Was seen by SG9 (Ranges) on 2022-03-14 (Full Minutes)

Polls

POLL: We support forwarding the change proposed in P2538R0 to LEWG.

Attendance: 11

# of Authors: 1

Author Position: SF

Outcome: Consensus in favor

Summary

Consensus for forwarding to LEWG. Author is aware that the change is unlikely to be ported as a DR, but that implementers will have the freedom to back-port.

Implementation experience in libcxx (in paper) and in STL by Casey Carter (https://github.com/CaseyCarter/STL/tree/p2538), all tests in STL pass normally.

A motion was made on april 4 2022 on the mailing list to send this paper to electronic polling https://lists.isocpp.org/lib-ext/2022/04/22898.php.

Motion passes with 10 votes.

P2538R1 ADL-proof std::projected (Arthur O'Dwyer, Casey Carter)

2022-05 Library Evolution Electronic Poll Outcomes

POLL: Send [P2538R0] ADL-Proof projected to Library Working Group for C++23, classified as an improvement of an existing feature ([P0592R4] bucket 2 item).

Strong consensus in favor.

After further discussion, we'll try and squeeze in another review of this in Library Evolution in time for C++23.

After even further discussion, the potential conflicts were only in P2248 (#944), so this paper doesn't need more Library Evolution attention.

2022-06-28 Library Evolution Telecon

P2248R4: Enabling list-initialization for algorithms

2022-06-28 Library Evolution Telecon Minutes

P2609R0: Relaxing Ranges Just A Smidge

2022-06-28 Library Evolution Telecon P2609 Minutes

P2538R1: ADL-proof projected

2022-06-28 Library Evolution Telecon P2538 Minutes

Chair: Bryce Adelstein Lelbach

Minute Taker: Ben Craig

Champion: Giuseppe D'Angelo (P2248) & John Eivind Helset (P2609)

Start P2248: 2022-06-28 11:47 Eastern

POLL: Modify P2248R4 (Enabling list-initialization for algorithms) by making projected_value a metafunction defined in terms of invoke_result_t, and return the revised paper to Library for C++23 classified as B2 - Improvement, to be confirmed with a Library Evolution electronic poll.

Attendance: 26

# of Authors: 1

Author Position: N

Outcome: Strong consensus in favor.

End: 11:55

POLL: Modify P2248R4 (Enabling list-initialization for algorithms) by making projected_value a metafunction defined in terms of invoke_result_t and extending the paper to ranges::fold and ranges::contains, and return the revised paper to Library for C++23 classified as B2 - Improvement, to be confirmed with a Library Evolution electronic poll.

Attendance: 26

# of Authors: 1

Author Position: WF

Outcome: Consensus in favor.

N: Too much work for C++23.

POLL: Modify P2248R4 (Enabling list-initialization for algorithms) by making projected_value a metafunction defined in terms of invoke_result_t and extending the paper to ranges::fold and ranges::contains, and return the revised paper to Library for C++23 classified as B2 - Improvement, to be confirmed with a Library Evolution electronic poll.

Attendance: 25

# of Authors: 1

Author Position: SF

Outcome: Strong consensus in favor.

We will go with the last poll.

End P2248: 12:03

Start P2609: 12:04

Does this paper have:

• Examples?
  • Yes.
• Field experience?
  • Implementation experience?
    • Yes, in ranges v3 and libstdc++.
• Wording?
  • Yes.
• Feature test macro?
  • No; we should modify the paper to bump the __cpp_lib_ranges feature
    test macro.

POLL: indirect_value_t should be exposition only.

Attendance: 25

# of Authors: 1

Author Position: N

Outcome: No consensus; we'll leave it to the author's discretion.

POLL: Modify P2609R0 (Relaxing Ranges Just A Smidge) by bumping the __cpp_lib_ranges feature test macro, and send the revised paper to Library for C++26 classified as B2 - Improvement, to be confirmed with a Library Evolution electronic poll.

Attendance: 24

# of Authors: 1

Author Position: SF

Outcome: Unanimous consensus in favor.

End: 12:37

If P2609 merges before P2248, P2248 should define projected_value in terms of indirect_value_t.

If P2248 merges before P2609, P2609 should update the definition of projected_value.

Summary

We discussed projected_value (from P2248) and indirect_value_t (from P2609), two metafunctions with a very similar definition.
We determined that we should try to ensure that they are defined in terms of each other, and that we wanted both to be concrete and not expositional.
We also decided that P2248 (Enabling list -initialization for algorithms) should be expanded to include two new algorithms that have been added since P2248 was forwarded, ranges::fold and ranges::contains.

Next Steps

• Produce and publish P2248R5 (Enabling list-initialization for algorithms), making projected_value a metafunction defined in terms of invoke_result_t and expanding the paper to include ranges::fold and ranges::contains
• Take an electronic poll to confirm sending P2248R5 (Enabling list-initialization for algorithms) to Library for C++23 classified as B2 - Improvement.
• Produce and publish P2609R1 (Relaxing Ranges Just A Smidge), bumping the __cpp_lib_ranges feature test macro.
• Take an electronic poll to confirm sending P2609R0 (Relaxing Ranges Just A Smidge) to Library for C++26 classified as B2 - Improvement.
• If P2609 merges before P2248, P2248 should define projected_value in terms of indirect_value_t.
• If P2248 merges before P2609, P2609 should update the definition of projected_value.

Correcting target label given the latest status

LWG reviewed and approved for C++26

https://wiki.edg.com/bin/view/Wg21telecons2023/P2538-20230419

Poll: Put P2538R1 in c++26