[Bf-committers] Blender 2.5 malicious scripting

Campbell Barton ideasman42 at gmail.com
Thu Feb 25 12:51:49 CET 2010


@Tyler, from conversations with python devs, the request of sandboxing
gets the response "dont even think about it!",
I'm not especially interested in security to the point where Id try
motivate others, but am not against someone working on it either.

so now this seems to boil down to "who wants to write a patch" :),
probably should be a command line argument like Dali wrote for 2.4x as
well as an option on load.
There are stull some issues still like, what happens when you double
click on a file to open so maybe something like this.

- user default for the startup auto-execution-script value.
- global flag for auto script execution that is reset on loading blend
files and can be set on file load.
would look something like G.flag & G_PY_AUTO_EXEC, U.py_auto_exec
which could be accessed anywhere.

again, I think hashing scripts would be hard to manage well, not to
mention hashing every pydriver (can be 100's) and having a place to
store this, varify etc.

On Thu, Feb 25, 2010 at 9:26 AM, Stefan Langer
<mailtolanger at googlemail.com> wrote:
> 2010/2/25 Tyler Tricker <tntricker at gmail.com>
>
>
>> [...] What about checking MD5 hashes on core scripts and having a command
>> line
>> option to shut down all other scripts? That way if there is a bad script, a
>> user still has the ability to open a file to try and extract useful data.
>> [...]
>>
> Use SHA cause MD5 is broken and can be easily faked now a days.
> _______________________________________________
> Bf-committers mailing list
> Bf-committers at blender.org
> http://lists.blender.org/mailman/listinfo/bf-committers
>



-- 
- Campbell


More information about the Bf-committers mailing list