This is the mail archive of the
gcc-prs@gcc.gnu.org
mailing list for the GCC project.
c++/9410: [3.4 regression] inability to access a base classes' member from a templated derived class
- From: andrew at andypo dot net
- To: gcc-gnats at gcc dot gnu dot org
- Date: 22 Jan 2003 23:04:36 -0000
- Subject: c++/9410: [3.4 regression] inability to access a base classes' member from a templated derived class
- Reply-to: andrew at andypo dot net
>Number: 9410
>Category: c++
>Synopsis: [3.4 regression] inability to access a base classes' member from a templated derived class
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: rejects-legal
>Submitter-Id: net
>Arrival-Date: Wed Jan 22 23:06:00 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator: Andrew Pollard
>Release: gcc-3.4-20030122
>Organization:
>Environment:
i686-pc-linux-gnu on RedHat8.0/PIII
>Description:
The following snippet of code no longer compiles.
I'm fairly certain it should compile (but I'm not a language laywer).
Previous versions of gcc compile it (including gcc-3.4-20030116)
hidden.cxx:
-----------
struct A {
int a;
};
template<typename T>
struct B : public A {
};
template<typename T>
struct C : public B<T> {
int foo() { return (a); }
int bar() { return (this->a); }
};
----------
% g++34 hidden.cxx -c
hidden.cxx: In member function `int C<T>::foo()':
hidden.cxx:11: error: `a' has not been declared
Note how the use with (this->a) works. From a language point of view I would have expected both to work or both to fail.
Is this a bug, or just another inability of mine to understand the complexity of templates in C++?
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: