Re: [Exist-development] [Exist-commits] SF.net SVN: exist:[11485] trunk/eXist/test/src/org/exist/validation

[Dannes W. <da...@ex...>]

Dmitriy,

sure I saw the repairs, that triggered me. I tried to analyze the
risks for the project. Are there more locations?
pity we do need see the deprecation warnings during build. We needed
to switch it off because otherwise the build is flooded.

cheers

Dannes

On Sat, Mar 20, 2010 at 12:32 PM, Dmitriy Shabanov <sha...@gm...> wrote:
> Not this. Check TC, 11467 & 11468 commits
>
> On Sat, 2010-03-20 at 11:32 +0100, Dannes Wessels wrote:
>> Hi,
>>
>> On Sat, Mar 20, 2010 at 7:24 AM,  <sha...@us...> wrote:
>> > [ignore] use SecurityManager.GUEST instead of UserImpl.DEFAULT
>>
>> >    trunk/eXist/test/src/org/exist/validation/DatabaseInsertResources_NoValidation_Test.java
>> >    trunk/eXist/test/src/org/exist/validation/DatabaseInsertResources_WithValidation_Test.java
>> > -            broker = pool.get(UserImpl.DEFAULT);
>> > +            broker = pool.get(SecurityManager.GUEST);
>>
>> I was a bit surprised that this little change made the junit test
>> fail. Do we have any idea why we did not see this before? Did we run
>> the test suite locally?
>>
>> Anyway, what are the risks here? At this moment we found this issue in
>> the not related validation tests by accident... did we check whether
>> there are more locations where this can happen?
>>
>> Couldn't these UserImpl.DEFAULT and SecurityManager.GUEST be made
>> identical, making it a pure replacement, so there is no risk?
>>
>> regards
>>
>> Dannes
>>
> --
> Cheers,
>
> Dmitriy Shabanov
>



-- 
eXist-db Native XML Database - http://exist-db.org
Join us on linked-in: http://www.linkedin.com/groups?gid=35624