https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88812
Jonathan Wakely <redi at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |ASSIGNED
Last reconfirmed| |2019-01-11
Assignee|unassigned at gcc dot gnu.org |redi at gcc dot gnu.org
Ever confirmed|0 |1
--- Comment #1 from Jonathan Wakely <redi at gcc dot gnu.org> ---
(In reply to emsr from comment #0)
> http://www.open-std.org/JTC1/SC22/WG21/docs/papers/2018/p0487r1.html
>
> TL;DR Just like deletion of gets() the operator>>(char* str) need to be
> replaced by ones that take sized buffers:
>
> #if __cplusplus > 201703L
> // Not in LWG2499/P0487R1
> template<std::size_t _Num>
> basic_istream<char>&
> operator>>(basic_istream<char>& __in, char (&__s)[_Num]);
>
> template<typename _Traits, std::size_t _Num>
> inline basic_istream<char, _Traits>&
> operator>>(basic_istream<char, _Traits>& __in, unsigned char
> (&__s)[_Num])
> { return (__in >> reinterpret_cast<char*>(__s)); }
This cast needs to be to char(&)[_Num] or it tries to call the old signature.
>
> template<typename _Traits, std::size_t _Num>
> inline basic_istream<char, _Traits>&
> operator>>(basic_istream<char, _Traits>& __in, signed char (&__s)[_Num])
> { return (__in >> reinterpret_cast<char*>(__s)); }
Same here.
> #endif
>
> No, I don't have a complete patch.
Coincidentally I just implemented this.
> We'll need to figure out migration deprecation strategies, ABI compatibility
> (keep old sigs in istream.cc, just don't declare publicly?).
I think we should keep the old declarations for C++98 - C++17 in the public
headers. We definitely need to keep exporting the definition from the library.
We can add the new declarations for C++2a only (and maybe for -std=gnu++NN
modes too, or expose them via some new macro, to allow users to either opt in
or opt out).
