Outline migration path from tag_invoke for codebases that have a homebrew implementation / use libunifex

[atomgalaxy]

[lewissbaker]

Not that having a generic override like this will require having some way to find that overload by adl. Currently we don't consider the cpo type itself as an associated entity but maybe we should.

This may also be useful for types to customise foo for their type T even if they don't have any args in their namespace. Eg parse<T>(std::string_view)

[cor3ntin]

Currently we don't consider the cpo type itself as an associated entity but maybe we should.

That seems extremely purpose defeating

[lewissbaker]

Because of the extra namespace to search?

[cor3ntin]

Oh, i think i misunderstood. You want the generic override to be found in the cpo's NS rather than the cpo itself being found by adl, which scared me.
Then yes, I think it would make sense, maybe. definitively worth considering.

[atomgalaxy]

... right yeah we probably should consider it an associated entity, otherwise we'll have major issues with the transition.

…

On Fri, Feb 18, 2022 at 2:03 PM cor3ntin ***@***.***> wrote: Oh, i think i misunderstood. You want the generic override to be found in the cpo's NS rather than the cpo itself being found by adl, which scared me. Then yes, I think it would make sense, *maybe*. definitively worth considering. — Reply to this email directly, view it on GitHub <#20 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAA5R5LRLNUA73L4IBBHFT3U3ZGUZANCNFSM5OVU5HZQ> . Triage notifications on the go with GitHub Mobile for iOS <https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675> or Android <https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[h-vetinari]

Any update on the R2 effort after R1 was seen by EWG?

[atomgalaxy]

It's probably dead in favor of traits. I'm working on a medley of papers, that'll together with a paper from Lewis (adl injection) and Barry (pmfs should be callable/std::invoke in the language) should give us the ability to do the equivalent ease-of-use in the library.

…

On Sun, Mar 5, 2023 at 6:52 AM h-vetinari ***@***.***> wrote: Any update on the R2 effort after R1 was seen by EWG? — Reply to this email directly, view it on GitHub <#20 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAA5R5KZLB4DB2SIPYHGR23W2QZ25ANCNFSM5OVU5HZQ> . You are receiving this because you authored the thread.Message ID: ***@***.***>