[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: abstract types

• To: "Maciejewski, Thomas" <Thomas.Maciejewski@l...>
• Subject: Re: abstract types
• From: ht@c... (Henry S. Thompson)
• Date: 12 Sep 2002 09:45:37 +0100
• Cc: "'xml-dev@l...'" <xml-dev@l...>
• In-reply-to: <7723AD8015FDD411B13F00508BE31C140ECED03A@e...>
• References: <7723AD8015FDD411B13F00508BE31C140ECED03A@e...>
• User-agent: Gnus/5.0808 (Gnus v5.8.8) XEmacs/21.2(beta13) (Demeter)

"Maciejewski, Thomas" <Thomas.Maciejewski@l...> writes:

> Is there a way to enforce abstract types.  For example lets say I have an
> interface : printable  
> 
> but then I have a concrete "class" wordDocument that derives from printable
> with some additional information.  Lets say I would like to serialize these
> objects to XML.  Lets say I also have a class queue that can have many
> "printable" objects ... 
> 
> so in my xml I would like to have :
> 
> <foo>
> 	<wordDocument name="toms document" />
> 	<excelDocument name="toms spreadsheet" />
> </foo>
> 
> but not allow:
> <foo>
> 	<printable />
> </foo>

There are serveral ways to do this.  Most straightforward is

1) Declare printable as an abstract element at the top-level with a
type which covers all the possibilities for your concrete classes;

2) Declare foo to have printable* as its content;

3) Declare wordDocument and excelDocument at the top level with
substitutionGroup='printable' and types derived from the type of
printable.

You're done.

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
          W3C Fellow 1999--2002, part-time member of W3C Team
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht@c...
		     URL: http://www.ltg.ed.ac.uk/~ht/
 [mail really from me _always_ has this .sig -- mail without it is forged spam]

• References:
  • abstract types
    • From: "Maciejewski, Thomas" <Thomas.Maciejewski@l...>

• Prev by Date: Re: Combining 2 XML into 1 XML via XSLT
• Next by Date: XML SIG 17 September: Elliotte Rusty Harold on What's Wrong with XML APIs (and How to Fix Them)
• Previous by thread: abstract types
• Next by thread: RE: abstract types
• Index(es):
  • Date
  • Thread