Re: [multi] Formal Review Begins

Andrzej Krzemienski via Boost <[email protected]>
Newsgroups gmane.comp.lib.boost.devel
Message-ID <CAOenAXhDXhQAObWPPJ5_a0guuoWcriCRghKrMSNrPiVwknpVTA@mail.gmail.com>
niedz., 8 mar 2026 o 05:35 Amlal El Mahrouss via Boost <
[email protected]> napisał(a):

> Hello Seth,
> Yes `multi::extensions_t` and `a.extensions()` does read as off.
> But Boost.Multi has already a container called 'extents' (
> https://github.com/correaa/boost-multi/blob/b7f4dcf08b18628e1ae7aba9f44f71246ac45c06/include/boost/multi/detail/extents.hpp
> )
> Which does make use of `extensions_type`.
>
> >From what I understand so far about the library's design,
> `extensions_type serves as an underlying type of the `extents` type, which
> could explain the odd naming.


I agree with Seth's position here. The correct English word for the concept
is `extents`, and this is what the users should see. There is even no clash
with using the same name for the public type and the member function:

multi::extents ext = arr.extents();

Boost.Multi has a PR issue for that:
https://github.com/correaa/boost-multi/issues/127

If the library needs a "private detail" type for being a base of something,
because this is a "private detail", it can afford to have a longer name,
like "basic_extents".

BTW, the above ticket also shows how the different initialization syntax
choices can lead to bugs.

Regards,
&rzej;
_______________________________________________
Boost mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://lists.boost.org/mailman3/lists/boost.lists.boost.org/
Archived at: https://lists.boost.org/archives/list/[email protected]/message/JP4GN4S7X3BRXBKVQ6K3X6FUM3CXYRBZ/
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.