Altova Mailing List Archives>Archive Index >microsoft.public.xml Archive Home >Recent entries >Thread Prev - Deciding XML structure [Thread Next] Re: Deciding XML structureTo: NULL Date: 8/5/2007 1:22:00 PM
Mark wrote:
> Basic question (I think); when is it best to use one element with
> attributes vs. one without and child elements? For example:
> <book title="Some Book" author="Anderson, Jones"></book>
> versus:
>
> <book>
> <title>Some Book</title>
> <author>Anderson</author>
> <author>Jones</author>
> </book>
>
> Is there a rule or best practice?
This is a FAQ: see http://xml.silmaril.ie/developers/attributes/
My personal preference is to use attributes for numeric, coded, or
categorical data, and elements containing PCDATA for other textual
material, so in this case your second option is better, IMHO.
But there isn't anything wrong with doing it the other way round: just
unusual. More to the point (to make the distinction) would be:
<book isbn="123456789" pub.date="2007-08-05">
<title>Some Book</title>
<authorgroup>
<author>Anderson</author>
<author>Jones</author>
</authorgroup>
</book>
(It's usually A Good Idea to group together consecutive elements of the
same type so that later processes can access them as a block without
having to detect whether there is one of them or many.)
///Peter
--
XML FAQ: http://xml.silmaril.ie/
| ||||||
| Company | Legal | Press | Partners | Careers | Sitemap | Contact Us | Altova Blog | Mobile | Full Site | |||
|
