Oleg Tkachenko wrote:
 > P.S. I believe full-fledged fo schema would be a very useful thing. 
Would you mind if I contribute to this project?
You just did contribute. Thanks. I'll include your change (and any 
others anyone sends me) in the next draft. I'll post it next week to 
give other people time to respond.

I've got nothing in my own FO documents that even come close to 
exercising the full spec. If we want to develop a full-fledged and 
validated schema, I'll need contributions from people who have a real 
need to validate complex documents.


Oleg Tkachenko wrote:
 > Hello!
 > Chuck Paussa wrote:
 > I've improved the FO schema adding much better type patterns and 
adding a bunch of attributes that I'd missed the first time around.
 > Well done! I tried to validate my fo documents and the first 
validation error I got was about <fo:declarations> content - I have some 
meta info there in my namespace. Spec says:
 > -----------
 > Contents:
 > (color-profile)+
 > The fo:declarations flow object may have additional child elements in 
a non-XSL namespace. Their presence
 > does not, however, change the semantics of the XSL namespace objects 
and properties. The permitted
 > structure of these non-XSL namespace elements is defined for their 
 > ------------
 > So, I suppose, schema declaration for the element should be as follows:
 > <xs:element name="declarations">
 >     <xs:complexType>
 >       <xs:sequence>
 >         <xs:element ref="fo:color-profile" maxOccurs="unbounded"/>
 >         <xs:any minOccurs="0" maxOccurs="unbounded" 
namespace="##other" processContents="skip"/>
 >        </xs:sequence>
 >       <xs:attributeGroup ref="fo:inheritable_properties_List"/>
 >     </xs:complexType>
 >   </xs:element>
 > btw, this declaration assumes that all <fo:color-profile> elements 
must come before other stuff, which is not required by spec, but I 
cannot see any way to express such constraints in schema, one could use 
<xs:choice> instead of <xs:sequence>, but this way we lose control over 
(color-profile)+ constraint.

To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]

Reply via email to