Dieter Maurer wrote:
Jim Fulton wrote at 2004-1-15 10:03 -0500:

Right. The name attribute was intended for attribute-based access.

IMO, it makes no sense to consider key values when doing security

I will let Jim comment on your use case.

What use case? I missed it. Where is it?

allows integers, strings, dictionary mapping names to integers
and function with signature "name,value --> boolean" as

The motivation is that some attributes may be accessible
while others should not. It is highly likely that
this decision is based on the attribute name.
When "None" is passed as name, you loose...

None is never passed when validating *attribute* access; it is (now) passed when validating *item* access (think sequence as well as mapping items). The rationale was that there were no known cases where item access was being discriminated based on the key, and that casually passing the key / index around was a source of bugs.

We can look at reverting that, if somebody has a convincing use case which requires protecting item access based on the key / index.

Tres Seaver                                [EMAIL PROTECTED]
Zope Corporation      "Zope Dealers"

Zope-Dev maillist - [EMAIL PROTECTED]
** No cross posts or HTML encoding! **
(Related lists - )

Reply via email to