Oh, thats
 https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2022/p2726r0.html
<https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2022/p2726r0.html>

Hewill Kang <hewillk_at_[hidden]> 於 2025年11月17日 週一 下午3:37寫道：

> This can now be easily done by providing an `operator[] ` that takes an
> `integral-constant-like` for `tuple` :
>
> std::tuple t{'a', 0.42, 100};
> t[std::cw<0>] = 'b';
> t[std::cw<1>] = 0.0;
> t[std::cw<2>] = 5;
>
> And the implementation can be simplified by duducing this.
> Therefore, is providing an operator[] for a tuple more desirable feature
> than the this original proposal?
>
>
> Jonathan Wakely via Std-Proposals <std-proposals_at_[hidden]> 於
> 2025年11月17日 週一 下午2:30寫道：
>
>>
>>
>> On Sun, 16 Nov 2025, 17:18 Bjorn Reese via Std-Proposals, <
>> std-proposals_at_[hidden]> wrote:
>>
>>> On 11/16/25 16:42, Hewill Kang via Std-Proposals wrote:
>>>
>>> > I think this might be a bit far-fetched, because it changes the
>>> > long-standing expected use case of operator[]. More precisely, it
>>> > provides a more complex overload for operator[]. I don't know if it's
>>> > worth it.
>>>
>>> Recall that we already have a special syntax with array bound checks for
>>> tuple-like classes:
>>>
>>> std::array<int, 4> a{};
>>> auto v = std::get<42>(a);
>>>
>>
>>
>> Yes, but if we could have supported accessing tuples as tup[N] I think
>> most people would have preferred that!
>>
>> The get<N> syntax is a less-than-ideal hack.
>>
>>
>> --
>> Std-Proposals mailing list
>> Std-Proposals_at_[hidden]
>> https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals
>>
>