Re: Relationship among our data model, patterns, and XML 1.0

[Kohsuke KAWAGUCHI <kohsuke.kawaguchi@eng.sun.com>]

> it is the only child of that <oneOrMore>.  Changing the syntax from
> 
>   <oneOrMore>
>     <attribute>...</>
>   </oneOrMore>
> 
> to
> 
>   <multipleAttributes>...</>
> 
> does not buy us anything.  Murata-san is right that we need a pattern that

I agree. The current syntax is fine. It's just that the processor should
treat

<oneOrMore>
  <attribute/>
</oneOrMore>

as <multipleAttributes/> internally, if it wants.


But my concern is, what happens if the following pattern is given?

<oneOrMore>
  <choice>
    <attribute>
      <nsName ns="foo"/>
    </attribute>
    <attribute name="bar:xyz"/>
  </choice>
</oneOrMore>

Is this still OK?



>   <group>
>     <attribute><anyName/></attribute>
>     <attribute><anyName/></attribute>
>   </group>

Yet another reason to prohibit this. If we allow this, then it becomes
hard to prohibit things like:

<group>
  <attribute><anyName/>
    <data type="positiveInteger"/>
  </attribute>
  <attribute><anyName/>
    <data type="negativeInteger"/>
  </attribute>
</group>

pattern matching algorithm also becomes complex.



--
Kohsuke KAWAGUCHI                          +1 650 786 0721
Sun Microsystems                   kohsuke.kawaguchi@sun.com