[XML Schema] Including a schema document with absent target namespace to a schema with specified tar

[Stanimir Stamenkov]

I'm trying to find out if it is permissible to include a schema
document with absent target namespace to a schema with specified
target namespace, and if it is, what are the rules to resolve the
target namespace of the components from the included schema document.

I'm confused because of the rules I read in the XML Schema spec

If the <element> element information item has <schema> as its parent, the
corresponding schema component is as follows:
...
{target namespace} The ·actual value· of the targetNamespace [attribute]
of the parent <schema> element information item, or ·absent· if there is none.
...
otherwise if the <element> element information item has <complexType> or
<group> as an ancestor...
...
{target namespace} If form is present and its ·actual value· is qualified,
or if form is absent and the ·actual value· of elementFormDefault on the
<schema> ancestor is qualified, then the ·actual value· of the
targetNamespace [attribute] of the parent <schema> element information item,
or ·absent· if there is none, otherwise ·absent·.

and I've tried using the Xerces2 XSModel implementation with the
following example:

-----"primary-example.xsd"

<?xml version="1.0" ?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
targetNamespace="urn:x-example:test"
xmlnsrimary="urn:x-example:test">
<xs:include schemaLocation="secondary-nonamespace.xsd" />
<xs:element name="doc-root">
<xs:complexType>
<xs:sequence>
<xs:element name="dummy" />
<xs:element ref="primary:test-elem" />
</xs:sequence>
</xs:complexType>
</xs:element>
</xs:schema>

-----"secondary-nonamespace.xsd"

<?xml version="1.0" ?>
<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
elementFormDefault="qualified">
<xs:element name="test-elem">
<xs:complexType>
<xs:sequence>
<xs:element name="foo" />
<xs:element name="bar" />
</xs:sequence>
</xs:complexType>
</xs:element>
</xs:schema>

-----

The result is the element components in the included schema are
effectively put in the namespace of the including schema, although
strictly following the rules I've quoted above it shouldn't be
possible, I think.

I guess I'm missing something - could someone point me out?

 

[Stanimir Stamenkov]

/Stanimir Stamenkov/:

I'm trying to find out if it is permissible to include a schema
document with absent target namespace to a schema with specified
target namespace, and if it is, what are the rules to resolve the
target namespace of the components from the included schema document. ....
I guess I'm missing something - could someone point me out?

O.k. I've found it <http://www.w3.org/TR/xmlschema-1/#src-include>, clause 3.2.

Feeling a bit silly now, but anyway.

 

[C. M. Sperberg-McQueen]

I'm trying to find out if it is permissible to include a schema
document with absent target namespace to a schema with specified
target namespace, and if it is, what are the rules to resolve the
target namespace of the components from the included schema
document.

If a schema document gives no target namespace, the normal
rules for processing that document will produce components
without any target namespace, just as you say the relevant
parts of the spec lead you to believe. In other words, your
reading of the part of the spec you quote is quite right.

When such a schema document is INCLUDED, however, special
processing takes place, as described in section 4.2.1 of the
specification [1]: The top level components without
namespaces are treated as orphans, and given a new home in
the namespace of the including document. (This doesn't
strictly speaking contradict the material you quoted,
because it's described not as a change to the mapping from
XML source to schema components, but as an operation
performed on the schema components derived from the
included schema document.)

[1] http://www.w3.org/TR/xmlschema-1/#compound-schema

Components described in a schema document without a target
namespace can thus take on the target namespace of any
including schema document; the phenomenon is known for
obvious reasons as 'chameleon include'.

If that's not what you want -- if you just want those
components, as is, without taking on the coloration of
their surroundings -- then you want to be using import,
not include. An import element in a schema document says
two things: (1) "I'll be referring to components in
another namespace, to wit: ...", and (2) (optionally)
"Some components in that other namespace are over
there; follow the schemaLocation info."

xsd:import imports foreign namespaces; xsd:include includes
new components for the current schema document's target
namespace (and imposes the current target namespace on them
if they don't have one of their own -- if they do have one,
and it's different, then an include should raise an error).

So if by 'include' you mean 'use the xsd:include mechanism',
the answer is no, you cannot do what you describe in XML
Schema 1.0. If by 'include' you just mean 'acquire the
components', then the answer is 'yes, you can, by using
the xsd:import mechanism rather than xsd:include'.

If what you want is not so much a particular behavior as
just to understand what's going on, then you may want to
look at the current working draft of XML Schema 1.1 [2], in
which part (not all, and not enough) of the treatment of
schema composition has been revised to try to make it
clearer. (No change of functionality, just the start of an
attempt to lay it out more clearly.)

[2] http://www.w3.org/TR/xmlschema11-1/

I hope this helps.

-C. M. Sperberg-McQueen
World Wide Web Consortium

 

[Stanimir Stamenkov]

/C. M. Sperberg-McQueen/:

If a schema document gives no target namespace, the normal
rules for processing that document will produce components
without any target namespace, just as you say the relevant
parts of the spec lead you to believe. In other words, your
reading of the part of the spec you quote is quite right.

When such a schema document is INCLUDED, however, special
processing takes place, as described in section 4.2.1 of the
specification [1]: ...

[1] http://www.w3.org/TR/xmlschema-1/#compound-schema

Thank you for pointing it out. I've found it just after I've posted
initially. It is that I use the spec more as a reference when I'm in
doubt and sometimes I can't find what I'm looking for easily.

[2] http://www.w3.org/TR/xmlschema11-1/

I've took a glance but haven't found much difference... may be I
should look more closely, again.