AttributeError with embedded Python

PlayDough pladow at
Tue Jul 1 23:49:23 CEST 2008

I've embedded Python in an extension for a program we are using here
at work.  And I'm a bit stumped as to why I am getting an
AttributeError only in the embedded Python.

First, a bit of what I am doing.  We use a simulator for a
microprocessor we are using in our systems.  Our simulator allows for
extensions that can be loaded as shared libraries.  Rather than code
the entire extension in C/C++, I would like to make use of Python to
script the extension.

So, I first initialize Python (examples below leave out the error
checking, but it is there):


And then I make sure the script directory is in the path with
Py_GetPath() and PySys_SetPath().

Finally, I import the script (say it is in a local file ''):

pName = PyString_FromString("script");
pModule = PyImport_Import(pName);

Once the module is imported, I get objects to the functions in the
script I want to call later, which I do with:

pName = PyString_FromString("foo");
pFunc  = PyObject_GetAttr(pModule, pName);

Later, I come back and call the function:

pResult = PyObject_CallObject(pFunc, NULL);

And it always fails whenever I iterate of a list.  Say for example my
Python function is:

def foo():
  a = ['cat', 'window', 'defenstrate']
  for x in a:
    print x, len(x)

Now, the function runs, i.e. I see the output ("cat 3\nwindow
6\ndefenstrate 11\n"), but I always get the following error message:

Traceback (most recent call last):
  File "./", line 3, in foo
    for x in a:
AttributeError: 'module' object has no attribute 'reset'

Now, if I run this exact same script using Python standalone, i.e.:

Python 2.3.4 (#1, Nov 20 2007, 15:18:15)
[GCC 3.4.6 20060404 (Red Hat 3.4.6-9)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import script
cat 3
window 6
defenstrate 11

(I know, and old version of Python.  But we are stuck with it because
our processes require us to validate a newer version if we change.)

This works great.  What is the difference?  Why does it work in one
context but not the other?


More information about the Python-list mailing list