https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64666
Harald van Dijk <harald at gigawatt dot nl> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |harald at gigawatt dot nl
--- Comment #3 from Harald van Dijk <harald at gigawatt dot nl> ---
Strictly speaking, this affects conformance because it causes GCC to wrongly
accept code without any diagnostic, in conforming modes (e.g. -std=c++14
-pedantic), where the standard requires a diagnostic for the syntax error:
struct S { constexpr int operator=(int) { return 1; } };
int main() { int a[S() = 1]; }
Sun C++ and Intel also accept it without any diagnostic. clang rejects it as a
syntax error.
Rejecting it as a syntax error when it was previously accepted seems like a bad
idea, though. Perhaps an extra warning if -pedantic is used would be
appropriate?
