[config] fix for BCB and BOOST_NO_MEMBER_TEMPLATE_FRIENDS

I believe config macro BOOST_NO_MEMBER_TEMPLATE_FRIENDS should be defined for BCB 6. Currently it is defined only for BCB 5.5.1 and less but with BCB 6: - compiling test file boost_no_mem_templ_frnds.cxx fails, - the problem shows in Serialization library too. So I recommend to change borland.hpp to define this macro for BCB 6 (and likely for Kylix which is simply BCB6 ported to Linux). BCB6 patch 4 == 0x564 Kylix = 0x570 /Pavel

I believe config macro BOOST_NO_MEMBER_TEMPLATE_FRIENDS should be defined for BCB 6.
Currently it is defined only for BCB 5.5.1 and less but with BCB 6:
No I don't think it's defined for either.
- compiling test file boost_no_mem_templ_frnds.cxx fails,
Works for me, both with bcb5 and bcb6.
- the problem shows in Serialization library too.
Can you provide a test case? Thanks, John.

"John Maddock" <john@johnmaddock.co.uk> wrote
I believe config macro BOOST_NO_MEMBER_TEMPLATE_FRIENDS should be defined for BCB 6.
- compiling test file boost_no_mem_templ_frnds.cxx fails,
Works for me, both with bcb5 and bcb6.
Oops, my mistake. Nevertheless I bellow is code snippet which fails on BCB 6 (and VC6) and compiles on Intel C++ 7 and Comeau online: ------------------------ template <class T> class foo; template <class T> bool must_be_friend_proc(const foo<T>& f); template <class T> class foo { private: template<typename Y> friend bool must_be_friend_proc(const foo<Y>& f); int i; public: foo(){ i = 0; } void call_friend() { must_be_friend_proc<T>(*this); } }; template <class T> bool must_be_friend_proc(const foo<T>& f) { return f.i != 0; } int main(int argc, char* argv[]) { foo<int> fi; fi.call_friend(); return 0; } ------------------------ The difference to file boost_no_mem_templ_frnds.cxx is that I actually call friend. /Pavel

Oops, my mistake. Nevertheless I bellow is code snippet which fails on BCB 6 (and VC6) and compiles on Intel C++ 7 and Comeau online:
The difference to file boost_no_mem_templ_frnds.cxx is that I actually call friend.
So it is, OK both the config test case, and the Borland config have been updated accordingly. Thanks, John.
participants (2)
-
John Maddock
-
Pavel Vozenilek