W3C home > Mailing lists > Public > xmlschema-dev@w3.org > September 2006

Re: Redefine and Import used together - is this valid?

From: Danny Vint <dvint@sack.dreamhost.com>
Date: Tue, 19 Sep 2006 08:52:43 -0700 (PDT)
To: "Henry S. Thompson" <ht@inf.ed.ac.uk>
Cc: xmlschema-dev@w3.org, xml-dev@lists.xml.org
Message-ID: <Pine.LNX.4.62.0609190847110.19085@gimli.dreamhost.com>

So your saying XSV is correct because a differnt rule/loophole allows it 
to ignore the problem causing statement. I guess this si similar to 
Michael Kaye's statement about different parsers having different 
workarounds.

So as long as I find tools that "workaround this problem" in some manner 
than I'm ok? I don't really like that answer either. I can live with 
problems in interpretation of the spec, but now it looks like the spec is 
creating loopholes if you want to take advanatage of them. I really want 
to provide the functionality as described in the earlier message, but I 
can't rely on this working or consistently quote the spec to say this is 
why some tool is wrong.

..dan

---------------------------------------------------------------------------
Danny Vint

Specializing in Panoramic Images of California and the West
http://www.dvint.com

Voice:510:522-4703
FAX: 801-749-3229

On Tue, 19 Sep 2006, Henry S. Thompson wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Danny Vint writes:
>
>> I have the following situation:
>>
>> 1) Base industry standard schema (ACORD)
>> 2) A schema that imports the ACORD schema (to reuse data types and
>> some elements) that defines my organizations new elements and
>> aggregates (ACME)
>> 3) A schema that redefines #1 ACORD to modify existing elements and
>> aggregates to include my new ACME elements.
>>
>> I then have a docuemnt instance the references #3.
>>
>> Xerces and XSV say my document and schemas are valid.
>
> XSV accepts your package because it never processes a namespaced
> schema document more than once if it can avoid it, so the nested,
> potentially problematic, import in (2), does not cause a problem
> because the ACORD schema document has already been processed (during
> the processing of (3), so the 'schemaLocation' of the import in (2) is
> ignored).
>
> This is conformant behaviour, because the REC says that
> 'schemaLocation' on <xs:import> is a hint, which need not be followed.
>
> But it's not _required_ behaviour, for better or worse.
>
> I'd like to see a way for users to mandate such behaviour. . .
>
> ht
> - --
> Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
>                     Half-time member of W3C Team
>    2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
>            Fax: (44) 131 650-4587, e-mail: ht@inf.ed.ac.uk
>                   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)
>
> iD8DBQFFD9YxkjnJixAXWBoRAhs3AJ4u9n2bQD/1nkYbP4YcnZ2PE9IXxgCeO4CY
> FiqsAOh2YY5s2Dfk/Y3uq1A=
> =3fsp
> -----END PGP SIGNATURE-----
>
Received on Tuesday, 19 September 2006 15:53:10 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 11 January 2011 00:14:55 GMT