David Abrahams <dave@boost-consulting.com> writes:

This failure: http://engineering.meta-comm.com/boost-regression/CVS-HEAD/developer/output/...

Appears to be due to vc7.1 badly misinterpreting a using-directive in the scope of the instantiating function. I suggest full qualification of ::boost::detail in shared_ptr.hpp as a remedy.

Correction about what stimulated the bug to appear: I now believe it was the following sequence in boost/graph/detail/read_graphviz_spirit.hpp using namespace std; using namespace boost; using namespace boost::spirit; using namespace phoenix; The amount of ambiguity being courted by bringing all those namespaces into a subnamespace of boost::detail (!) would scare me away from doing it, especially in a header file. But maybe that's just me. -- Dave Abrahams Boost Consulting www.boost-consulting.com

David Abrahams wrote:

This failure: http://engineering.meta-comm.com/boost-regression/CVS-HEAD/developer/output/...

Appears to be due to vc7.1 badly misinterpreting a using-directive in the scope of the instantiating function. I suggest full qualification of ::boost::detail in shared_ptr.hpp as a remedy.

Most 'detail's in shared_ptr.hpp were already fully-qualified, but since there is no shared_ptr test covering this case, I've managed to introduce two unqualified instances. Which leads me to think that my first reaction of "does anyone have a simple test case" was correct. Unfortunately producing a simple test for this problem doesn't seem easy.