P3709 R2 Reconsider parallel ranges::rotate_copy and ranges::reverse_copy

[wg21bot]

P3709R0 Reconsider parallel ranges::rotate_copy and ranges::reverse_copy (Ruslan Arutyunyan, Alexey Kukanov)

[foonathan]

I'd like to see this in SG9 first, as the use of the in_in_result types for unary range algorithms is novel.

[inbal2l]

(SG9 have seen this paper first, minutes will be posted)

---

2025-06-16 LEWG Sofia Meeting

P3709R1: Reconsider parallel ranges::rotate_copy and ranges::reverse_copy

2025-06-16 Library Evolution Sofia Minutes

Champion: Ruslan Arutyunyan
Chair: Fabio Fracassi / Andreas Weis
Minute Taker: Robert Leahy

Summary

POLL: We want to change the return type of “rotate_copy” and “reverse_copy” to something similar to in_in_out

SF	F	N	A	SA
3	7	2	0	0

Attendance: 22 (IP) + 4 (R)
Author's Position: SF
Outcome: Consensus in favor

POLL: In P3709R1 rotate_copy and reverse_copy should have a dedicated return type instead of re-using in_in_out_result

SF	F	N	A	SA
7	5	2	2	0

Attendance: 22 (IP) + 5 (R)
Author's Position: A
Outcome: Consensus in favor

Next Steps

LEWG approved to change “rotate_copy” and “reverse_copy”’s return types to something in sme structure as std::in_in_out (but a new utility with a different name and different members' names). Author should come back with a new revision that have such a proposed type (or types), to present in LEWG.