Is there any plan for standardizing `local_shared_ptr`?

[frederick-vs-ja]

I'm sorry for bothering here. This is not an issue for boost::local_shared_ptr.

Arthur O'Dwyer said that local_shared_ptr has been rejected for standardization (on std-proposals). But it seems that there's not even a P-paper for it.

I wonder whether the standardization is really rejected, and if not, whether there is any plan for it.

[pdimov]

I'm not aware of any attempts to propose local_shared_ptr for standardization. I'm not sure what Arthur was referring to. @Quuxplusone?

[Quuxplusone]

@pdimov: You think I remember what I was thinking in a parenthetical from two years ago? ;) Searching for "non-thread-safe shared_ptr" turns up some old discussions like

• https://groups.google.com/a/isocpp.org/g/std-proposals/c/LNaFSlj_Y-0 (2015, Ville seems amenable)
• non thread-safe shared_ptr WG21-SG14/SG14#59 (2016, @onqtam considered making a formal paper but didn't)
• https://stackoverflow.com/questions/6593770/creating-a-non-thread-safe-shared-ptr (no WG21-related discussion)

I might have been hazily remembering the 2015 and 2016 discussions, which were WG21-adjacent and didn't result in anything being adopted, but also weren't formal proposals either.

[pdimov]

Reading that SO link reminds me that I need to find the time to implement local_weak_ptr one of these days.