[Zope3-dev] Form framework, adapters and pau

[Dominik Huber]

Jim Fulton wrote:

> Dominik Huber wrote:
>
>> Excuse me late response I was busy that weekend...
>>
>> Jim Fulton wrote:
>>
>>>> I fixed that issue within the branch 
>>>> 'Zope3/branches/dominik-locatableadapters'
>>>> Jim, could you take a look at that please. Thank you very much in 
>>>> advance!
>>>
>>>
>>>
>>>
>> [...]
>>
>>>
>>> We should *only* add the location if the adapter requires a 
>>> permission other
>>> than zope public.  Key rereferences don't require a permission and 
>>> should not
>>> be location proxied. Doing so makes them unpicklable.
>>>
>>> We're going to need to fix the trusted adapter factory to do this 
>>> correctly.
>>> I suggest we create a new kind of trusted adapter factory, perhaps
>>> "LocatingTrustedAdapter" factory that adds the location. This should
>>> only be used when the adapter directive specified a permission other
>>> than zope.Public. Otherwise, the original trusted adapter factory 
>>> should
>>> be used.
>>
>>
>>
>> May be I do not understand the problem correctly:
>>
>> IMO the permission declaration within the adapter directive has 
>> nothing to do
>> with the permission declaration wihtin the class directive that mus 
>> be used if
>> trusted adapters are invoked.
>
> >
>
>>
>> Example KeyReference Adapter:
>>
>>  <adapter
>>      for="persistent.interfaces.IPersistent"
>>      provides=".interfaces.IKeyReference"
>>      factory=".persistent.KeyReferenceToPersistent"
>>      trusted="y"
>>      />
>>
>>  <class class=".persistent.KeyReferenceToPersistent">
>>    <require permission="zope.Public" 
>> interface=".interfaces.IKeyReference" />
>>  </class>
>
>
> That's odd.  I didn't notice the class directive.  I don't know why the
> class directive is being used here. This is one of those rare cases where
> we have to make the security declaration on the adapter class. (In this
> case, it's because the adapters are pickled, which is also unusual.)
>
> It is not generally the case that you need to use separate security
> declarations with trusted adapters.

I declare those additional class directive all the time if I'm using 
trusted adapters.
IMO this kind of registration is the common pattern

For example stephan richter uses this pattern within the wiki:

  <adapter
      factory=".wikipage.WikiPageHierarchyAdapter"
      provides=".interfaces.IWikiPageHierarchy"
      for=".interfaces.IWikiPage"
      trusted="true" />

  <class class=".wikipage.WikiPageHierarchyAdapter">
    <require
        permission="zwiki.ViewWikiPage"
        attributes="parents path findChildren"
        />
    <require
        permission="zwiki.ReparentWikiPage"
        attributes="reparent"
        set_attributes="parents"
        />
  </class>

>> permission declaration of the possible adapter instances (<class 
>> ...). Therefore we can't switch the
>> adapter factories within the adapter directive processing..
>
> So, during the registration of an trusted adapter  factory 
> (<adapter...) I can't do any prediction about the
>
> No, you can't do so implicitly.  You can know about the permission
> used if a non-public permission was supplied.  If a non-public permission
> is supplied, then it is reasonable to add the location. Otherwise, you 
> should
> not try to guess.

But the permission declaration within the adapter directive defines the 
permission that is nessecary to invoke the adapter factory.
It's implict to asume that the  instance by a certain factory desires 
the same permissions like the invocationg of the factory.

>> - How should we handle NullAdapters correctly?
>
> All the same I implemented your sugestion. The following question 
> appears:
>
> What do you mean by NullAdapters?  Are you talking about factories that
> return None?
>
> Id so, then, obviously, you should not proxy None.

No, 'NullAdapter' are parameterless factories, for example x = IX()

>> IMO this solution is pretty implicit and I do not like it at all.
>
>
> What solution?  If you are talking about my suggestion to
> add a location to any adapter for whoch a non-public permission
> is specified, then I consider this far less impleicit than what you
> had before.  It doesn't guess. If it knows a permission is required,
> it adds a location.
>
Yes, the current solution in the branch.

>> We should find a way to check the permission of the adapter instances 
>> inside a single trusted adapter factory:
>>
>> class IntelligentTrustedAdapterFactory(TrustedAdapterFactoryMixin):
>>
>>    def _customize(self, adapter, context):
>>        p = ProxyFactory(adapter)
>>        checker = getChecker(p)
>>        if checker is CheckerPublic:
>>            return adapter
>>        else:
>>           return assertLocation(adapter, context)
>
> >
>
>> Questions:
>> - performance
>> - in the key refernce example it would be an dedicated checker. how 
>> could I guess if such an checker is Public?
>
>
> This is just too much magic.  It has too many rules and
> is too slow.
>
> BTW, we are not just taking about trusted adapters.  We are talking about
> *any* adapter for which a non-public permission is required.

Yes, thats clear.
We have two cases:
1. untrusted
2. trusted

1. if we register an untrusted adapter using the adapter directive we 
can make an correlation between adapter factory permission and
adapter instance: if I have no permission to invoke the factory, I 
should not be able to use its instances.

2. if we register a trusted adapter this correlation breaks because the 
adapter instance itself might have highly differented security-permission
(compare <class...)

So 1 is easy to handle in the adapter directive, but 2 not. The 
suggestion above tries to respect your requirments, but also allow to use.
 the registration pattern <adapter... and <class.. which is failry 
common and explicit.

The current solution in the branch is not able to handle 2. correctly.
Stephan's zwiki example:
- The adapter directive has no permission (which is ok, because he 
doesn't need to protect the invocation of the adapter factory itself).
- In that case the locating trusted adapter factory is not invoked 
within the registration.
- Therefore the adapter instance will not have any location.
- Therefore we can't invoke the local authentication.

My suggestion:
IMO the intid and keyrefernce stuff is an exceptional case. Nobody else 
using the zope framework tries to memorize adapters persistently.
Everyone is aware that adapters regularly are very transient.
If somebody needs a pickable adapter he should be aware that those 
adapter must provide ILocation and must be trusted.

-> Derive KeyRerenceToPersistent from Location
-> if adapter are trusted, the locating trusted adapter factory is 
invoked all the time.

Regards,
Dominik