find a way to keep the old and new way coexisting. I think we can build some `opts` and...