[snipSteven Bethard wrote:
Michael Spencer wrote:
Safe eval recipe posted to cookbook: http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/364469
This recipe only evaluates constant expressions
Indeed. But it's easy to extend this to arbitrary constructs. You just need to decide what code to emit for the other 50 or so ast node types. Many of those are boiler-plate binops.
Ahh, gotcha. Thanks for the clarification.
I haven't ever spent much time dealing with Python's ASTs, but my guess is doing anything here is probably worth putting off until the AST branch is merged into main CVS for Python 2.5. (I understand there are supposed to be some substantial changes, but I don't know exactly what they are or what they affect.)
Right - the crux of the problem is how to identify dangerous objects. My point is that if such as test is possible, then safe exec is very easily implemented within current Python. If it is not, then it is essentially impossible.[snip]
It might still be possible to have a reliable test within a problem-specific domain i.e., white-listing.
Yeah, that was basically my intent -- provide a white-list of the usable objects. I wonder how complicated this would be... You also probably have to white-list the types of all the attributes of the objects you provide...
Steve -- http://mail.python.org/mailman/listinfo/python-list
