losing annotations when using XMLGrammarPreparser with DOMInputSource

[Nikolai Grigoriev <ng...@gmail.com>]

Hello!

I am trying to develop an app that will use XML schema with some additional
annotations. Annotations come in various forms, including the attributes
added to XML element definitions. I am using a different namespace for
these attributes. I know this is not exactly canonical for Schema but still
XML-legal :) Plus, I have found (thanks to this list) the feature "
http://apache.org/xml/features/generate-synthetic-annotations" that puts
all my custom stuff into the annotations that I can access via
XSElementDeclaration.getAnnotation().

Initially I have found a bit hacky way to load my schemas by following this
path:

    XSImplementation impl = new XSImplementationImpl();
    XSLoader loader = impl.createXSLoader(impl.createStringList(new
String[]{"1.0"}));
    ...
    loader.getConfig().setParameter("
http://apache.org/xml/properties/internal/entity-resolver", resolver);
    loader.getConfig().setParameter("
http://apache.org/xml/properties/internal/error-handler", errorHandler);
    loader.getConfig().setParameter("
http://apache.org/xml/features/generate-synthetic-annotations",  true);

    XSModel model = loader.loadURI("file://" + schemaFileName);

This works, I can use the API to find my element and its annotation. Then I
have realized that there are two ways to handle the annotation data - write
it to DocumentFragment or to SAX handler. There is more value for me in DOM
since I need to deal with some additional document elements anyway, so,
again following the suggestions from this list I have tried another
approach:

    DocumentBuilderFactory dbf = DocumentBuilderFactory.newInstance();
    dbf.setNamespaceAware(true);
    dbf.setValidating(false);

    DocumentBuilder db = dbf.newDocumentBuilder();
    ...setting entity resolver and error handler...
    Document doc = null;
    try ( BufferedReader br = new BufferedReader(new FileReader(modelFile))
) {
      doc = db.parse(new InputSource(br));
    } catch (IOException e) {
    ...
    }


    XMLGrammarPreparser preParser = new XMLGrammarPreparser();
    preParser.setEntityResolver(resolver);
    preParser.setErrorHandler(errorHandler);
    preParser.setFeature("
http://apache.org/xml/features/generate-synthetic-annotations",  true);
    preParser.setFeature("
http://apache.org/xml/features/validation/schema-full-checking", true);
    preParser.setFeature("http://apache.org/xml/features/validation/schema",
true);
    preParser.setFeature("
http://apache.org/xml/features/honour-all-schemaLocations", true);
    preParser.setFeature("
http://apache.org/xml/features/validate-annotations", false);
    preParser.setFeature("
http://apache.org/xml/features/generate-synthetic-annotations", true);
    preParser.registerPreparser(XMLGrammarDescription.XML_SCHEMA, null);

    DOMInputSource modelSource = new DOMInputSource(doc);
    XSGrammar grammar =
(XSGrammar)preParser.preparseGrammar(XMLGrammarDescription.XML_SCHEMA,
modelSource);

    XSModel model = grammar.toXSModel();


I do get the model here and I can use the same methods to find my element.
However, with this approach I get no annotations! Element's getAnnotation()
returns null. By the way, I do have also a real annotation for this element
(documentation). Using Xerces 2.11.0.SP1.

If needed, I can reduce my test schema and code to minimum and post here -
but maybe someone already knows the answer?

Thanks!




-- 
Nikolai Grigoriev