Find The Bug

[Pete Becker]

BZZT. Caught myself here. Not undefined behavior as there are inherent
sequence points in the function calls. operator ++ is overloaded here.

Ah, that's the point. Okay, unspecified here. Undefined for builtin
types. Bad business either way.

 

[Ron Natalie]

Where is the sequence point between the two increments?

They are done inside a SUBROUTINE. Operator++ is overloaded (these
are ENUMs). There is an sequence point after the value p is passed to the
function and when the value is returned.

 

[osmium]

cout << p << " " << p++ << " " << ++p << endl;
}

The output from g++ and several other compilers is _not_

0 0 2

as expected. In g++ it is

2 1 1

Did you allow for the fact that shift left has a lower precedence than
postfix ++ and prefix ++?

 

[Pete Becker]

They are done inside a SUBROUTINE.

A SUBROUTINE? This isn't FORTRAN. In C++ we have functions. <g>

 

[Ron Natalie]

A SUBROUTINE? This isn't FORTRAN. In C++ we have functions. <g>

Hey, at least I didn't invoke it with GOSUB.

 

[David Rasmussen]

Where is there a sequence point problem?

There is none, as others have pointed out. I meant the undefined
evaluation order problem.

/David