well, I was hoping the links I provided would explain what needs to be
done. http://issues.apache.org/jira/browse/JCR-257 contains some sort of
'migration instructions'. Is there anything more specific you are
looking for?
The fact is that if I did a //* search and then check
node.isNodeType("my:type"), I could find what I was looking for, but
doing //element(*,my:type) directly did not return nothing.
yes, there's a little magic behind that use case ;)
the query '//*' gets a special treatment because this can be seen as a
workspace traversal and does not need any index lookup whatsoever.
See: [1]
regards
marcel
[1]
http://incubator.apache.org/jackrabbit/xref/org/apache/jackrabbit/core/query/lucene/WorkspaceTraversalResult.html
Martin Perez wrote:
:-(
That explains everything. The info that I was looking was under jcr:system.
The fact is that if I did a //* search and then check
node.isNodeType("my:type"),
I could find what I was looking for, but doing //element(*,my:type) directly
did not return nothing.
Where can I find information about that system query handler?
On 1/9/06, Marcel Reutegger <[EMAIL PROTECTED]> wrote:
Hi Martin,
I just tried all your queries and they worked well on the current svn
checkout.
There was a recent change in the query code to separate the workspace
index from the system index [1] and [2]. Did you try your test with the
new configuration using a system query handler?
regards
marcel
[1] http://issues.apache.org/jira/browse/JCR-257
[2] http://thread.gmane.org/gmane.comp.apache.jackrabbit.devel/4057
