[Bug libstdc++/114336] New: Manual should document implementation-defined behaviour in C++20 and C++23

[Bug libstdc++/114336] New: Manual should document implementation-defined behaviour in C++20 and C++23

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114336

            Bug ID: 114336
           Summary: Manual should document implementation-defined
                    behaviour in C++20 and C++23
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Keywords: documentation
          Severity: normal
          Priority: P3
         Component: libstdc++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: redi at gcc dot gnu.org
            Blocks: 88322
  Target Milestone: ---

Similar to
https://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html#iso.2017.specific
and the equivalent for earlier standards, we need to document all
implementation-defined choices in the C++20 and C++23 features.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88322
[Bug 88322] Implement C++20 library features.

[Bug libstdc++/114336] Manual should document implementation-defined behaviour in C++20 and C++23

[de34 at live dot cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114336

Jiang An <de34 at live dot cn> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |de34 at live dot cn

--- Comment #1 from Jiang An <de34 at live dot cn> ---
IMO the column number obtained from std::source_location::current is somehow
curious and might be critical. Implementations behaves differently, and
currently there doesn't seem any official documentation for this value.

[Bug libstdc++/114336] Manual should document implementation-defined behaviour in C++20 and C++23

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114336

--- Comment #2 from Jonathan Wakely <redi at gcc dot gnu.org> ---
I don't see how it can be critical, since it might just be 0 always, so it's
not a good idea to write code where that impl-defined value is critical.

For G++ the column number is the opening paren of
std::current_location::current() which doesn't seem curious to me.

We could subtract strlen("std::source_location::current") from that, but that
would give a weird value for weirdos doing weird things like:

auto loc = std::
  source_location::
  current
  ();

Since the line() value is always going to be the location of the () I think it
would be strange if the column was a different location.

But yes, this is one of the things that should be documented for conformance to
[defns.impl.defined].