[C++-sig] GCC 3.2 side-effect?

[David Abrahams]

"William Trenker" <wtrenker at hotmail.com> writes:

> The following code fragment compiled without error yesterday with GCC
> 2.95.3.
>
> handle<> h_result;
> ...
> h_result = PyRun_String(
>     "import time\n",
>     Py_file_input, main_namespace.get(), main_namespace.get()
>     );
> ...
>
> Today I upgraded to GCC 3.2.  I was able to re-build boost.python
> without error, although a number of warnings popped up that weren't
> there with the older compiler.
>
> But the code fragment above, taken from a small project I'm working
> on, gives the following errror:
>
> cm.cpp:28: no match for `boost::python::handle<PyObject>& = PyObject*'
> operator
> /root/cvsroot/boost/boost/python/handle.hpp:193: candidates are:
>    boost::python::handle<T>& boost::python::handle<T>::operator=(const
>    boost::python::handle<T>&) [with T = PyObject]
>
> What is missing in my code?

The handle<> constructor is explicit.

    h_result = handle<>(PyRun_String(...));

-- 
                       David Abrahams
   dave at boost-consulting.com * http://www.boost-consulting.com
Boost support, enhancements, training, and commercial distribution