Hello Nico, I understand your usecase. You could also choose to only have a property set on a document when it is not empty. We had a customer who wanted this. Would that result in the same behavior you have? Because, then a property will be only set when it is not empty, and when in the wrong namespace, it will be empty...or do I miss something.
What I do not get from the extractors.xml below (though from the top of my head, so I might miss something) is that for example in the first extractor, you have the xpath: xpath="//c:catalogName"/> Now, where is the c: prefix namespace declared? Regards Ard > > Ard, > > The fact that only the namespace of the root element is taken > into account is what's all about. The idea behind it is that > you can definea set of different properties on different > documents while they are stored within the same location > (directory). Imagine a webshop where you store product > information and catalog information alongside eachother. This > makes it possible to use DASL queries that act only on > product files or catalog files. > I don't know if there are many usecases. I know one at least ;-). > > The extractor.xml file might look like: > > <extractor > classname="nl.hippo.slide.extractor.HippoSimpleXmlExtractor" > uri="/files/default.preview/content/appdata/webshop" > content-type="text/xml > | text/xml; charset=UTF-8 | application/xml"> > <configuration > targetNamespace="http://www.mycompany.com/webshop/catalog > "> > <instruction property="catalog_name" namespace=" > http://hippo.nl/cms/1.0"; xpath="//c:catalogName"/> > </configuration> > </extractor> > > <extractor > classname="nl.hippo.slide.extractor.HippoSimpleXmlExtractor" > uri="/files/default.preview/content/appdata/webshop" > content-type="text/xml > | text/xml; charset=UTF-8 | application/xml"> > <configuration > targetNamespace="http://www.mycompany.com/webshop/product > "> > <instruction property="product_name" namespace=" > http://hippo.nl/cms/1.0"; xpath="//p:productName"/> > </configuration> > </extractor> > > > Kind regards > > Nico Tromp > > > On Thu, Oct 9, 2008 at 11:14 AM, Ard Schrijvers > <[EMAIL PROTECTED]>wrote: > > > Hello Nico, > > > > Thx first of all for sharing your code. Could you also post me an > > example of the extractor using this? > > > > Also, I see in the patch: > > > > String documentNamespace = > > document.getRootElement().getNamespaceURI(); > > > > This means that the target namespace must match the > namespace of the > > root element, right? So, the root element namespace defines in this > > case whether a property will be set or won't be set. I am not sure > > whether this is a really common usecase. Are there many > usecases for > > this behavior? > > > > Regards Ard > > > > > > > > Hi all, > > > > > > as part of the project I'm currently working on we needed > to be able > > > to have a extractors that only act on documents depending on the > > > namespace of the document. (See a ealier post > > > http://www.nabble.com/Setting-properties-per-document-namespac > > > e-td19380940.htmlor > > > http://www.nabble.com/Mixed-content-and-extractors-td19319614. > > > html). I finally made some changes to the Hippo extractors. > > > They are now able to take into account the namespace of the > > > document. If you specify a so called target namespace the > properties > > > are only set if the document has the same namespace. If you don't > > > specify the target namespace the properties are set on every > > > document. The target namespace is set as a attribute on the > > > 'configuration' element within the 'extractor' element. I > think it > > > should have been a attribute of the 'extractor' element but that > > > would mean a lot more changes and I just wanted it to work. > > > > > > If you like the idea, please feel free to use the code. And maybe > > > the Hippo guys can aply them to there code so everybody > can benifit > > > from it. > > > > > > Please see the attached diff files for the patches. > > > > > > > > > Have fun > > > > > > Nico Tromp > > > > > ******************************************** > > Hippocms-dev: Hippo CMS development public mailinglist > > > > Searchable archives can be found at: > > MarkMail: http://hippocms-dev.markmail.org > > Nabble: http://www.nabble.com/Hippo-CMS-f26633.html > > > > > ******************************************** > Hippocms-dev: Hippo CMS development public mailinglist > > Searchable archives can be found at: > MarkMail: http://hippocms-dev.markmail.org > Nabble: http://www.nabble.com/Hippo-CMS-f26633.html > > ******************************************** Hippocms-dev: Hippo CMS development public mailinglist Searchable archives can be found at: MarkMail: http://hippocms-dev.markmail.org Nabble: http://www.nabble.com/Hippo-CMS-f26633.html
