W3C home > Mailing lists > Public > xmlschema-dev@w3.org > November 2004

Re: non deterministic any element???

From: Xan Gregg <xan.gregg@jmp.com>
Date: Mon, 29 Nov 2004 10:52:54 -0500
Message-Id: <BBA8D3EF-421E-11D9-80EA-000A95B51846@jmp.com>
Cc: <xmlschema-dev@w3.org>
To: "Dean Hiller" <dean@xsoftware.biz>

Schema B is non-deterministic because after encountering a 
<callbackLocation> element it will accept either an element matching 
the "NewElementInVersion2" element particle or an element matching the 
"any" particle.  Since < NewElementInVersion2> matches both of those 
possibilities, it is ambiguous.  (There is no "unique particle" 
attributed to the element, so it violates the "Unique Particle 
Attribution" constraint of XML Schema.)

Same for schema C.

B and C would be OK in XML Schema 1.1 as proposed in the public Working 
Draft [1] because wildcard particles will be subordinate to explicit 
element particles. (RQ-36i)

xan

[1] http://www.w3.org/TR/2004/WD-xmlschema11-1-20040716/structures.html


On Nov 25, 2004, at 7:44 PM, Dean Hiller wrote:
> After reading the Versioning XML Languages Draft, I am a bit confused 
> on why the any element is non-deterministic.  Below, I whipped 
> together 3 schemas.  Say A is schema version1, B is schema 
> version2(same namespace), and C is companyX extending version 1.  Why 
> would the xml below the schema's be non-deterministic???  ie. if I 
> throw doc B at a program that only knows schema A or throw doc C at a 
> program that only knows schema B, etc.  thanks for any help in 
> understanding here.  I am really just trying to understand section 
> 9.1(http://www.w3.org/2001/tag/doc/versioning-20031003#d0e971)
Received on Monday, 29 November 2004 15:52:56 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 23:40:23 UTC