diff options
author | Marek Polacek <polacek@redhat.com> | 2020-10-05 17:48:19 -0400 |
---|---|---|
committer | Marek Polacek <polacek@redhat.com> | 2020-10-06 17:35:30 -0400 |
commit | 85307b4e938d42201d6c232f5d9259f91133a303 (patch) | |
tree | 21d685bef3085640fec38d61c6077ac012f15e15 /libgomp | |
parent | a9802204603616df14ed47d05f1b86f1bd08d8fb (diff) | |
download | gcc-85307b4e938d42201d6c232f5d9259f91133a303.zip gcc-85307b4e938d42201d6c232f5d9259f91133a303.tar.gz gcc-85307b4e938d42201d6c232f5d9259f91133a303.tar.bz2 |
c++: typename in out-of-class member function definitions [PR97297]
I was notified that our P0634R3 (Down with typename) implementation has
a flaw: when we have an out-of-class member function definition, we
still required 'typename' for its parameters. For example here:
template <typename T> struct S {
int simple(T::type);
};
template <typename T>
int S<T>::simple(/* typename */T::type) { return 0; }
the 'typename' isn't necessary per [temp.res]/5.2.4. We have a qualified
name here ("S<T>::simple") so we know it's already been declared so we
can look it up to see if it's a function template or a variable
template.
In this case, the P0634R3 code in cp_parser_direct_declarator wasn't
looking into uninstantiated templates and didn't find the member
function 'simple' -- cp_parser_lookup_name returned a SCOPE_REF which
means that the qualifying scope was dependent. With this fix, we find
the BASELINK for 'simple', don't clear CP_PARSER_FLAGS_TYPENAME_OPTIONAL
from the flags, and the typename is implicitly assumed.
gcc/cp/ChangeLog:
PR c++/97297
* parser.c (cp_parser_direct_declarator): When checking if a
name is a function template declaration for the P0634R3 case,
look in uninstantiated templates too.
gcc/testsuite/ChangeLog:
PR c++/97297
* g++.dg/cpp2a/typename18.C: New test.
Diffstat (limited to 'libgomp')
0 files changed, 0 insertions, 0 deletions