You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@directory.apache.org by Emmanuel Lecharny <el...@apache.org> on 2009/10/06 07:38:51 UTC

[Schema refactoring] Keep going...

Hi,

this morning, I validated the Add operation for the AttributeTypes, 
following this algorithm :

Add operation
-------------

We try to add a new SchemaObject into the server. Here is the algorithm :

(ou=schema)
    the SO must not exist in the schemaPartition
    if SO is disabled
      then
        store in the schemaPartition
      else
        the SO must be part of an enabled schema
        check the entry
        if OK
          store into the Registries
          store in the schemaPartition
          update the schemaModifiers AT


Sadly, I found new bugs I had to fix : when an AT does not have a SYNTAX 
(this is legal), no check was done for an existing SUP (we must have 
either a SUP or a SYNTAX). Also MR weren't inherited when they were null 
and we had a SUP with the declared MR.

Not complicated, but that shows we need some more regression tests !

-- 
--
cordialement, regards,
Emmanuel Lécharny
www.iktek.com
directory.apache.org