Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Secure interpreter...haven't we had that?

1 view
Skip to first unread message

Robert Hicks

unread,
Jun 1, 2007, 5:13:34 PM6/1/07
to
http://sayspy.blogspot.com/2007/05/i-have-finished-securing-python.html

Robert Hicks

unread,
Jun 1, 2007, 5:14:33 PM6/1/07
to
On Jun 1, 5:13 pm, Robert Hicks <sigz...@gmail.com> wrote:
> http://sayspy.blogspot.com/2007/05/i-have-finished-securing-python.html

Or maybe he means something different...

Darren New

unread,
Jun 1, 2007, 5:42:11 PM6/1/07
to

No, he means a safe interpreter. AFAIK, Python still only supports one
interpreter per process tho, yes? Makes it difficult to do some of the
good stuff.

--
Darren New / San Diego, CA, USA (PST)
His kernel fu is strong.
He studied at the Shao Linux Temple.

Michael Schlenker

unread,
Jun 2, 2007, 1:24:58 AM6/2/07
to
Darren New schrieb:

> Robert Hicks wrote:
>> On Jun 1, 5:13 pm, Robert Hicks <sigz...@gmail.com> wrote:
>>> http://sayspy.blogspot.com/2007/05/i-have-finished-securing-python.html
>>
>> Or maybe he means something different...
>
> No, he means a safe interpreter. AFAIK, Python still only supports one
> interpreter per process tho, yes? Makes it difficult to do some of the
> good stuff.
>
IIRC they had some safe interpreter like features but threw them out
because extensions didn't play nice with them (introduced security
breaches).

Tcls safe interpreters have the same problem if a C coded extension
exposes unsafe features in its Safe_Init Function. So review the
extensions you allow to be used.

Michael

0 new messages