P1701 Inline Namespaces: Fragility Bites CWG2505

[wg21bot]

P1701R0 Inline Namespaces: Fragility Bites (Nathan Sidwell)

[jensmaurer]

EWG in Cologne

[jfbastien]

I talked to Nathan in person, this needs revision.

[wg21bot]

P1701R1 Inline Namespaces: Fragility Bites (Nathan Sidwell)

[jfbastien]

This was discussed in today's EWG telecon.
My intent is to see P1701r1 again, and poll "revert DR2061" is tentatively ready to be sent to Core. No poll taken today, Nathan received feedback, we'll get back to this.

[wg21bot]

P1701R2 Inline Namespaces: Fragility Bites (Nathan Sidwell)

[jfbastien]

This was seen in today's EWG telecon.

POLL: The problem exposed in 2.1 PR90291 (which is the same as 2.3 Modules) is one that we wish to address with a language change.

inline namespace A {
 namespace detail { // #1
 void foo() {} // #3
 }
}
namespace detail { // #2
 inline namespace C {
 void bar() {} // #4
 }
}

SF	F	N	A	SA
0	2	3	2	0

Result: No consensus for a change.

POLL: The problem exposed in 2.2 Unnamed Namespaces is one that we wish to address with a clarification of the Standard’s intent.

namespace {}
inline namespace bob {
 namespace {}
}
namespace {} // error, ambiguous

SF	F	N	A	SA
0	7	1	0	0

Result: Consensus.

POLL: Direct Core to clarify the Standard’s intent in 2.2 Unnamed Namespaces by adopting P1701r2’s 5.2 Option B. Resolve this as a Core issue, apply as a DR.

SF	F	N	A	SA
2	6	0	0	0

Result: Consensus.

Process: Nathan to talk to Mike, will open a Core issue with the proposed resolution. No need for EWG electronic polling, but Nathan should email EWG once the Core issue is open.

[jfbastien]

Moving the paper to Core to track the issue that will be open. Once the issue is open, I believe that the issue can be closed as the paper will see no further progress.

[jensmaurer]

CWG telecon 2021-12-09: CWG2505 considered; author is pondering updates.