I need to evaluate a Python expression from C++. This code seems to work:
PyObject * dict = PyDict_New();
PyObject * val = PyRun_String(expression, Py_eval_input, dict, 0);
Py_DECREF(dict);
Unfortunately, it fails horribly if expression is "True" of "False" (that is, val is 0 and PyErr_Occurred() returns true). What am I doing wrong? Shouldn't they evaluate to Py_True and Py_False respectively?
PyObject* PyRun_String(const char *str, int start, PyObject *globals, PyObject *locals);
If you want True and False they will have to be in the *globals
dict passed to the interpreter. You might be able to fix that by calling PyEval_GetBuiltins
.
From the Python 2.6 source code:
if (PyDict_GetItemString(globals, "__builtins__") == NULL) {
if (PyDict_SetItemString(globals, "__builtins__",
PyEval_GetBuiltins()) != 0)
return NULL;
}
If that doesn't work, you could try to PyRun_String("import __builtin__ as __builtins__", globals, locals)
before calling PyRun_String("True", ...)
.
You might notice the Python interactive interpreter always runs code in the __main__
module which we haven't bothered to create here. I don't know whether you need to have a __main__
module, except that there are a lot of scripts that contain if __name__ == "__main__"
.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With