Peter, Using a key pair (namespaceURI and schemaId) will cause a big overhead in document validation. The XML document does not know anything about the schema id, we only have the namespace URI information (through the prefix). So, in that case the validation process will be slowed down significantly.
Khaled "Peter A. Volchek" wrote: > Khaled, > > If I understood your strategy right, you are using the namespaceUri as a key > to get a schema grammar. > But why not use key as pair( namespaceUri, schemaId ) ? > This uniquely identifies the schema to use as well as allows to prevent the > circular imports. > > Peter. > > ----- Original Message ----- > From: "Khaled Noaman" <[EMAIL PROTECTED]> > To: <[EMAIL PROTECTED]> > Sent: Thursday, October 17, 2002 11:18 PM > Subject: Re: schemaLocation question > > > This is allowed. The schema spec says that the schemaLocation is only a > > hint and it is open to application to ignore it for a given namespace if > > that namespace was previously resolved. We chose to implement it that > > way to eliminate the need for duplicate processing of the same namespace > > and to prevent endless processing in the case of circular imports (i.e A > > imports B which imports A). > > > > Khaled > > > > "Peter A. Volchek" wrote: > > > > > I have the question related to "schemaLocation" attribute > > > processing.As I noticed in your code, when parser meeths the > > > schemaLocation attribute with the uri that was already defined > > > previously it just gets the schema grammar from cache and uses > > > it. This causes the trouble when validating the following instance > > > document: > a.xml----------------------------------------------------------------------- > ------<ns:A > > > xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"; > > > xmlns:ns="http://www.sample.com"; > > > xsi:schemaLocation="http://www.sample.com sch1.xsd"> > > > > > > <ns:B xsi:schemaLocation="http://www.sample.com > > > sch2.xsd">Hello</ns:B> > > > > > > > > > > /ns:A>---------------------------------------------------------------------- > ------- In > > > this document the element A is declared in in schema sch1.xsd, and > > > element B in sch2.xsdas > > > follows: > sch1.xsd-------------------------------------------------------------------- > ---------<?xml > > > version="1.0"?> > > > <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"; > > > targetNamespace="http://www.sample.com";> <xs:element > > > name="A"> > > > <xs:complexType> > > > <xs:sequence> > > > <xs:any namespace="http://www.sample.com"; minOccurs="1" > > > maxOccurs="unbounded"/> > > > </xs:sequence> > > > </xs:complexType> > > > > > > </xs:element> > </xs:schema>---------------------------------------------------------------- > ------------- > sch2.xsd-------------------------------------------------------------------- > ---------<?xml > > > version="1.0"?> > > > <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"; > > > targetNamespace="http://www.sample.com";> <xs:element > > > name="B" > > > type="xs:string"/> > </xs:schema>---------------------------------------------------------------- > ------------- Now, > > > if you parse with DOMPrint sample you'll get the following error: > > > > DOMPrint.exe -v=always -n -s a.xmlError at file "a.xml", line 5, > > > column 61 > > > Message: Unknown element 'ns:B' THe reason is that parser ignores > > > the second schemaLocation attribute, meaning that the grammar already > > > exists for it in cache.Is this expected behaviour? What does XMLSchema > > > rec. says about it ? Peter A. Volchek > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
