On 09.08.2011, at 16:17, Lukas Kahwe Smith wrote:

> Hi,
> 
> I am undergoing some frustrations with SQL2, which hopefully someone on this 
> list can resolve.
> 
> I have a structure like follows
> 
> /issues
> /issues/2011
> /issues/2011/08
> /issues/2011/08/09
> /issues/2011/08/09/foo
> /issues/2011/08/09/bar
> /issues/2011/08/02
> /issues/2011/08/02/foo
> /issues/2011/08/02/bar
> 
> now I am trying to get a list of all the "dates"
> 
> aka fetch all nodes that match LIKE '/issues/____/__/__' (not sure if 
> underscore is supported in SQL)
> 
> I tried two approaches.
> 
> The first the obvious one just trying to match the PATH (note for now I 
> hardcoded the 2011 and 08 but eventually I would want those to be place 
> holders):
> 
> HTTP 400: Query:
> SELECT * FROM [nt:unstructured] WHERE PATH()(*)LIKE '/issue/2011/08/%'; 
> expected: LENGTH, NAME, LOCALNAME, SCORE, LOWER, UPPER, or CAST
> 
> Passing  [nt:unstructured] to PATH() doesnt help. Nor does using jcr:path.
> 
> The other approach I tried was finding the node by matching the child nodes 
> aka (/issues/2011/08/09/*) using an INNER JOIN ON ISDECENDENTNODE() but that 
> either got me no matches or a 403. Not sure if I am getting the 403 if there 
> are too many matches because if I remove certain conditions the 403 happend 
> if I left all relevant conditions in I got no result. But without the join 
> just looking without filters I get results, so I was pretty certain my 
> queries worked.
> 
> Any tipps? I am also wondering if there are any tools that can help debug the 
> internals of whats going on inside Jackrabbit when sending SQL2 queries via 
> davex.


BTW trying to ORDER BY jcr:path also does not seem to have any effect.

regards,
Lukas Kahwe Smith
[email protected]



Reply via email to