• From: Rick Marshall <rjm@z...>
• To: pau carre <pau.carre@g...>
• Date: Sun, 04 Feb 2007 17:22:05 +1100

curious about this. why not the traditional data solution of an extra 
attribute specifying the type of measure? as in
<item name="water" measure="volume" value="22" />

it also opens up lots of extra, easy, possibilities for analysis.

Rick

pau carre wrote:
> I am trying to code a XML Schema (XSD file) for a web service in such 
> a way:
>
> <items>
>   <item name = "water" volume ="22">
>   <item name = "meat" weight = "10" >
> </items>
>
> It is to say, if name is "water" , then use the "volume" attribute.
> When name is "meat" then use "weight" attribute.
>
> What I have done is to define a simple type for name attribute
>
> <xs:simpleType name="nameType">
>        <xs:restriction base="xs:string">
>                <xs:enumeration value="water"/>
>                <xs:enumeration value="meat"/>
>        </xs:restriction>
> </xs:simpleType>
>
> Then I defined an abstract complex type for "item" element:
>
> <xs:complexType name="itemType" abstract = "true">
>        <xs:attribute name = "name" type = "nameType"/>
> </xs:complexType>
>
> I restricted the "name" attribute to "meat":
>
> <xs:complexType name="itemTypeMeat">
>        <xs:complexContent>
>                <xs:restriction base="itemType">
>                        <xs:attribute name = "name" type = "nameType"
> fixed="meat"/>
>                </xs:restriction>
>        </xs:complexContent>
> </xs:complexType>
>
> ... and finally I added an attribute "weight":
>
> <xs:complexType name="itemTypeMeatWeight">
>        <xs:complexContent>
>                <xs:extension base="itemTypeMeat">
>                        <xs:attribute name = "weight" type = "xs:int"/>
>                </xs:extension>
>        </xs:complexContent>
> </xs:complexType>
>
> For the "water" is the same as "meat". First I restricted the "name"
> attribute to "water":
>
> <xs:complexType name="itemTypeWater">
>        <xs:complexContent>
>                <xs:restriction base="itemType">
>                        <xs:attribute name = "name" type = "nameType"
> fixed="water"/>
>                </xs:restriction>
>        </xs:complexContent>
> </xs:complexType>
>
> ... and finally I added the "volume" attribute
>
> <xs:complexType name="itemTypeWaterVolume">
>        <xs:complexContent>
>                <xs:extension base="itemTypeWater">
>                        <xs:attribute name = "volume" type = "xs:int"/>
>                </xs:extension>
>        </xs:complexContent>
> </xs:complexType>
>
> The problem is to define the type for "item" element. If I use
> "itemType" as type it the XML does not validated because
> "itemType" is abstract. I I use:
>
> <xs:element  name="items">
>        <xs:complexType>
>                <xs:sequence>
>                        <xs:choice>
>                                <xs:element name="item"
> type="itemTypeWaterVolume"/>
>                                <xs:element name="item"
> type="itemTypeMeatWeight"/>
>                        </xs:choice>
>                </xs:sequence>
>        </xs:complexType>
> </xs:element>
>
> the XML does not validate because the validator assumes that the
> "item" type is the first defined (itemTypeWaterVolume). If I
> change the declaration of "item"  element then the validator
> assumes that the type of "item" is "itemTypeMeatWeight".
>
> Do you know how to solve this problem?
>
> Thanks in advance.
>
> Pau
>
> _______________________________________________________________________
>
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
>
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@l...
> subscribe: xml-dev-subscribe@l...
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>

• References:
  • An element with more than one possible type
    • From: "pau carre" <pau.carre@g...>