I think it can be a good tradeoff
From mobile
Dnia 16 paź 2013 o godz. 12:59 Trong Tran <trongtt(a)gmail.com> napisał(a):
So should we make this case insensitive test optional from TCK ? To
let it be managed by the implementation itself.
> On 3 October 2013 14:28, Boleslaw Dawidowicz <bdawidow(a)redhat.com> wrote:
> PicketLink is capable of doing case sensitive queries (there is a config switch to
disable it). I cannot guarantee that we don't have any users relying on this
functionality.
>
>> On Oct 3, 2013, at 5:58 AM, Trong Tran <trongtt(a)gmail.com> wrote:
>>
>> Do you have any further information about case-sensitive support of LDAP queries
in PL ?
>>
>>
>>> On 23 September 2013 19:51, Bolesław Dawidowicz <bdawidow(a)redhat.com>
wrote:
>>>
>>>
>>> On 16.09.2013 10:18, Julien Viet wrote:
>>> > Hi,
>>> >
>>> > recently we worked on making the eXo Core OrganizationService TCK
implemented for the picket link implementation. (I think it was not implemented because
the TCK came after the PL implementation and then nobody was aware that it was kind of
existing, anyway better late than never…)
>>> >
>>> > We came through a few conflicting behavior between what the TCK defines
and the original PL implementation testsuite:
>>> >
>>> > 1/
testRemoveGroup(org.exoplatform.services.tck.organization.TestGroupHandler)
>>> > conflict between TCK and PL test when removing a group having at
least 1 child
>>> > + TCK assumes an exception is thrown
>>> > + PL assumes this group will be removed and its children will be
removed too
>>>
>>> Ok. I think we could align on exception - this is safest.
>>>
>>> >
>>> > 2/
testFindUserProfileByName(org.exoplatform.services.tck.organization.TestUserProfileHandler):
Expected: <null> but was:
org.exoplatform.services.organization.impl.UserProfileImpl@40773f4b
>>> > conflict between TCK and PL test when a non existing user profile of
existing user is searched via find:
>>> > + TCK suppose null is returned
>>> > + PL suppose an empty profile is returned
>>>
>>> I think null is returned in other cases so we can allign on that.
>>>
>>> >
>>> > 3/
>>> > -
testFindUsersByQuery(org.exoplatform.services.tck.organization.TestUserHandler):
expected:<1> but was:<0>
>>> > -
testFindUsers(org.exoplatform.services.tck.organization.TestUserHandler):
expected:<1> but was:<0>
>>> > In method: UserHandler#findUsersByQuery(Query query)
>>> > TCK suppose that case sensitivity does not matter in a search user
but PicketLink implementation is case sensitive
>>> >
>>> > somehow we need to make this consistent.
>>>
>>> IIRC case sensivity is actually supported in case of LDAP queries in PL.
>>> I don't think it would be safe to just disable this part. Need to dig
in.
>>>
>>> >
>>> > Our take is that we should align Picket Link with the TCK.
>>> >
>>> > Comments are open.
>>> >
>>> > cheers
>>> >
>>> > Julien
>>> > _______________________________________________
>>> > gatein-dev mailing list
>>> > gatein-dev(a)lists.jboss.org
>>> >
https://lists.jboss.org/mailman/listinfo/gatein-dev
>>> >
>>>
>>>
>>>
>>> --
>>> Bolesław Dawidowicz
>>> JBoss Portal Platform Architect | GateIn Portal Project Lead
>>> _______________________________________________
>>> gatein-dev mailing list
>>> gatein-dev(a)lists.jboss.org
>>>
https://lists.jboss.org/mailman/listinfo/gatein-dev
>>
>>
>>
>> --
>> Trong Tran
>> (+84) 983841909 | trongtt(a)gmail.com
>> Twitter:
http://twitter.com/trongtt
--
Trong Tran
(+84) 983841909 | trongtt(a)gmail.com
Twitter:
http://twitter.com/trongtt