\chapter{Restricted Execution \label{restricted}} \begin{notice}[warning]   In Python 2.3 these modules have been disabled due to various known   and not readily fixable security holes.  The modules are still   documented here to help in reading old code that uses the   \module{rexec} and \module{Bastion} modules.\end{notice} \emph{Restricted execution} is the basic framework in Python that allowsfor the segregation of trusted and untrusted code.  The framework is based on thenotion that trusted Python code (a \emph{supervisor}) can create a``padded cell' (or environment) with limited permissions, and run theuntrusted code within this cell.  The untrusted code cannot break outof its cell, and can only interact with sensitive system resourcesthrough interfaces defined and managed by the trusted code.  The term``restricted execution'' is favored over ``safe-Python''since true safety is hard to define, and is determined by the way therestricted environment is created.  Note that the restrictedenvironments can be nested, with inner cells creating subcells oflesser, but never greater, privilege. An interesting aspect of Python's restricted execution model is thatthe interfaces presented to untrusted code usually have the same namesas those presented to trusted code.  Therefore no special interfacesneed to be learned to write code designed to run in a restrictedenvironment.  And because the exact nature of the padded cell isdetermined by the supervisor, different restrictions can be imposed,depending on the application.  For example, it might be deemed``safe'' for untrusted code to read any file within a specifieddirectory, but never to write a file.  In this case, the supervisormay redefine the built-in \function{open()} function so that it raisesan exception whenever the \var{mode} parameter is \code{'w'}.  Itmight also perform a \cfunction{chroot()}-like operation on the\var{filename} parameter, such that root is always relative to somesafe ``sandbox'' area of the filesystem.  In this case, the untrustedcode would still see an built-in \function{open()} function in itsenvironment, with the same calling interface.  The semantics would beidentical too, with \exception{IOError}s being raised when thesupervisor determined that an unallowable parameter is being used. The Python run-time determines whether a particular code block isexecuting in restricted execution mode based on the identity of the\code{__builtins__} object in its global variables: if this is (thedictionary of) the standard \refmodule[builtin]{__builtin__} module,the code is deemed to be unrestricted, else it is deemed to berestricted. Python code executing in restricted mode faces a number of limitationsthat are designed to prevent it from escaping from the padded cell.For instance, the function object attribute \member{func_globals} andthe class and instance object attribute \member{__dict__} areunavailable. Two modules provide the framework for setting up restricted executionenvironments: \localmoduletable \begin{seealso}  \seetitle[http://grail.sourceforge.net/]{Grail Home Page}           {Grail, an Internet browser written in Python, uses these            modules to support Python applets.  More            information on the use of Python's restricted execution            mode in Grail is available on the Web site.}\end{seealso}