Hi Emmanuel, Emmanuel Lecharny schrieb: > Hi, > > just wanted to have a second thought. Here is what bugs me : > > - we have a valid entry > - we modify it with a new ObjectClass, which is not valid ( "test", for > instance) > > Q1 : should the server accept this? > > My personnal guess would be : no, but the DIRSEVER-791 test accept it. So
No, I think the server must reject it. Do you mean the org.apache.directory.server.core.jndi.DIRSERVER791IT.java? The test testDefect2() tries to add an non-existing objectClass "test", the server rejects this and when reading the entry after that the objectClass attribute doesn't contain the "test" objectClass. So I think this test is correct. We should just change the Javadoc because it still describes the defect but not the intention of the test. > Q2: Should we modify the test accordingly to my guess ? > > Thanks ! >
