- From: Scott, Michael Gordon <Michael.Gordon.Scott@kla-tencor.com>
- Date: Tue, 18 Mar 2003 10:11:28 -0800
- To: "'Jeni Tennison'" <jeni@jenitennison.com>, "Scott, Michael Gordon" <Michael.Gordon.Scott@kla-tencor.com>
- Cc: Sam Carleton <sam@linux-info.net>, xmlschema-dev <xmlschema-dev@w3.org>
Jeni, I was afraid of that. Thanks for all your insight. It's been extremely helpful, and very educational. I'll start using <xs:*> instead of <xsd:*>. I've seen both used in the MSMXL Parser SDK (which I only use for syntax help). Michael -----Original Message----- From: Jeni Tennison [mailto:jeni@jenitennison.com] Sent: Tuesday, March 18, 2003 9:59 AM To: Scott, Michael Gordon Cc: Sam Carleton; xmlschema-dev Subject: Re: allowing zero to unbounded elements in any order? Hi Michael, > Actually, NumWafers and Wafers are mandatory, they just want them in > any order. It's everything in : > > <xsd:extension base="Type_ToolLotSummary"> > > The Type_ToolLotSummary, which is above <xs:all> that is optional > (everyting), and also needed to be in any order. I'm sorry, I overlooked the fact that you were extending the type. <xs:all> can't be used in that context because when you extend a type you effectively create a sequence of the content from the base type followed by the content from the extension, and <xs:all> isn't allowed within a sequence. There's no way to build a type by extension and allow all the elements (from both the base type and the extension) to be present in any order. Sorry that I didn't spot that sooner. > Dumb question : Is <xs:???> and <xsd:???> the same thing? Yes; I tend to use <xs:*> because that's what's used in the XML Schema Rec, and because it's shorter. Cheers, Jeni --- Jeni Tennison http://www.jenitennison.com/
Received on Tuesday, 18 March 2003 13:11:44 UTC