Jukka,
But the name TransientRepository kind of implies that changes made to the
repository are gonna be lost when the program terminates... so I expected
that whenever I run the program, the repository be empty... :-)
What do you think? Shouldn't TransientRepository behave this way?
Regards,
Behi
Jukka Zitting-3 wrote:
>
> Hi,
>
> On 10/4/06, behrangsa <[EMAIL PROTECTED]> wrote:
>> Compile and run the following class:
>> [...]
>> Now, change the value of BLOG_COUNT to 20 and rerun the class. When I do
>> this, I get 20 exceptions:
>>
>> javax.jcr.PathNotFoundException: blog11
>> [...]
>> If I delete the repository directory and repository.xml and derby.log and
>> rerun the program, the exceptions fade away...
>>
>> Could someone please confirm this bug?
>
> You probably ran the program first with BLOG_COUNT set to 10. This
> created nodes /blogs and /blogs/blogN (0 <= N < 10) and those nodes
> were persisted for the next invocation of the program. Then, when you
> ran the program again, the statement s.getRootNode().addNode("blogs")
> created a *same-name sibling* for the /blogs node, namely /blogs[2],
> after which your program created nodes /blogs[2]/blogN (0 <= N < 20).
> Thus, nodes /blogs/blogN (10 <= N < 20) were never created, causing
> your PathNotFoundExceptions.
>
> You can solve the issue by
> a) removing the entire repository between program runs,
> b) deleting the /blogs content tree at the end of each run, or
> c) checking if the nodes being created already exist (see Node.hasNode).
>
> BR,
>
> Jukka Zitting
>
> --
> Yukatan - http://yukatan.fi/ - [EMAIL PROTECTED]
> Software craftsmanship, JCR consulting, and Java development
>
>
--
View this message in context:
http://www.nabble.com/TransientRepository-bug--tf2379243.html#a6630984
Sent from the Jackrabbit - Users mailing list archive at Nabble.com.
