W3C home > Mailing lists > Public > w3c-ietf-xmldsig@w3.org > July to September 2001

RE: X509 Schema Tweaks (Was: XML Signature schema implementation)

From: Hallam-Baker, Phillip <pbaker@verisign.com>
Date: Thu, 27 Sep 2001 14:32:12 -0700
Message-ID: <2F3EC696EAEED311BB2D009027C3F4F4058697A2@vhqpostal.verisign.com>
To: "'reagle@w3.org'" <reagle@w3.org>, Tom Gindin <tgindin@us.ibm.com>, Peter Tornberg <tberg@x-obi.com>
Cc: xmldsig <w3c-ietf-xmldsig@w3.org>, bal@microsoft.com

If we are opening this up I would prefer to simply have a rule that we have
a type declared for every element.

It may look ugly in some folk's view, but the fault lies in XML schema. The
distinction between elements and element types is unnecessary, especially
since they invented abstract types. But they did wot they did.

		Phill


Phillip Hallam-Baker FBCS C.Eng.
Principal Scientist
VeriSign Inc.
pbaker@verisign.com
781 245 6996 x227


> -----Original Message-----
> From: Joseph Reagle [mailto:reagle@w3.org]
> Sent: Thursday, September 27, 2001 5:18 PM
> To: Tom Gindin; Peter Tornberg
> Cc: xmldsig; bal@microsoft.com
> Subject: X509 Schema Tweaks (Was: XML Signature schema implementation)
> 
> 
> On Friday 21 September 2001 06:34 pm, Tom Gindin wrote:
> >      By the way, all of the elements in this case except 
> for X509SKI are
> > plausible candidates for reuse.  X509Certificate, X509CRL, and
> > X509SubjectName are all more likely to be reused in another 
> spec than
> > X509IssuerSerial.
> 
> In that case, I think the tweaked schema would need to look 
> like [1]. This 
> wouldn't affect parser or schema validation performance I 
> don't think. It 
> does permit  people to borrow our natural language 
> specification of how 
> these things are encoded and such. However, it is rather 
> ugly, if someone 
> wants to re-use it, they could redefine/import them in a new 
> namespace, and 
> it divorces these element types from their context/meaning as 
> properties of 
> a single X509Data structure. 
> 
> These are all minor points, but given our late stage in the 
> game, I'd like 
> to hear more voices in support of this change...
> 
> 
> [1]  Tweaked X509DataType
> <complexType name="X509DataType">
>    <sequence maxOccurs="unbounded">
>       <choice>
>          <element ref="ds:X509IssuerSerial"/>
>          <element ref="ds:X509SKI"/>
>          <element ref="ds:X509SubjectName"/>
>          <element ref="ds:X509Certificate"/>
>          <element ref="ds:X509CRL"/>
>          <any namespace="##other" processContents="lax"/>
>       </choice>
>    </sequence>
> </complexType>
> 
> <element name="X509IssuerSerial" type="ds:X509IssuerSerialType"/>
> <element name="X509SKI" type="base64Binary"/>
> <element name="X509SubjectName" type="string"/>
> <element name="X509Certificate" type="base64Binary"/>
> <element name="X509CRL" type="base64Binary"/>
> 
> <complexType name="X509IssuerSerialType">
>    <sequence>
>       <element name="X509IssuerName" type="string"/>
>       <element name="X509SerialNumber" type="integer"/>
>    </sequence>
> </complexType>
> 



Received on Thursday, 27 September 2001 17:32:07 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:21:36 UTC