On Jan 18, 2023, at 12:27 PM, Arthur O'Dwyer <arthur.j.odwyer@gmail.com> wrote:

Hi Robert,

This would be what we'd call an "editorial change" — one with no functional/behavioral change, just fiddling with the wording (hopefully for the better).

You've identified an inconsistency in style between

https://eel.is/c++draft/string.view.template.general

and

https://eel.is/c++draft/basic.string.general

https://eel.is/c++draft/vector.overview

https://eel.is/c++draft/deque.overview

https://eel.is/c++draft/flat.set.defn

etc. etc.

I agree with you that this is worth fixing — not primarily because the wording in [string.view] is confusing to the reader (as Jason said, there's really only one interpretation that makes sense), but rather because the wording in [string.view] is inconsistent with the established practice in all those other sections. Deviations from the best practice ought to have a reason; and in this case I don't think there is a valid reason — it's just gratuitously inconsistent.

Editorial changes — unless unusually large, subtle, or controversial — are generally handled via pull requests, not papers.

So your next step would be to clone https://github.com/cplusplus/draft , grep for all instances of `using reverse_iterator =`, find all the inconsistent instances, make a commit that fixes them all at once, and then pull-request it (N.B.: humbly ;) and including a link to this discussion).  I think that'd be a good idea, and I'm willing to help with it if you need help. Unless of course someone shows the reason why [string.view] is inconsistent — it's possible there is a reason and I'm just not seeing it.

N.B.: Pull requests are the appropriate venue for editorial (non-behavioral) changes. Pull requests are not an appropriate mechanism for feature requests or behavioral changes of any kind.

–Arthur

On Wed, Jan 18, 2023 at 12:08 PM Robert Allan Schwartz via Std-Proposals <std-proposals@lists.isocpp.org> wrote:

Hi Jason,

You're correct, there is nothing that *needs* to be changed.

I was only proposing an improvement for purposes of clarity, and to prevent a potential error if a future edition of the Standard mistakenly reverses the order of those two lines.

Thanks,

Robert

> On Jan 18, 2023, at 12:03 PM, Jason McKesson via Std-Proposals <std-proposals@lists.isocpp.org> wrote:
>
> On Wed, Jan 18, 2023 at 11:54 AM Robert Allan Schwartz via
> Std-Proposals <std-proposals@lists.isocpp.org> wrote:
>>
>> Hello,
>>
>> Today I am making my first posts to this group. Please forgive me if I make mistakes. Thank you.
>>
>> Here is the context:
>>
>> Header <string_view>:
>>
>> template<class charT, class traits = char_traits>
>> class basic_string_view {
>> public:
>> // types
>> using traits_type = traits;
>> using value_type = charT;
>> using pointer = value_type*;
>> using const_pointer = const value_type*;
>> using reference = value_type&;
>> using const_reference = const value_type&;
>> using const_iterator = implementation-defined ; // see 24.4.2.2
>> using iterator = const_iterator;228
>> using const_reverse_iterator = reverse_iterator<const_iterator>;
>> using reverse_iterator = const_reverse_iterator;
>>
>> I understand that the alias declaration for const_reverse_iterator uses the only declaration of reverse_iterator that is in scope at that moment, which is std::reverse_iterator, but then reverse_iterator is defined again on the next line, which creates an ordering dependency.
>>
>> Wouldn't it be safer and clearer to say:
>>
>> using const_reverse_iterator = std::reverse_iterator<const_iterator>;
>>
>> This would eliminate the ordering issue.
>
> There is no ordering issue. Because `basic_string_view` exists inside
> the `std` namespace, `reverse_iterator` will refer to the namespace
> declaration. There is nothing that needs to be changed.
> --
> Std-Proposals mailing list
> Std-Proposals@lists.isocpp.org
> https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals

--
Std-Proposals mailing list
Std-Proposals@lists.isocpp.org
https://lists.isocpp.org/mailman/listinfo.cgi/std-proposals