4 The Jinja2 sandbox can be used to evaluate untrusted code. Access to unsafe
5 attributes and methods is prohibited.
7 Assuming `env` is a :class:`SandboxedEnvironment` in the default configuration
8 the following piece of code shows how it works:
10 >>> env.from_string("{{ func.func_code }}").render(func=lambda:None)
12 >>> env.from_string("{{ func.func_code.do_something }}").render(func=lambda:None)
13 Traceback (most recent call last):
15 SecurityError: access to attribute 'func_code' of 'function' object is unsafe.
18 .. module:: jinja2.sandbox
20 .. autoclass:: SandboxedEnvironment([options])
21 :members: is_safe_attribute, is_safe_callable
23 .. autoclass:: ImmutableSandboxedEnvironment([options])
25 .. autoexception:: SecurityError
27 .. autofunction:: unsafe
29 .. autofunction:: is_internal_attribute
31 .. autofunction:: modifies_known_mutable
35 The Jinja2 sandbox alone is no solution for perfect security. Especially
36 for web applications you have to keep in mind that users may create
37 templates with arbitrary HTML in so it's crucial to ensure that (if you
38 are running multiple users on the same server) they can't harm each other
39 via JavaScript insertions and much more.
41 Also the sandbox is only as good as the configuration. We stronly
42 recommend only passing non-shared resources to the template and use
43 some sort of whitelisting for attributes.
45 Also keep in mind that templates may raise runtime or compile time errors,
46 so make sure to catch them.