Peeter Piegaze wrote:
On Wed, 16 Mar 2005 16:31:44 +0100, Andreas Hartmann
<[EMAIL PROTECTED]> wrote:
Tobias Strasser wrote:
well: "/*[local-name()='hugo' and
namespace-uri()='http://foo.bar.org']" is not a valid path :-)
What's wrong with it?
LocationPath -> AbsoluteLocationPath AbsoluteLocationPath -> '/' RelativeLocationPath RelativeLocationPath -> Step Step -> AbbreviatedAxisSpecifier NodeTest '[' PredicateExpr ']' NodeTest -> NameTest ("*") PredicateExpr -> Expr Expr -> AndExpr AndExpr -> EqualityExpr 'and' EqualityExpr EqualityExpr (1) -> PrimaryExpr (1) '=' Literal EqualityExpr (2) -> PrimaryExpr (2) '=' Literal PrimaryExpr (1) -> FunctionCall PrimaryExpr (2) -> FunctionCall
It's not about the quotation marks, is it? :)
The first parameter of Session.exportDocView is not an XPath. It is a
JCR path. A JCR path is similar to a file system path (e.g.,
"/a/b/c"). It points to a single item in a JCR workspace tree. The
concept of the JCR path is foundational to the design of the system
(see section 4.6 of the spec) and is used in many many methods.
got that, but let me do a concrete example:
1) I create a node "/authoring"
2) I import an XML with the following namespaces:
<?xml version="1.0" encoding="UTF-8"?> <site i18n:attr="label" label="Authoring" xmlns="http://apache.org/cocoon/lenya/sitetree/1.0"; xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; xsi:schemaLocation= "http://apache.org/cocoon/lenya/sitetree/1.0 ../../../../resources/entities/sitetree.xsd" xmlns:i18n="http://apache.org/cocoon/i18n/2.1"; >....
and attach it to the node "/authoring"
3) After the import I take a look at the repository with the dump method of the FirstSteps example and see
Node: /authoring/_pre10:site
Property: /authoring/_pre10:site/jcr:primaryType = nt:unstructured
Property: /authoring/_pre10:site/i18n:attr = label
Property: /authoring/_pre10:site/label = Authoring
Property: /authoring/_pre10:site/xsi:schemaLocation = http://apache.org/cocoon/lenya/sitetree/1.0 ../../../../resources/entities/sitetree.xsd
which means Jackrabbit "renamed" the prefix of the default namespace
http://apache.org/cocoon/lenya/sitetree/1.0
to "_pre10" for its "internal" representation
4) I want to retrieve the node
"/authoring/_pre10:site"
which works of course ;-) but I don't want to hardcode the prefix "_pre10" of course
because it could be anything, only the namespace should matter. That's why I thought
one has to do it as with XPath.
Makes sense? Just wanted to explain how I got to this thinking.
XPath
is used in a JCR repository only in the context of search.
ok, I could do this, but I guess the performance wouldn't be that good.
Let me take now a look at Jukka's DocumentViewExportVisitor.java ;-)
Thanks
Michi
Cheers, Peeter
-- Michael Wechner Wyona Inc. - Open Source Content Management - Apache Lenya http://www.wyona.com http://lenya.apache.org [EMAIL PROTECTED] [EMAIL PROTECTED]
