From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 2122) id F0B2238515D7; Wed, 26 Jan 2022 04:23:13 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org F0B2238515D7 MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="utf-8" From: Jason Merrill To: gcc-cvs@gcc.gnu.org Subject: [gcc r12-6872] c++: alias template and typename [PR103057] X-Act-Checkin: gcc X-Git-Author: Jason Merrill X-Git-Refname: refs/heads/master X-Git-Oldrev: e0b8716f53ed6455e9f18931940141692793068d X-Git-Newrev: f3e6ef7d873885ffaa0db7f4185364bbd7a70505 Message-Id: <20220126042313.F0B2238515D7@sourceware.org> Date: Wed, 26 Jan 2022 04:23:13 +0000 (GMT) X-BeenThere: gcc-cvs@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-cvs mailing list List-Unsubscribe: , List-Archive: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 26 Jan 2022 04:23:14 -0000 https://gcc.gnu.org/g:f3e6ef7d873885ffaa0db7f4185364bbd7a70505 commit r12-6872-gf3e6ef7d873885ffaa0db7f4185364bbd7a70505 Author: Jason Merrill Date: Tue Jan 25 15:12:16 2022 -0500 c++: alias template and typename [PR103057] Usually we handle DR1558 substitution near the top of tsubst, but in this case while substituting TYPENAME_TYPE we were passing an alias specialization to tsubst_aggr_type, which ignored its aliasness. PR c++/103057 gcc/cp/ChangeLog: * pt.cc (tsubst_aggr_type): Call tsubst for alias template specialization. gcc/testsuite/ChangeLog: * g++.dg/cpp0x/alias-decl-void1.C: New test. Diff: --- gcc/cp/pt.cc | 8 ++++++++ gcc/testsuite/g++.dg/cpp0x/alias-decl-void1.C | 18 ++++++++++++++++++ 2 files changed, 26 insertions(+) diff --git a/gcc/cp/pt.cc b/gcc/cp/pt.cc index 8f50b9c4d3c..6fbda676527 100644 --- a/gcc/cp/pt.cc +++ b/gcc/cp/pt.cc @@ -13585,6 +13585,14 @@ tsubst_aggr_type (tree t, if (t == NULL_TREE) return NULL_TREE; + /* If T is an alias template specialization, we want to substitute that + rather than strip it, especially if it's dependent_alias_template_spec_p. + It should be OK not to handle entering_scope in this case, since + DECL_CONTEXT will never be an alias template specialization. We only get + here with an alias when tsubst calls us for TYPENAME_TYPE. */ + if (alias_template_specialization_p (t, nt_transparent)) + return tsubst (t, args, complain, in_decl); + switch (TREE_CODE (t)) { case RECORD_TYPE: diff --git a/gcc/testsuite/g++.dg/cpp0x/alias-decl-void1.C b/gcc/testsuite/g++.dg/cpp0x/alias-decl-void1.C new file mode 100644 index 00000000000..accc1a45abc --- /dev/null +++ b/gcc/testsuite/g++.dg/cpp0x/alias-decl-void1.C @@ -0,0 +1,18 @@ +// PR c++/103057 +// { dg-do compile { target c++11 } } + +template struct A { }; +template struct B { using type = A; }; +template struct C { + using type = typename T::foo; // { dg-error "int" } +}; +template using L = B; + +template +typename L::type>::type +f(T) { }; + +int main() +{ + f(42); // { dg-error "no match" } +}