jackrabbit-oak-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "angela (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (OAK-5689) AbstractSecurityTest: enforce test-failure for traversal queries
Date Mon, 20 Feb 2017 14:27:44 GMT

    [ https://issues.apache.org/jira/browse/OAK-5689?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15874598#comment-15874598
] 

angela edited comment on OAK-5689 at 2/20/17 2:27 PM:
------------------------------------------------------

Revision 1783769: explicity set {{QueryEngineSettings}} in {{AbstractSecurityTest}} that will
fail upon traversal and adjust tests that perform queries for which no index exists. 

Note: the following classes contain queries that resulted in a traversal (and thus exception)
despite the fact that there exist an index, which IMHO should have been used. earlier tests
revealed that limiting the query to a subtree of the indexed tree can cause the query to end
in a traversal, which seems unexpected to me... for the time being I marked these tests with
a FIXME comment:

- {{PrincipalProviderImplTest}}
- {{EveryoneGroupTest}}
- {{UserManagerImplTest}}
- {{UserPrincipalProviderTest}}
- {{AutoSaveEnabledManagerTest}}.

[~tmueller] as discussed recently, I would appreciate if you could take a look to see if it
is really expected to have a traversal in these cases and if that is also sensible. Once we
are done with the analysis and depending on the outcome I would remove the FIXME comment or
the code that resets the {{QueryEngineSettings}} for these tests.


was (Author: anchela):
Revision 1783769: explicity set {{QueryEngineSettings}} in {{AbstractSecurityTest}} that will
fail upon traversal and adjust tests that perform queries for which no index exists. 

Note: the following classes contain queries that resulted in a traversal (and thus exception)
despite the fact that there exist an index, which IMHO should have been used. earlier tests
revealed that limiting the query to a subtree of the indexed tree can cause the query to end
in a traversal, which seems unexpected to me... for the time being I marked these tests with
a FIXME comment:

- {{PrincipalProviderImplTest}}
- {{EveryoneGroupTest}}
- {{UserManagerImplTest}}
- {{UserPrincipalProviderTest}}
- {{AutoSaveEnabledManagerTest}}.

@tmueller as discussed recently, I would appreciate if you could take a look to see if it
is really expected to have a traversal in these cases and if that is also sensible. Once we
are done with the analysis and depending on the outcome I would remove the FIXME comment or
the code that resets the {{QueryEngineSettings}} for these tests.

> AbstractSecurityTest: enforce test-failure for traversal queries
> ----------------------------------------------------------------
>
>                 Key: OAK-5689
>                 URL: https://issues.apache.org/jira/browse/OAK-5689
>             Project: Jackrabbit Oak
>          Issue Type: Test
>          Components: core
>            Reporter: angela
>            Assignee: angela
>             Fix For: 1.8
>
>




--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message