F
franco
I'll work up an example soon, but in the mean time let me describe my
problem.
I have a schema with a type that contains <xs:any namespace="##any"
processContents="lax"/> <bucket/>.
In an instance document I have an element inside a <bucket/>:
<xxx:thing xsi:type='foo'/>. The namespace prefix xxx is declared to
a namespace but there is no schema supplied during validation for it.
with processContents set to lax the validator should attempt to
validate <xxx:thing xsi:type='foo'/> if definitions are available.
since they are not it should skip it.
But since xsi:type='foo' implies there is a type named foo, the
validator tries to find a type named 'foo' and fails because there is
no schema available and type foo cannot be resolved.
Is this the proper behavior of a validator in this case? is the
presence of xsi:type enough to make processContents="lax" to attempt
validation?
thanks in advance,
-franco
problem.
I have a schema with a type that contains <xs:any namespace="##any"
processContents="lax"/> <bucket/>.
In an instance document I have an element inside a <bucket/>:
<xxx:thing xsi:type='foo'/>. The namespace prefix xxx is declared to
a namespace but there is no schema supplied during validation for it.
with processContents set to lax the validator should attempt to
validate <xxx:thing xsi:type='foo'/> if definitions are available.
since they are not it should skip it.
But since xsi:type='foo' implies there is a type named foo, the
validator tries to find a type named 'foo' and fails because there is
no schema available and type foo cannot be resolved.
Is this the proper behavior of a validator in this case? is the
presence of xsi:type enough to make processContents="lax" to attempt
validation?
thanks in advance,
-franco