Anton,

I will try to provide more detailed answers, but let me comment upon on
thing now.

Anton Brass wrote:
> Hey Werner,
> 
> so for answering the questions:
> 
> 1) Are those classes by any chance part of a class hierarchy ?
> 
> Yes they are in a class hirarchie, it can be found in the zip file of the 
> last email.
> 
> 2)  Why would you want that just at this very location ? Why not let Castor 
> decide if it's needed ? Again, a short example might be helpful.
> 
> I want them in this very locations, because i must fit an european standard 
> for exchange of health information. The XSD can be found here:
> http://pangea.upv.es/linkehr/wp-content/uploads/2009/09/en13606.zip
> 
> Maybe I should generate the code out of it?
Yes, most definitely. If you happen to have a set of XML schemas
available, code generation is your friend.
> 
> 3) More specific example.
> 
> So here we go.
> 
> I just take 6 classes (included in the first email) which I want to combine:
> 
> - RecordComponent with the attributes:
>     - meaning : CV
>     - name : SimpleText
> 
> - Composition extends RecordComponent and no needed attributes for the example
> 
> - DataValue with no attributes
> 
> - CD extends DataValue with attributes:
>     - codedText : String
>     - codingSchemeName: String
> 
> - Text extends DataValue with the attribute:
>     - originalText : String
> 
> - SimpleText with no attribute, just specification
> 
> Now I want a Composition with the attributes meaning and name set. After 
> Marshalling it with my xml mapping (included in the first email) I get this 
> transformation:
> 
> <COMPOSITION>
> <meaning>
>     <codingSchemeName>AntonBrass</codingSchemeName>
>     <codeValue>anton</codeValue>
> </meaning>
> <name xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; 
> xsi:type="SIMPLETEXT">
>     <originalText>anton</originalText>
> </name>
> </COMPOSITION>
> 
> 
> The attribute name behaves like I want it to, setting the xsi:type I defined 
> in the mapping as map-to, meaning is not. 
> I want instead it like this:
> 
> <COMPOSITION>
> <meaning xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; xsi:type="CV">
>     <codingSchemeName>AntonBrass</codingSchemeName>
>     <codeValue>anton</codeValue>
> </meaning>
> <name xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; 
> xsi:type="SIMPLETEXT">
>     <originalText>anton</originalText>
> </name>
> </COMPOSITION>
> 
> Is this possible anyhow? I played with the different settings of the 
> marshaller, but with no success. The reference guide doesn't help me, because 
> I have entirely the same mapping and do not want a auto-naming that is 
> derived by the class.
> 
> If you want I can create a dummy project with the example.
> 
> Feel free to ask further questions!
> 
> Thank you a lot!
> 
> Best regards
> 
> Anton Brass
> 
> 
> 
> 
> 
> ----- Original Message ----- 
> From: "Werner Guttmann" <[email protected]>
> To: <[email protected]>
> Sent: Tuesday, September 22, 2009 2:00 PM
> Subject: Re: [castor-user] [XML] XSI:TYPE behavior
> 
> 
>> Hi Anton,
>>
>> let me ask you a few questions .....
>>
>> Anton Brass wrote:
>>> Hey there,
>>>
>>> I want to use castor for simply mapping existing java objects to xml.
>>>
>>>
>>> But my problem is, that some objects are adding xsi:type
>>> automatically and others not. With others I do not mean the simple
>>> types directly supported by castor, but own classes, which are not
>>> much different to the classes, which are working correctly.
>> Are those classes by any chance part of a class hierarchy ?
>>
>>> I don't really understand how I can switch on/off the xsi:type or can
>>> impact the output for this attribute, even after reading the
>>> ReferenceGuide chapter many times.
>> Well, your options are limited in some way, but there are options.
>>
>> a) Turn it off completely (option on the Marshaller).
>> b) Read
>>
>> http://castor.org/reference/html/XML%20data%20binding.html#xml.mapping.mappingfile.xsitype
>>
>> again, and try to play with the options defined (and explained) therein.
>>
>> c) Showcase a small (self-sufficient) problem that allows us to answer
>> your questions more specifically.
>>
>>> I want that every element which is a field of a class A and type of
>>> class B adds my defined map-to type in form of the xsi:type
>>> attribute.
>> Why would you want that just at this very location ? Why not let Castor
>> decide if it's needed ? Again, a short example might be helpful.
>>
>>> I added all my classes and the mapping files I use as attachment in
>>> this E-Mail. The class and its extensions that are not working the
>>> same way like the others is called "es.ehrflex.core.en13606.text.CD".
>> Sorry, but you will have to break down things so that your questions are
>> a bit moire specific.
>>
>>> I'm using Castor 1.3rc1 (and others tested), JDK 1.6, Apache Tomcat
>>> 6.0.0.20, Maven 2 and GWT 1.6.4.
>> Why not switch to 1.3 final ?
>>
>>> Thank you a lot for any helpful suggestions.
>>>
>>> Best regards
>>>
>>> Anton Brass
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>>
>>> ---------------------------------------------------------------------
>>>  To unsubscribe from this list, please visit:
>>>
>>> http://xircles.codehaus.org/manage_email
>> ---------------------------------------------------------------------
>> To unsubscribe from this list, please visit:
>>
>>    http://xircles.codehaus.org/manage_email
>>

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email


Reply via email to