• "Matthew" <matthew hat.stlsoft.dot.org> Dec 26 2005
• "Walter Bright" <newshound digitalmars.com> Dec 26 2005

DMC++ 8.45 has an ICE when asked to compile code using the non-member !=
operator for stlsoft::pod_vector. 8.44 doesn't.

This is reminiscent of a behaviour I've seen with many "older" compilers,
whereby non-member (in)equality operators of class templates are not
properly resolved, leading to the need to workaround by conditionally
defining the operators as member functions. The surprise here is that 8.44
likes it, and 8.45 doesn't and _appears_ to have regressed.

I'm afraid I haven't time to pare it down, but just wanted to give you a
heads up on the off-chance it might provoke a Eureka moment

"Matthew" <matthew hat.stlsoft.dot.org> wrote in message 
news:dops5r$phn$1 digitaldaemon.com...
 DMC++ 8.45 has an ICE when asked to compile code using the non-member !=
 operator for stlsoft::pod_vector. 8.44 doesn't.

 This is reminiscent of a behaviour I've seen with many "older" compilers,
 whereby non-member (in)equality operators of class templates are not
 properly resolved, leading to the need to workaround by conditionally
 defining the operators as member functions. The surprise here is that 8.44
 likes it, and 8.45 doesn't and _appears_ to have regressed.

 I'm afraid I haven't time to pare it down, but just wanted to give you a
 heads up on the off-chance it might provoke a Eureka moment


I don't know what might be causing the problem.