Date: Thu, 5 Aug 2021 16:55:37 -0700
On 8/3/2021 11:23 AM, Peter C++ via Std-Proposals wrote:
> As the author of the paper with unique_resource and scope guards, I believe that unique_resource has some merit, but that scope guards (as in LFTS) are a design error, because using one means one has forgotten an abstraction. so any (generic) scope guard is a license to "cheat" C++ clean resource management.
>
> Also I burnt out on that paper snd therefore refrained from proposing it any further. Feel free, to pick it up (the unique_resource part), I won't.
Do you (or does someone else) know where to see the current
status/issues with unique_resource? The closest I could find was
p1411r0[1], but perhaps I'm missing something.
I have a slightly-modified implementation of unique_resource that I've
used to great effect and I wouldn't mind trying to get this across the
finish line, provided I know all what's holding it up.
- Jim
[1]: http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2019/p1411r0.pdf
> As the author of the paper with unique_resource and scope guards, I believe that unique_resource has some merit, but that scope guards (as in LFTS) are a design error, because using one means one has forgotten an abstraction. so any (generic) scope guard is a license to "cheat" C++ clean resource management.
>
> Also I burnt out on that paper snd therefore refrained from proposing it any further. Feel free, to pick it up (the unique_resource part), I won't.
Do you (or does someone else) know where to see the current
status/issues with unique_resource? The closest I could find was
p1411r0[1], but perhaps I'm missing something.
I have a slightly-modified implementation of unique_resource that I've
used to great effect and I wouldn't mind trying to get this across the
finish line, provided I know all what's holding it up.
- Jim
[1]: http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2019/p1411r0.pdf
Received on 2021-08-05 19:00:17