Preserve XSD annotations in bindings

[hcw70]

When preparing my XSD for PyXB binding, i would love to have access to the annotations of the type / attribute /element i am at in the XSD so i can inspect the bindings later on (while parsing the XML instance document) to see the annotations.

I would like i.e. to define an XSD annotation which states an attribute as a reference to another element so i can iterate over the instance document to resolve the reference later.

So in XSD i would like to write:

    <xs:complexType name="TVariable">
        <xs:attribute name="name" use="required"/>
        <xs:attribute name="default"/>
        <xs:attribute name="type" use="required">
            <xs:annotation>
                <xs:appinfo>
                    <dma:annotations>
                        <xsdTypeRef name="TTypeDefinition"/>
                    </dma:annotations>
                </xs:appinfo>
            </xs:annotation>
        </xs:attribute>
    </xs:complexType>

    <xs:group name="TPacketContent">
        <xs:sequence>
            <xs:element name="var" minOccurs="0" maxOccurs="unbounded" type="TVariable"/>
        </xs:sequence>
    </xs:group>

And in my code i would like to write:

import myxsd
doc=myxsd.CreateFromLocation(...)
for var in doc.var:
  if var._type.annotation.appinfo ... :   
    # look for source type
    # add attribute to var:
    var.type_resolved=findInDoc(doc,appinfo.xsdTypeRef())

(Roughly draftet so you hopefully get the idea).

The purpose of this is to annotate the schema so i can create a post-processing step over the instance XML document which looks at the annotations from the XSD to perform project-specific tasks like resolving references in the XML document or add specific consistency checks.

The reason to put it into the XSD is to have a single place where the datamodel is defined (XSD) and have a generic postprocessing based on PyXB and annotations which works for all data models.

[hcw70]

Currently i am able to work around this by misusing the doucmentation field:

from build import msgdb
doc=msgdb.CreateFromDocument(open("MessageDB-Example.xml").read())


print("Gives type doc:")
print(doc.__doc__)

print("gives type name:")
print(doc._element().typeDefinition()._ExpandedName.localName())

So i need to parse the type doc to look out for my "annotation" and handle that.