Re: derefInSearching (was Re: protocol-27 comments #3)

[Hallvard B Furuseth <h.b.furuseth@usit.uio.no>]

This was either lost or ignored - maybe deservedly so.  If nobody
replies, I'll consider the issue dead.  (As I note below, it still seems
to say dereferencing the base object is not recursive, but OTOH I don't
expect anyone to misunderstand.)

Hallvard B Furuseth writes:
>Jim Sermersheim writes:
>>>>> Hallvard B Furuseth <h.b.furuseth@usit.uio.no> 11/8/04 5:59:28 AM
>>>> derefInSearching: While searching, dereference any alias entry
>>>> subordinate to the base object which is also in the search
>>>> scope. The filter is applied to the dereferenced object(s). If
>>>> the search scope is wholeSubtree, the search continues in the
>>>> subtree of any dereferenced object. Aliases in that subtree are
>>>> also dereferenced.
>>>
>>> This seems to say that recursive dereferencing stops at alias entries
>>> that are not subordinate to the base object, and that onelevel
>>> searches do not recursively dereference aliases.
>>
>> How's this:
>>
>> While searching subordinates of the base object, dereference any alias
>> within the search scope (the act of dereferencing an alias includes
>> recursively dereferencing aliases which refer to aliases). Dereferenced
>> objects become the vertices of further search scopes where the Search
>> operation continues. If the search scope is wholeSubtree, the Search
>> continues in the subtree(s) of any dereferenced object. If the search
>> scope is singleLevel, the search is applied to any dereferenced objects,
>> and is not applied to their subordinates.
>
> I want to mix the old text with your new text, but I'm not sure how...
> Suggestions:
>
>   Move the part about aliasing being recursive directly under
>   derefAliases.  Move up 'Servers MUST detect looping...' as well.
>
>   It was perhaps unfortunate to have 'The filter is applied to the
>   dereferenced object(s).' _before_ saying that a subtree search
>   continues, but OTOH 'Aliases in that subtree are also dereferenced'
>   seems worth keeping.

-- 
Hallvard