The repr of the ctypes.CArgObject instance will fail when the value is a non-ascii character.

The code is:

        sprintf(buffer, "<cparam '%c' (%c)>",
            self->tag, self->value.c);
    ...
    return PyUnicode_FromString(buffer);

If self->value.c is out of range 0-127, buffer will contain a string not decodable with UTF-8.

There is a similar problem with non-ascii self->tag.

The following PR is purposed to fix this, but I don't know how to test it. Current tests only create CArgObject instances with tag='P' (in byref()).