P3663 R1 Future-proof `submdspan-mapping`

[wg21bot]

P3663R0 Future-proof submdspan-mapping? (Mark Hoemmen)

[inbal2l]

2025-04-15 Library Evolution Telecon

P3663R0: Future-proof submdspan_mapping?

2025-04-15 Library Evolution Hagenberg Minutes

Champion: Mark Hoemmen
Chair: Inbal Levi
Minute Taker: Guy Davidson, Andreas Weis

Summary

POLL: Change section 5.8.3 “Requirements” to add under section 3 the ability to accept “std::tuple” and “std::pair”.

SF	F	N	A	SA
1	2	4	2	0

Attendance: 19
Author's Position: N
Outcome: No consensus for a change

POLL: Apply the changes above, bump the feature test macro and forward to LWG for C++29 (and with recommendation to apply as a DR for 26).

SF	F	N	A	SA
3	5	3	1	0

Attendance: 20
Author's Position: F
Outcome: Consensus in favor

POLL: Apply the changes above, bump the FTM and forward to LWG for C++26.

SF	F	N	A	SA
4	6	2	1	0

Attendance: 18
Author's Position: F
Outcome: Consensus in favor

A: The magic word performance was mentioned, but no measurements were provided. My concerns would go away if those measurements were provided.

ACTION: Add a section for performance evaluation.

Next Steps

he paper will be updated to include the changes, the updated recision (P3663R1) was forwarded to LWG for C++29 (and with recommendation for a DR) and as a C++26 fix if possible.

[wg21bot]

P3663R1 Future-proof submdspan-mapping (Mark Hoemmen)