W3C meta-schema - useless?

[Ian Pilcher]

I was just wondering if anyone has ever been able to use the W3C meta-
schema (from http://www.w3.org/2001/XMLSchema) to actually validate a
schema, including the meta-schema itself.

The tools I've tried (xmllint from libxml2 and whatever parser Sun's
Java 1.5.0 uses) barf at various points. Is the meta-scehma flawed, or
does it just push the parsers too hard?

TIA

 

[Martin Honnen]

I was just wondering if anyone has ever been able to use the W3C meta-
schema (from http://www.w3.org/2001/XMLSchema) to actually validate a
schema, including the meta-schema itself.

It has been some time I have looked at that but my understanding is that
the schema serves some documentary purpose but validating against it
does not work as there is some bootstrapping necessary. However schema
parsers usually allow you to compile a schema and then give you errors
if the schema is not valid, there is no need to try to validate against
that meta schema if you want to check one of your own schemas.

 

[Henry S. Thompson]

I was just wondering if anyone has ever been able to use the W3C meta-
schema (from http://www.w3.org/2001/XMLSchema) to actually validate a
schema, including the meta-schema itself.

Yes -- XSV [1] uses it to validate every schema document that comes in
for use.

The tools I've tried (xmllint from libxml2 and whatever parser Sun's
Java 1.5.0 uses) barf at various points. Is the meta-scehma flawed, or
does it just push the parsers too hard?

Surprised that Xerces doesn't accept the sForS -- error messages?

ht

[1] http://www.ltg.ed.ac.uk/~ht/xsv-status.html
--
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: (e-mail address removed)
URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]

 

[Henry S. Thompson]

It has been some time I have looked at that but my understanding is
that the schema serves some documentary purpose but validating against
it does not work as there is some bootstrapping necessary.

The W3C XML Schema REC makes clear that all schema documents must be
schema-valid with respect to the sForS. . .

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: (e-mail address removed)
URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]

 

[Ian Pilcher]

Surprised that Xerces doesn't accept the sForS -- error messages?

UndeclaredPrefix: Cannot resolve 'xs:string' as a QName: the prefix 'xs'
is not declared.

I get this error whether I'm trying to validate my own schema or the
meta-schema itself.

 

[Henry S. Thompson]

UndeclaredPrefix: Cannot resolve 'xs:string' as a QName: the prefix 'xs'
is not declared.

That's downright weird -- needless to say there's a namespace
declaration for 'xs' at the top of
http://www.w3.org/2001/XMLSchema.xsd -- can anybody else replicate
this?

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: (e-mail address removed)
URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]