https://github.com/Endilll created https://github.com/llvm/llvm-project/pull/121654
This patch adds test for [CWG156](https://cplusplus.github.io/CWG/issues/156.html). The most relevant piece of current wording is [[basic.lookup.unqual]/5](https://eel.is/c++draft/basic.lookup#unqual-5): > An unqualified name that is a component name > ([[expr.prim.id.unqual]](https://eel.is/c++draft/expr.prim.id.unqual)) of a > [type-specifier](https://eel.is/c++draft/dcl.type.general#nt:type-specifier) > or [ptr-operator](https://eel.is/c++draft/dcl.decl.general#nt:ptr-operator) > of a > [conversion-type-id](https://eel.is/c++draft/class.conv.fct#nt:conversion-type-id) > is looked up in the same fashion as the > [conversion-function-id](https://eel.is/c++draft/class.conv.fct#nt:conversion-function-id) > in which it > appears[.](https://eel.is/c++draft/basic.lookup#unqual-5.sentence-1) If that lookup finds nothing, it undergoes unqualified name lookup; in each case, only names that denote types or templates whose specializations are types are considered[.](https://eel.is/c++draft/basic.lookup#unqual-5.sentence-2) Per resolution of [CWG1111](https://cplusplus.github.io/CWG/issues/1111.html), additional lookup in the context of the entire postfix-expression, which originally was intended to cross-check lookup in the context of object-expression, was effectively turned into a fallback for it. Check out "Calling a conversion function" example in [P1787R6](https://wg21.link/p1787r6) for step-by-step explanation of the current lookup mechanics for conversion functions. Clang rejects one of the well-formed examples, hence partial status. Clang is the only implementation which rejects it: https://godbolt.org/z/ohhbx8Mfs >From dcd29ca8c77e24c532ca8300a7e46f5498ffebbb Mon Sep 17 00:00:00 2001 From: Vlad Serebrennikov <serebrennikov.vladis...@gmail.com> Date: Sat, 4 Jan 2025 19:37:46 +0300 Subject: [PATCH] [clang] Add test for CWG156 "Name lookup for conversion functions" --- clang/test/CXX/drs/cwg1xx.cpp | 43 +++++++++++++++++++++++++++++++++++ clang/www/cxx_dr_status.html | 2 +- 2 files changed, 44 insertions(+), 1 deletion(-) diff --git a/clang/test/CXX/drs/cwg1xx.cpp b/clang/test/CXX/drs/cwg1xx.cpp index 6aec8b65c91f12..eddad2e6a87b00 100644 --- a/clang/test/CXX/drs/cwg1xx.cpp +++ b/clang/test/CXX/drs/cwg1xx.cpp @@ -922,6 +922,49 @@ namespace cwg155 { // cwg155: dup 632 // expected-warning@-1 {{braces around scalar initializer}} } +namespace cwg156 { // cwg156: partial +namespace ex1 { +struct A { + operator int(); +} a; +void foo() { + typedef int T; + a.operator T(); // T is found using unqualified lookup + // after qualified lookup in A fails. +} +} // namespace ex1 + +namespace ex2 { +struct A { + typedef int T; + operator T(); +}; +struct B : A { + operator T(); +} b; +void foo() { + b.A::operator T(); // FIXME: qualified lookup should find T in A. + // expected-error@-1 {{unknown type name 'T'}} +} +} // namespace ex2 + +namespace ex3 { +template <class T1> struct A { + operator T1(); +}; +template <class T2> struct B : A<T2> { + operator T2(); + void foo() { + // In both cases, during instantiation, qualified lookup for T2 wouldn't be able + // to find anything, so T2 has to be found by unqualified lookup. + // After that, 'operator T2()' is found in A<T2> by qualfied lookup. + T2 a = A<T2>::operator T2(); + T2 b = ((A<T2> *)this)->operator T2(); + } +}; +} // namespace ex3 +} // namespace cwg156 + // cwg158 is in cwg158.cpp namespace cwg159 { // cwg159: 3.5 diff --git a/clang/www/cxx_dr_status.html b/clang/www/cxx_dr_status.html index c069e155fd547c..bbdca49aad0533 100755 --- a/clang/www/cxx_dr_status.html +++ b/clang/www/cxx_dr_status.html @@ -981,7 +981,7 @@ <h2 id="cxxdr">C++ defect report implementation status</h2> <td><a href="https://cplusplus.github.io/CWG/issues/156.html">156</a></td> <td>NAD</td> <td>Name lookup for conversion functions</td> - <td class="unknown" align="center">Unknown</td> + <td class="partial" align="center">Partial</td> </tr> <tr class="open" id="157"> <td><a href="https://cplusplus.github.io/CWG/issues/157.html">157</a></td> _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits