jackrabbit-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From ruchi goel <Ruchi.G...@Sun.COM>
Subject Re: session.move vs. workspace.move
Date Thu, 24 May 2007 10:41:19 GMT
Hi Christophe,
I had workspace.move(src,dest) in the original patch for   JCR-910.  I 
had tested it  and it worked perfect i.e.  deleted the node from src  
and  added it to dest. I see that you are now using 
session.move(src,dest). But then session.save() is also required.

Thanks,
Ruchi

Christophe Lombart wrote:
> Yes, I can make a simple basic test to check if I have the same issue.
> I would like to modify one unit test in jackrabbit-jcr-tests and run 
> it but
> the unit tests are not in the folder 'test'. So, how can I run it with 
> maven
> ?
>
> br,
> Christophe
>
> On 5/23/07, Stefan Guggisberg <stefan.guggisberg@gmail.com> wrote:
>>
>> hi christophe
>>
>> On 5/21/07, Christophe Lombart <christophe.lombart@gmail.com> wrote:
>> > Hi all,
>> >
>> > In the ocm framework, we have the feature to move object. When the
>> > implementation uses the jcr call 'session.move (src, dst)', it works
>> fine.
>> > the source node is deleted but if the implementation used the jcr 
>> call '
>> > workspace.move(src, dst)', the source node is not deleted.
>> > I'm wondering why the both methods are not providing the same 
>> result. Is
>> it
>> > not a bug or do I miss something ?
>>
>> the difference between Session#move and Workspace#move is that the
>> latter immediately persists the changes while the former causes 
>> transient
>> changes only (which need to be persisted separatly using a save() call).
>>
>> however, the current session's resulting view should be the same.
>>
>> personally i doubt that it's a bug in Workspace#move since we should
>> have noticed it earlier. however, if you have a test case that
>> demonstrates
>> the issue please create a jira issue.
>>
>> cheers
>> stefan
>> >
>> > br,
>> > Christophe
>> >
>>
>


Mime
View raw message