Re: Help on XML Schema generation using XSLT

Hi Austin,
Vielen Dank :) for your tips. I agree absolutely to the concerns you have mentioned. I do see this problem along with similar issues like identifying enumerations, identifying decimal values that are based on European conventions, occurances, set of values and the list goes on.
I have no idea of how the xml would be and the limits of its existance. It is highly specific to the customers environment. But on the other hand, I hae the possibility to refine my XSLT once I treat succesive documents. hence my Schema would get better and better every time it is loaded with the same type of XML file at that customers place.

At the moment I am working on the XSLT and I do land on issues where it is misty. I try to find my solutions, I think I would post some code where I am stuck and cant get further.

But thank you for giving some hopes that it is possible to generate a schema using a non-procedural language like XSLT.
cheers
Domnic

--- On Fri, 20/8/10, Cheney, Edward A SSG RES USAR USARC <[email protected]> wrote:

> From: Cheney, Edward A SSG RES USAR USARC <[email protected]>
> Subject: Re: Help on XML Schema generation using XSLT
> To: "Silent lights" <[email protected]>
> Cc: [email protected]
> Date: Friday, 20 August, 2010, 1:58
> Densil,
> 
> I would say converting a basic XML document to a schema
> document is not probable unless there exists a certain
> quantity of known information that you are aware of outside
> the XML instance.  I say this because there are many
> details associated with a schema that cannot be inferred
> from a single XML instance, such as how many times a certain
> child should occur under a parent element or if a certain
> parent element should be confined to only a limited
> specified choice of child elements.
> 
> The only way that such a task would become probable is if
> the XML document is auto-generated with known and defined
> meta-data and structure.  This is so because then you
> known exactly what all the bounds of those generated
> instances of XML should be limited to and allow expression
> of, which is the breath of information necessary to write a
> schema.
> 
> If you would supply additional information about the nature
> of the XML instances and the method by which they are
> generated and the depth of their consumption I believe
> anybody on this mailing list could supply you with tips to
> achieve a highly efficient solution.
> 
> Austin
>