Ah, now I see. Yes that makes perfect sense. Regards Julian
On Tue, Oct 11, 2016 at 2:59 PM, Roy Teeuwen <[email protected]> wrote: > Hey Julian, > > No, because then you get as result from the query the some-component, what I > want the result returned by the query to be the element cq:PageContent (that > contains a parsys with some-component on) > > Greets, > Roy >> On 11 Oct 2016, at 14:54, Julian Sedding <[email protected]> wrote: >> >> Hi Roy >> >> I see, thanks for the pointer. I didn't know that was allowed in XPath. >> >> Glad you got a solution. I'm almost certain that the following would >> also work: >> /jcr:root/content/site//element(*,cq:PageContent)/parsys/some-component >> for your case. >> >> Regards >> Julian >> >> >> On Tue, Oct 11, 2016 at 1:44 PM, Roy Teeuwen <[email protected]> wrote: >>> Hey Julian, >>> >>> This is standard XPath and worked in JCR 2, for example if you look at >>> these samples provided by microsoft: >>> https://msdn.microsoft.com/en-us/library/ms256086(v=vs.110).aspx >>> <https://msdn.microsoft.com/en-us/library/ms256086(v=vs.110).aspx> >>> book[excerpt] : All <book> elements that contain at least one <excerpt> >>> element child. >>> >>> A real life used example: >>> /jcr:root/content/site//element(*,cq:PageContent)[parsys/some-component] to >>> select pages with a specific component in the parsys >>> But they gave me a workable solution in the oak-dev channel, although it >>> should still work in the old way according to xpath :) >>> >>> The solution they provided: something like >>> /jcr:root/content/site//element(*,cq:PageContent)[parsys/some-component/@jcr:primaryType] >>> >>> Greetings, >>> Roy >>> >>>> On 11 Oct 2016, at 11:54, Julian Sedding <[email protected]> wrote: >>>> >>>> Not sure why this worked in JR2 (and whether that was a bug or a feature). >>>> >>>> IMO you need to test for /jcr:root//*[fn:name() = 'test'] (or if there >>>> are no other conditions simply /jcr:root//test). >>>> >>>> Regards >>>> Julian >>>> >>>> >>>> On Mon, Oct 10, 2016 at 1:14 PM, Roy Teeuwen <[email protected]> wrote: >>>>> Hey all, >>>>> >>>>> I’m not really getting a reply in the oak-dev channel, so I might as well >>>>> also try here because I’m trying it in a sling application :) >>>>> >>>>> When doing the following XPath query in JCR 2, it would select me all the >>>>> nodes that has a subnode named test. But since upgrading to oak, this >>>>> query does not work anymore. Is there a reason this stopped working or a >>>>> way to make it work again >>>>> >>>>> Some query example: >>>>> /jcr:root//*[test] or >>>>> /jcr:root/content/site//element(*,nt:unstructured)[@jcr:createdBy='admin' >>>>> and test] >>>>> >>>>> Greetings, >>>>> Roy >>> >
