The sentence you are referencing is P0144R0: Structured bindings . The Kona mailing list lists these documents according to the Evolution working group. It is not covered by the current Evolution Working Group (EWG) active issues list, and Nikol Bolos indicates that the EWG active issues list has not been updated for Kona. Once it appears in the active EWG list, you can track this offer by tracking the problem.
There are some really good travel reports, and in particular the Botond Ballo trip report , applies to this offer in the section "Offers for which additional work is offered, and it says:
The proposal to initialize the destructuring, which will allow to write auto {x, y, z} = expr; where type expr was a tuple-like object whose elements will be associated with the variables x, y, and z (which the construct declares). Tuple objects include std :: tuple, std :: pair, std :: array, and aggregate structures. The proposal did not have a mechanism for adapting a non-aggregate user type for "tuple-like" and work with this syntax; Feedback from the EWG was that such a mechanism is important. In addition, the EWG recommended that this sentence be extended to allow (optionally) to specify types for x, y and z, rather than deriving their types.
We can find the WG21 mailing list by going to the WG21 website and going to the section.
As TC notes there is also a competing proposal P0151R0: Multi-Declarators proposal , which states:
We offer a better approach for “Structured bindings” as defined in P0144R0, where “better” is defined as a terrier, more orthogonal, more general, more expressive, less (parsing) ambiguous.
source share