[Python-Dev] Question on imports and packages

[Guido van Rossum]

FWIW, I also agree with Michael that static analysis would be much
preferred. You never know what side effects importing a module has.
(This could even be construed as an attack vector.)

--Guido

On Tue, Nov 2, 2010 at 7:54 PM, Brett Cannon <brett at python.org> wrote:
> On Tue, Nov 2, 2010 at 17:35, Guido van Rossum <guido at python.org> wrote:
>> If you are importing the code, the __module__ attribute on each class
>> should tell you where it is actually defined (as opposed to where you
>> imported it from). Then sys.modules gives you the module object which
>> has a __file__ attribute, etc.
>
> What Guido said. It's the equivalent of browsing an object that a
> function returned to you. Working backwards to where something is
> defined has nothing to do with imports and more to do with __module__,
> __class__, etc. Import has nothing to do with introspection for things
> that you access off of a module that happened to have imported the
> object.
>
>>
>> On Tue, Nov 2, 2010 at 4:44 PM, Raymond Hettinger
>> <raymond.hettinger at gmail.com> wrote:
>>> Brett,  Does the import mechanism for importing packages preserve enough information to be able to figure-out where all the components are defined?  I'm wondering if it is possible for the class browser to be built-out to scan/navigate class structure across a module that has been split into a package.
>>
>> --
>> --Guido van Rossum (python.org/~guido)
>>
>



-- 
--Guido van Rossum (python.org/~guido)