redefine and target namespace

[Florent Georges-3]

  Hi,

  I have a schema document that redefines a complex type as
following:

    <xs:redefine schemaLocation="redefine.xsd">
       <xs:complexType name="root">
          <xs:complexContent>
             <xs:extension base="tns:root">
                <xs:sequence>
                   <xs:element name="b" type="xs:string"/>
                </xs:sequence>
             </xs:extension>
          </xs:complexContent>
       </xs:complexType>
    </xs:redefine>

  I thought the element 'b' was defined in the target namespace,
but it seems it has to be in no namespace.  Is it really the
intent of the spec?  Did I do anything wrong?  I use Xerces 2 for
Java (as in Java SE 6.)

  redefine.xsd:

    <?xml version="1.0" encoding="UTF-8"?>

    <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
               xmlns:tns="tns"
               targetNamespace="tns"
               elementFormDefault="qualified">

       <xs:element name="root" type="tns:root"/>

       <xs:complexType name="root">
          <xs:sequence>
             <xs:element name="a" type="xs:string"/>
          </xs:sequence>
       </xs:complexType>

    </xs:schema>

  redefine-red.xsd:

    <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
               targetNamespace="tns"
               xmlns:tns="tns">

       <xs:redefine schemaLocation="redefine.xsd">
          <xs:complexType name="root">
             <xs:complexContent>
                <xs:extension base="tns:root">
                   <xs:sequence>
                      <xs:element name="b" type="xs:string"/>
                   </xs:sequence>
                </xs:extension>
             </xs:complexContent>
          </xs:complexType>
       </xs:redefine>

       <xs:element name="redefined" type="tns:root"/>

    </xs:schema>

  Valid redefine.xml:

    <tns:redefined xmlns:tns="tns">
       <tns:a>bla</tns:a>
       <b>bla</b>
    </tns:redefined>

  Expected valid redefine.xml (but actually invalid):

    <redefined xmlns="tns">
       <a>bla</a>
       <b>bla</b>
    </redefined>

  Regards,

--
Florent Georges
http://www.fgeorges.org/






















   

[Michael Kay]

> I thought the element 'b' was defined in the target namespace, but it
seems it has to be in no namespace.

Just as if you didn't use redefine, your local element declaration will only
be in the target namespace if you specify elementFormDefault="qualified" on
your xs:schema element. (Or, of course, form="qualified" on the local
element declaration itself, but no-one ever does that.)

> Is it really the intent of the spec?

Yes. I've never heard an adequate explanation of why this extraordinary
decision was made, but in practice everyone changes the default by using
elementFormDefault="qualified".

Regards,

Michael Kay
http://www.saxonica.com/
http://twitter.com/michaelhkay 

 

[Florent Georges-3]

Michael Kay wrote:

  Hi,

> > I thought the element 'b' was defined in the target
> > namespace, but it seems it has to be in no namespace.

> Just as if you didn't use redefine, your local element
> declaration will only be in the target namespace if you specify
> elementFormDefault="qualified"

  <blush>Ouch!</blush>  It's been a long time since I've been
fooled by this one.  I guess I was focused on the redefine...
Sorry for the noise and thanks for the quick answer!

  Regards,

--
Florent Georges
http://www.fgeorges.org/

[Henry S. Thompson]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Michael Kay writes:

> Yes. I've never heard an adequate explanation of why this extraordinary
> decision was made, but in practice everyone changes the default by using
> elementFormDefault="qualified".

At the time the spec. was written, the community was pretty evenly
divided between people who thought the attribute model (i.e. not in a
namespace as such) was right for leaf node elements as well, and those
who thought that all elements should be qualified.

Without going into tedious and confidential details about
personalities and working group discussions, suffice it to say that
the design we have was the only one that included both possibilities
which the WG could reach consensus on.

ht
- --
       Henry S. Thompson, School of Informatics, University of Edinburgh
                         Half-time member of W3C Team
      10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440
                Fax: [phone], e-mail: ht@...
                       URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFK5Y5ekjnJixAXWBoRAh2UAJ4gBVgrOq4C+DK22kFTJNRvnLkSFwCfZnsJ
jgNXeNgag92u+CLvC5gXK9o=
=yYHE
-----END PGP SIGNATURE-----

[Florent Georges-3]

Henry S. Thompson wrote:

  Hi,

> Without going into tedious and confidential details about
> personalities and working group discussions, suffice it to say
> that the design we have was the only one that included both
> possibilities which the WG could reach consensus on.

  Yes, but that does not answer the interesting question: why the
chosen default value ;-)

  Regards,

--
Florent Georges
http://www.fgeorges.org/

[Henry S. Thompson]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Florent Georges writes:

> Henry S. Thompson wrote:
>
>> Without going into tedious and confidential details about
>> personalities and working group discussions, suffice it to say
>> that the design we have was the only one that included both
>> possibilities which the WG could reach consensus on.
>
>   Yes, but that does not answer the interesting question: why the
> chosen default value ;-)

That was part of the compromise.

ht
- --
       Henry S. Thompson, School of Informatics, University of Edinburgh
                         Half-time member of W3C Team
      10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440
                Fax: [phone], e-mail: ht@...
                       URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFK5beVkjnJixAXWBoRAoT5AJ4yFDcwJ5ffeptvNEzg6ORziuap5gCdFrbr
df9/0oDB1kNHtEXZPaHqEks=
=ciuI
-----END PGP SIGNATURE-----

[noah_mendelsohn]

Henry Thompson writes:

> That was part of the compromise.

Stated differently:  you can infer that there were a significant number of
people involved who strongly took the opposite position from yours, and
the default you see was in deference to their views of what the community
needed.  As already noted, it was a long and difficult discussion, with
strongly held opinions on both sides.  I don't think anyone thought the
resulting design was particularly appealing, but given the strongly
divergent opinions about what users needed and/or about how namespaces
were supposed to be used, the design had the advantage of letting users
choose.  FWIW:  I tend to sympathize with George's feeling that most users
do want the qualified idiom, that qualified would have been a better
default, etc.  Still, it's way too late to reopen it now, and there are
schemas out in the wild that use both options as far as I know.

Noah

--------------------------------------
Noah Mendelsohn
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------








ht@... (Henry S. Thompson)
Sent by: xmlschema-dev-request@...
10/26/2009 10:52 AM
 
        To:     lists@...
        cc:     Michael Kay <mike@...>, xmlschema-dev@...,
(bcc: Noah Mendelsohn/Cambridge/IBM)
        Subject:        Re: redefine and target namespace


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Florent Georges writes:

> Henry S. Thompson wrote:
>
>> Without going into tedious and confidential details about
>> personalities and working group discussions, suffice it to say
>> that the design we have was the only one that included both
>> possibilities which the WG could reach consensus on.
>
>   Yes, but that does not answer the interesting question: why the
> chosen default value ;-)

That was part of the compromise.

ht
- --
       Henry S. Thompson, School of Informatics, University of Edinburgh
                         Half-time member of W3C Team
      10 Crichton Street, Edinburgh EH8 9AB, SCOTLAND -- (44) 131 650-4440
                Fax: [phone], e-mail: ht@...
                       URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged
spam]
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFK5beVkjnJixAXWBoRAoT5AJ4yFDcwJ5ffeptvNEzg6ORziuap5gCdFrbr
df9/0oDB1kNHtEXZPaHqEks=
=ciuI
-----END PGP SIGNATURE-----