[ 
https://issues.apache.org/jira/browse/DIRSERVER-824?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12464304
 ] 

Emmanuel Lecharny commented on DIRSERVER-824:
---------------------------------------------

Injecting the collective attributes in the entry before storing the data will 
be shouting you in the feet : this is exactly what CA have been creating to  : 
to avoid such duplication of data.

Now, having said that does not help at all to solve the pb ! (sorry :)

The good point is that filtering on CA helps a lot : you can discard a full 
branch if the c-l is not found in the subentry. But this can also becoming a 
hell, if the subtree selection is a complex one...

Now, what about doing something which is close to the idea you had :
- adding CA to the selected entries, before discarding or returning them ?

Suppose you select with a filter like (&(ObjectClass=person)(c-l=ankara)), and 
you get the entry :
dn: uid=ersiner,c=turkey,ou=people,dc=apache,dc=org
objectclass: top
objectclass: person
uid: ersiner

in the c=turkey,ou=people,dc=apache,dc=org associated subtree you have a c-l: 
ankara
now you clone the entry, add c-l= ankara, then apply the filters, then if it 
matches, return the entry

wdyt ?


> Collective attributes are not evaluated in search operations when they are 
> used in filter expressions
> -----------------------------------------------------------------------------------------------------
>
>                 Key: DIRSERVER-824
>                 URL: https://issues.apache.org/jira/browse/DIRSERVER-824
>             Project: Directory ApacheDS
>          Issue Type: Bug
>    Affects Versions: 1.0.1, 1.5.0
>            Reporter: Ersin Er
>
> As collective attributes never hit the database, they are not evaluated upon 
> search operations if they are used in filters. This is an hard issue to fix 
> but we may also support this kind of virtual attributes in the future, so we 
> need a way to handle such cases. One solution can be injecting the collective 
> attributes into entries before search operations. Other LDAP servers may be 
> doing this. (I had felt this when I read their documents.) Hoever I am not 
> sure about the negative effects of that way on performance.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: 
https://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Reply via email to