Re: [xsl] Using 'validation' in xsl:result-document

Subject: Re: [xsl] Using 'validation' in xsl:result-document
From: a kusa <akusa8@xxxxxxxxx>
Date: Tue, 10 Nov 2009 16:33:11 -0600
Thank you. So it is a correct understanding that it the
result-document goes after the schema in
<xsl:import-schema schema-location="test.xsd"/> to validate the document?


On Tue, Nov 10, 2009 at 4:13 PM, Michael Kay <mike@xxxxxxxxxxxx> wrote:
>>
>> Can anyone tell me what exactly the options (strict, lax, strip,
>> preserve)  for attribute validation under element
>> <xsl:result-document> mean?
>
> The spec says *exactly* what they mean, but I expect you really want to
know
> in plain language what they mean.
>
> strict means the system must find an element declaration in a schema for
the
> named element at the root of your result document, and must validate the
> result document against that declaration, throwing an error if it's found
to
> be invalid.
>
> lax means the system should try to find an element declaration matching the
> root element name; if it finds one, it must perform validation against this
> element and throw an error if it's invalid; if it doesn't find one, then it
> doesn't perform validation (except that it keeps looking for schema
> declarations that match elements further down the tree).
>
> preserve means output the result tree with type annotations as they are
> (some branches of the tree might have been validated while they were being
> constructed, other branches might be untyped).
>
> strip means remove all type annotations from the result tree.
>
>>
>> When I use validation=strict, I get XTTE1550 error in XML Spy.
>
> That is most likely to mean that your result tree is invalid against the
> schema. It could also mean that the schema is invalid, or that it does not
> contain an element declaration for the outermost element of your result
> document.
>
> To debug this, if the error messages aren't helpful, you might find it
> useful to output the result document with validation="strip", and then
> validate it against the schema "by hand".
>
> Michael Kay
> Saxonica

Current Thread