https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123289

--- Comment #10 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Does it perhaps make sense to support overloading here as a GNAT extension
> as well as another extension relating to RM 4.3.5(22/5) to allow for
> disambiguation of positional_container_aggregate and
> named_container_aggregate? This would allow for very clean declarations of
> arbitrary JSON objects such as:
> 
> ["A" => 1, "B" => [1, 2], "C" => ["a" => 1]]

But would probably be an implementation nightmare, given that the resolution of
aggregates is already quite complicated.  I'd suggest creating an issue with
the ARG at https://github.com/Ada-Rapporteur-Group/User-Community-Input
instead.

Reply via email to