Hi Bo,

Did you specify an XPath expression to query the members from this xml file? Can I have a peek at the complete xml structure somewhere?

Thanks,

'Richard

one possible solution for you is to create a c# app that opens the xml file, and then sticks members into a sql table in the database and get cmsimport to import from there.

Hi Richard and Stephen,

Thanks for your fast replies! :)

@Richard, I've tried some various XPath Expressions to query it, but the closets I've come so far is to the shop element by explicitly doing something like ./shop. It seems that the member element becomes the root element by default. Shouldn't it be the members element? :) So the xpath expression would look something like: //member/*/* ?

@Stephen, that's definately a way of doing it, but to me it seems like double-trouble ;) Might have to go with such a solution if anything else fails.

The complete XML structure for a member is as f

Ops, a bit too fast right there on the reply button ;)

Xml structure:

<members>

 <member xmlns="mydomain" status="1" aktiv="1">

  <shop>

   <firmaNoter/>

   <firmaUrl>www.domain.dk</firmaUrl>

   <firmaUrlAlternativ/>

   <firmaGodkendelsesDato>2002-03-07T12:00:00</firmaGodkendelsesDato>

   <firmaSidstKontrolleretDato>2010-08-30T12:00:00</firmaSidstKontrolleretDato>

   <firmaBeskrivelse>

    <p>Description company</p>

   </firmaBeskrivelse>

   <firmaTags/>

   <kategori>1943</kategori>

  </shop>

  <Kunde>

   <firmaSagsbehandler>0</firmaSagsbehandler>

   <firmaStatusBesked/>

   <firmaNavn>Company name</firmaNavn>

   <firmaSelskabsform/>

   <firmaEjer/>

   <firmaAdresse>Company address</firmaAdresse>

   <firmaPostnummer>28000</firmaPostnummer>

   <firmaBy>Company city</firmaBy>

   <firmaTelefon>12345678</firmaTelefon>

   <firmaEmail/>

   <firmaCvr>12345678</firmaCvr>

   <firmaEtableringsaar/>

   <firmaAntalAnsatte>5-24</firmaAntalAnsatte>

   <firmaStatement/>

   <firmaStatementWriter/>

  </Kunde>

  <kontakter>

   <kontakt kontaktId="1">

   <firmaKontaktperson>Contact name</firmaKontaktperson>

   <firmaKontaktTelefon/>

   <firmaKontaktMobil>12345678</firmaKontaktMobil>

   <firmaKontaktEmail>[email protected]</firmaKontaktEmail>

   </kontakt>

  </kontakter>

  <logo>

   <firmaLogo/>

   <firmaLogoCustomerUpload/>

  </logo>

  <SEO>

   <metaTitle>metatitle</metaTitle>

   <metaDescription>metadescription</metaDescription>

   <metaKeywords>meta keywords</metaKeywords>

   <seoTopText/>

   <seoBottomText/>

  </SEO>

 </member>

</members>

<members>

Thanks again both of you!

Hi Bo,

No By default CMSImport takes the children of the root. In that case you don't need to specify some xpath at all. What are the elements  you want to map for this import?

Thanks,

Richard 

Hi again Richard,

I'm interested in getting the values that's in the third level elements. I.e.: <metaDescription>metadescription</metaDescription> and all attributes. Is this possible? :-) I've tried the following xpath expressions: */* and /*/* and ./*/* unfortunately without any results :/

Have I missed something here? :)

Thanks again,

Bo

Hi Bo,

Yes that is possible although not with a normal xPath. I've created a custom DataAdapter last year  for a client that had a very complex xml structure and flatten it down so it was supported by CMSImport.I can have a look at your DataAdapter might be easiest to send me the source (richard [at] soetemansoftware[dot]nl) Then I will have a look what is missing in your case and add the Methods that you can use to flatten down the xml structure.

Best,

Richard

Hi Richard,

Thanks for your email! Much appreciated :-) I decided to simply try and flatten the XML documents structure which worked right away. The code I used for flattening a deep structured XML document is this:

// Start flattening the XML output structure
            var doc = XDocument.Load(xml.CreateReader());
            var member = doc.Root.Element(NameSpace + "member");
            var descendants = member.Descendants().ToList();

            foreach (var nested in descendants.Elements().ToList())
            {
                nested.Remove();
            }

            // Replace nodes
            member.ReplaceNodes(descendants);

            // Make sure empty tags are closed like this: <tag></tag> and not like this: <tag/>
            foreach (XElement elm in doc.Root.Element(NameSpace + "member").Descendants())
            {
                if (string.IsNullOrEmpty(elm.Value))
                {
                    elm.SetValue(string.Empty);
                }
            }

Thanks again!

All the best

Bo

Hi Bo,

Great to hear that!

Thanks,

Richard

Actually, there was a mistake in my above code - it only took the first occurence of

Updated code (though not the prettiest solution):

// Start flattening the XML output structure
            var doc = XDocument.Load(xml.CreateReader());
            var member = doc.Root.Element(NameSpace + "member");

            foreach (XElement xelm in doc.Root.Elements(NameSpace + "member"))
            {
                var descendants = xelm.Descendants().ToList();

                foreach (var nested in descendants.Elements().ToList())
                {
                    nested.Remove();
                }

                // Replace nodes
                xelm.ReplaceNodes(descendants);
            }

            foreach (XElement elm in doc.Root.Element(NameSpace + "member").Descendants())
            {
                if (string.IsNullOrEmpty(elm.Value))
                {
                    elm.SetValue(string.Empty);
                }
            }

Thanks again.

- Bo

is working on a reply...