Hello, I do not think the query is that slow, rather the iteration of the nodes (hasNext() for example). Try in your repository.xml (workspace.xml if already existing repository) to configure :
<param name="respectDocumentOrder" value="false"/> or add an 'order by' clause in your query. This way not all 2000 results have to be loaded. see [1] for quite extended explanation on some queries. Do realize that the first query will be somewhat slower because nothing in cache, but still will complete pretty fast. Consecutive queries are fast because the hierarchical cache will be built up -Ard http://www.nabble.com/Explanation-and-solutions-of-some-Jackrabbit-queri es-regarding-performance-td15028655.html > Hi, All! > > I know this question was discussed before but I can't find > good solution. > > My content placed somewhere like > "/myRoot/news/en/SomeCategory/2008/01" and I want to get all > nodes from this path by executing guery like > "//myRoot/news/en//*/2008/01/ ..." > It takes up to minute with only 2000 items in "/myRoot/news/en/" node. > > How can I mangle query or configure repository to get > acceptable search time? > > I may use NodeIterator to get all subnodes but this is not > good solution because I use queries in JSP pages to give more > flexibility to JSP desigber. > > -- > SY, Alex Lukin > RIPE NIC HDL: LEXA1-RIPE >
