Date: Thu, 15 Sep 2022 23:55:21 +0200
On 14/09/2022 20.07, Tom Honermann via SG16 wrote:
> On 9/14/22 10:43 AM, Victor Zverovich via SG16 wrote:
>> > The wcwidth implementation does use East Asian width. No magic there.
>>
>> Sure.
>>
>> > *Adds it to the pile of NB comments*
>>
>> I don't think it can be an NB comment because width estimation was added in C++20.
>
> I believe NB comments may cover any aspect of the standard regardless of when the relevant text was introduced (and doing so is appropriate because times change; e.g., updates needed for newer Unicode versions).
Yes.
However, an NB comment that says "change C++20 to say X instead"
is, of course, ill-formed. The NB comment always (implicitly)
says "C++23 should say X instead of what the CD says".
Jens
> On 9/14/22 10:43 AM, Victor Zverovich via SG16 wrote:
>> > The wcwidth implementation does use East Asian width. No magic there.
>>
>> Sure.
>>
>> > *Adds it to the pile of NB comments*
>>
>> I don't think it can be an NB comment because width estimation was added in C++20.
>
> I believe NB comments may cover any aspect of the standard regardless of when the relevant text was introduced (and doing so is appropriate because times change; e.g., updates needed for newer Unicode versions).
Yes.
However, an NB comment that says "change C++20 to say X instead"
is, of course, ill-formed. The NB comment always (implicitly)
says "C++23 should say X instead of what the CD says".
Jens
Received on 2022-09-15 21:55:26