Hi,
I find that the xml-maven-plugin beta-1 requires us to specify the
schema file at the <systemId> element, even if the schema location is
specified in the xml file.
I don't know if it is a designed behavior, but it is annoying to use
when there are a lot of different types of XML files to be validated.
In the source code of org/codehaus/mojo/xml/ValidateMojo.java, there
is a function "newSAXParserFactory" :
private SAXParserFactory newSAXParserFactory( ValidationSet
pValidationSet ) {
SAXParserFactory spf = SAXParserFactory.newInstance();
spf.setValidating( pValidationSet.isValidating() );
spf.setNamespaceAware( true );
return spf;
}
It does not turn on the schema validation of the SAX parser. I tried
to change the method :
private SAXParserFactory newSAXParserFactory( ValidationSet
pValidationSet ) throws SAXException, ParserConfigurationException
{
SAXParserFactory spf = SAXParserFactory.newInstance();
spf.setValidating( pValidationSet.isValidating() );
if (pValidationSet.isValidating()) {
spf.setFeature("http://apache.org/xml/features/validation/schema";,
true);
}
spf.setNamespaceAware( true );
return spf;
}
Then the schema validation go well without specifying the schema file
at <systemId>.
It would become Xerces dependent, but the dependency is specified in
the POM anyway.
regards,
Herman.
---------------------------------------------------------------------
To unsubscribe from this list please visit:
http://xircles.codehaus.org/manage_email
