jackrabbit-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jukka Zitting" <jukka.zitt...@gmail.com>
Subject Re: XPath ignoring predicate
Date Tue, 06 May 2008 18:21:53 GMT

On Mon, May 5, 2008 at 9:57 AM, Martin Zdila <m.zdila@mwaysolutions.com> wrote:
>  I have nodes named "aaa" and if I execute XPath query "//aaa[nonexistingnode]
>  then all "aaa" nodes are returned regardless if they contain nonexisting node
>  or not. Is that a bug?

The JCR spec doesn't require support for such queries, and though we
have added some support for the child axis within predicates (see for
example JCR-247), there still are a number of cases like this that
don't work.

You can file a feature request for such child node existence
predicates, or a bug report for the fact that such unsupported queries
are not rejected by the Jackrabbit query manager.

>  Second strange thing: //*[name() = 'aaa'] returns nothing even I have nodes
>  named "aaa" in my repository.

Try //*[fn:name() = 'aaa'], see JCR-857.

Again, Jackrabbit should reject queries with functions or other
constructs that are not supported.

>  Third strange thing: if I execute //*[aaa] query, then nodeIterator.getSize()
>  returns -1. Another bug?

A feature. RangeIterator.getSize() is allowed to return -1 in case the
size of the collection is unknown.


Jukka Zitting

View raw message