On Feb 27, 2008, at 9:51 AM, Mike D. Jones wrote:
Since we're on the subject of XPath (or SQL for that matter), does
anyone know how to sort on an ancestor node? So, if I have a node
path, such as:
/website/articles/title-of-article/articles/0/categories/0/@category
Here's the Xpath that I want to work.
//articles//element(articles,mgnl:contentNode)/*[xs:dateTime
(10/9/2007) <= @startDate]/[EMAIL PROTECTED]'ckiis009320dkeis-39820']
order by @startDate descending
In essence, this pulls all the article nodes less than the date
with the appropriate category.
/articles/article1/articles/0/categories/0
/articles/article2/articles/0/categories/01
Unfornately, the sort descending doesn't work. I think it's
because the context node is the child category node, not the node
(articles) where startDate is an attribute.
If you know how to make this work via SQL I'm happy to use that
instead!
Thanks in advance!
Mike D. Jones
I can't help much with the first query - but as a general comment
about the SQL vs Xpath query... Jackrabbit translates both queries
into their abstract query language before processing them. It has a
mechanism inside of jackrabbit to take an abstract query and spit it
back out as either an SQL or Xpath query - and in their documentation
they mention that it is useful to view the other version of a
query... Perhaps we could tweak the JCR Query module to have a
checkbox or something that says "output detailed query information
before results" that would instruct it to dump out both the xpath and
sql version of the query that Jackrabbit is using to the results window.
Or maybe a separate tool on that same query page that you could use
to do the query translation.
Ryan
----------------------------------------------------------------
for list details see
http://documentation.magnolia.info/docs/en/editor/stayupdated.html
----------------------------------------------------------------
