[
Lists Home |
Date Index |
Thread Index
]
Dare, I do not think you can say that XPath data model is the standard
data model for all applications. I know you have done work on querying XML
Schemas using XPath etc.
SAX is also a data model, it provides a streaming API, it has different
operations we can use to access the data, DOM is another data model for
XML, and so on. XPath/XQuery data model is just one of the data models, it
can probably be used for several applications, but it is not right to say
that it is to be used for all applications..
Anyways, I guess we might differ here. I do not think I will accept that
XPath/XQuery data model is *the* data model for all XML applications..
regards - murali.
I think the confusion is because relational representations are only for
one particular class of applications (database), whereas XML
representations encompass multiple classes of applications..
I would like to see an XPath query which will say given an XML schema and
an element tag, I want to determine all element tags that can possibly
occur in any valid instance document as descendants..
On Sat, 24 May 2003, Dare Obasanjo wrote:
> a.) The data model is a foundation upon which application semantics
> can be built. I can say similar things about how the relational
> model's quaint notions of tuples and keys does not make it easy to
> answer questions like "Who is my best customer?". However this doesn't
> mean I can't create databases that hold my customer information nor
> can I create queries that answer such questions. Similarly, it doesn't
> mean once I have done this my application semantics will directly
> expose the underlying structure of my database or the underlying SQL
> queries being run.
> b.) Since XML Schema documents are constructed as XML as long as I
> have all the input schemas accessible as documents I can construct
> queries in XQuery that can retrieve information like "find all
> possible element tags that could be descendants of an element"
> actually I'm sure you meant to say find all elements whose type
> definition is a descendent of the type definition of another element
> since XPath provides a trivial way to get the descendants of an
> element.
>
> ________________________________
>
> From: Murali Mani [mailto:mani@CS.UCLA.EDU]
> Sent: Sat 5/24/2003 10:29 AM
> To: xml-dev@lists.xml.org
> Subject: RE: Re: [xml-dev] Syntax + object model
>
>
>
>
> Because I am working on data models, I will share my two cents worth of
> impressions..
>
> First off, data model consists of 3 parts: structures, constraints, and
> operations.
>
> You can standardize such data models only for applications.
>
> The XQuery/XPath data model is for applications that want to retrieve data
> from XML documents. The operations are defined using XPath/XQuery.
>
> But this does not make it a standard for all applications, we can write
> xml schemas also as XML. Do the same model and operations as defined in
> XPath/XQuery suitable? probably not..
>
> It probably is not useful to define, for example, what are the children of
> a complex type? it could be xsd:sequence.. what do you get from it?
>
> Similarly, suppose you want to write a query like: I want to find all
> possible element tags that could be descendants of an element? It is
> definitely not easy to write such a query using XPath..
>
> So when you want to analyze XML Schemas, you want to probably come up with
> a different data model, and probably a different set of operations..
>
> Conclusion: I believe it is not right to say XPath/XQuery data model is
> *the* data model for all XML documents..
>
> cheers and regards - murali.
>
>
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
>
> The list archives are at http://lists.xml.org/archives/xml-dev/
>
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://lists.xml.org/ob/adm.pl>
>
>
>
|
