P1830 std::dependent_false

[brycelelbach]

P1830R0 std::dependent_false (Ruslan Arutyunyan)

[brycelelbach]

Cologne 2019-07 LEWGI Minutes

P1830R0 std::dependent_false: Direction Review

Champion: Ruslan Arutyunyan

Minute Taker: Jayesh Badwaik

Start Overview: 07-18 13:36

Start Polling: 13:55

POLL: We should promise more committee time to pursuing this proposal, knowing that our time is scarce and this will leave less time for other work.

Strongly For	Weakly For	Neutral	Weakly Against	Strongly Against
4	5	2	1	0

Attendance: 18

// 0
template <typename Dep>
inline constexpr dependent_false = false;

// 1
template <typename... Deps>
inline constexpr dependent_false = false;

// 2
template <bool B, typename... Deps>
inline constexpr dependent_bool_value = B;

// 3
template <typename T, T t, typename... Deps>
inline constexpr dependent_value = false;

// 4
template <typename T, T t, typename... Deps>
struct dependent_constant : std::integral_constant<T, t> {};
// Plus template aliases and variable aliases.

// 0
template <typename Dep>
inline constexpr bool dependent_false = false;

// 1
template <typename... Deps>
inline constexpr bool dependent_false = false;

POLL: Which do you prefer?

Option	Votes
# 0	3
# 1	7

Attendance: 18

Option 1 has consensus.

// 1
template <typename... Deps>
inline constexpr bool dependent_false = false;

// 2
template <bool B, typename... Deps>
inline constexpr bool dependent_bool_value = B;

// 3
template <auto V, typename... Deps>
inline constexpr auto dependent_value = V;

POLL: Which do you prefer?

Option	Votes
# 1	6
# 2	5
# 3	1

End: 14:05

Author's choice between either 1 or 2.

CONSENSUS: Bring a revision of P1830R0, pursuing one of the two design options below, to LEWGI for further design review.

template <typename... Deps>
inline constexpr bool dependent_false = false;

// Or

template <bool B, typename... Deps>
inline constexpr bool dependent_bool_value = B;

[wg21bot]

P1830R1 std::dependent_false (Ruslan Arutyunyan)

[sempuki]

Belfast 2019-11 LEWGI Minutes

P1830R1 std::dependent_false: Design Review

Chair: Ryan McDougall

Champion: Ruslan Arutyunyan

Minute Taker: CJ Johnson

Start Review: 6 10:30

Start Polling: 11:10

POLL: We believe there is a library solution that will sufficiently (ease of use, failure prone) solve the problems described in P1830.

Strongly For	Weakly For	Neutral	Weakly Against	Strongly Against
1	3	2	2	1

NO CONSENSUS

Attendance: 17

SF: unlikely to get a language solution

POLL: We are willing to live with the solution presented in P1830R1 if no better library solution can be found.

Strongly For	Weakly For	Neutral	Weakly Against	Strongly Against
2	5	5	0	0

CONSENSUS

Attendance: 17

End: 11:24

We need more information.

[brycelelbach]

Belfast 2019-11 LEWGI Minutes

P1830R0 std::dependent_false: Direction Review

Chair: Ryan McDougall

Champion: Corentin Jabot

Minute Taker: CJ Johnson

Start: 11-06 10:37

End: 11:23

CONSENSUS: Bring a revision of P1882R0 (std::dependent_false), with the guidance below, to LEWGI for further direction review.

• Explore a consteval function that always returns false as a possible approach.

[mserdarsanli]

Last link points to a different paper

[jensmaurer]

@brycelelbach , there is some confusion with your latest comment here (from Nov 2019), confusing P1830 with P1882. (Also, we had direction review for P1830R0 in Cologne, so why do we have it again in Belfast?)

[cor3ntin]

@jensmaurer To add to the confusion, I remember championing P1882 (poorly) but not P1830

[dfrib]

P2593R0 (#1251) explores a new language solution to this problem.

[jfbastien]

@brycelelbach please assign to EWG once ready.

[jensmaurer]

P2593 was CWG-approved in Issaquah and is on track to be plenary-approved for C++26 in Varna. With that development, I think the motivation for P1830 has evaporated.

[cor3ntin]

@jensmaurer agreed, I'm tentatively closing this, the author can contact me or Bryce if they think this paper is still motivated.