Problems with generating enumLeafs

[duschata]

I'm using this schema file
`
<?xml version="1.0" encoding="UTF-8"?>
<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"
xmlns="http://foo.bar.EnumLeafSchema"
targetNamespace="http://foo.bar.EnumLeafSchema"
elementFormDefault="qualified">

<xsd:element name="UsesEnumLeaf">
<xsd:complexType>
<xsd:sequence>
<xsd:element name="personTitle" type="PersonTitleCodesEnumeration"></xsd:element>
<xsd:element name="personName" type="xsd:string"></xsd:element>
</xsd:sequence>
</xsd:complexType>
</xsd:element>

<xsd:simpleType name="PersonTitleCodesEnumeration">
<xsd:restriction base="xsd:string">
<xsd:enumeration value="mr" />
<xsd:enumeration value="ms" />
<xsd:enumeration value="mrs" />
<xsd:enumeration value="miss" />
<xsd:enumeration value="dr" />
</xsd:restriction>
</xsd:simpleType>

</xsd:schema>
`

This is generated by the compiler:

var enumleafschema_bar_foo_Module_Factory = function () {
  var enumleafschema_bar_foo = {
    name: 'enumleafschema_bar_foo',
    defaultElementNamespaceURI: 'http:\/\/foo.bar.EnumLeafSchema',
    typeInfos: [{
        localName: 'UsesEnumLeaf',
        typeName: null,
        propertyInfos: [{
            name: 'personTitle',
            required: true
          }, {
            name: 'personName',
            required: true
          }]
      }, {
        type: 'enumInfo',
        localName: 'PersonTitleCodesEnumeration',
        values: ['mr', 'ms', 'mrs', 'miss', 'dr']
      }],
    elementInfos: [{
        elementName: 'UsesEnumLeaf',
        typeInfo: '.UsesEnumLeaf'
      }]
  };
  return {
    enumleafschema_bar_foo: enumleafschema_bar_foo
  };
};

But I expected:

localName: 'UsesEnumLeaf',
        typeName: null,
        propertyInfos: [{
            name: 'personTitle',
            required: true,
//This was missing in the generated mapping
            typeInfo: '.PersonTitleCodesEnumeration'
          }, {
            name: 'personName',
            required: true
          }]

Is this a bug? My failure?

Kind regards,

Tom

[highsource]

See this:

https://github.com/highsource/jsonix-schema-compiler/blob/master/compiler/src/main/java/org/hisrc/jsonix/compilation/mapping/CreateTypeInfoDeclarationVisitor.java#L167

This is where the typeInfo declaration is generated. If it gives String then it's completely omitted. I'd try return createTypeInfoDeclaration(info); there. I actually have no idea why I didn't do it before. I'm puzzled, but I think it should work.

[duschata]

I checked a couple of .xsd files containing enums and everything seems to work as I expected. Thanks a lot!

[highsource]

Hi,

unfortunately this does not solve the problem completely yet. This works for strings but other types won't work well. We need Java-side conversion between lexical and JSON representation to render enum values correctly.
So don't close it yet.

Best wishes,
Alexey

[benmarch]

Hi there,

I am running into a bit of an issue with this myself. I am able to get the enum values to show up in the mapping.js file, but not in the jsonschema file. Here is what I am seeing in the mapping file:

{
    type: 'enumInfo',
    localName: 'ClassEnum',
    baseTypeInfo: 'NormalizedString',            //I have tried "string" as well
    values: ['standard', 'steps', 'example']
}

And here is the jsonschema file:

"ClassEnum":{
    "allOf":[
        {
            "$ref":"http://www.jsonix.org/jsonschemas/w3c/2001/XMLSchema.jsonschema#/definitions/normalizedString"
        }
    ],
    "typeType":"enumInfo",
    "typeName":{
        "localPart":"classEnum",
        "namespaceURI":"http://blahblahblah"
    }
}

It would be awesome if the schema could have the "enum" property as well:

"ClassEnum":{
    "allOf":[
        {
            "$ref":"http://www.jsonix.org/jsonschemas/w3c/2001/XMLSchema.jsonschema#/definitions/normalizedString"
        }
    ],
    "typeType":"enumInfo",
    "typeName":{
        "localPart":"classEnum",
        "namespaceURI":"http://blahblahblah"
    },
    "enum": ["standard", "steps", "example"]      //this would be the same format as "values" in the mapping except with double quotes
}

Is this something that can be done easily? I could take a stab at a PR if you can point me in the right direction.

Thanks,
Ben

[highsource]

It's not quite easy but doable.

Currently, the compiler just associates XML Schema types with names of Jsonix type. That's a simple mapping.

What we need to do is to replace string names with something more intelligent, something like "type descriptor". Type descriptor should be able to parse lexical representations of types and convert then in JSON structures which could then used in mappings and JSON schemas.

You can help by providing sample enumerated simple types for all the simple types for test:

https://github.com/highsource/jsonix-schema-compiler/blob/master/compiler/src/test/resources/basic/zero/schema.xsd

Next, I'll implement type descriptors for a few type and you could help by implementing further type descriptors. Essentially, we have to cover all of the supported types (from CreateTypeInfoDeclarationVisitor).

[benmarch]

Thank you for the quick response! I will create a PR with the enum type additions to the sample XSD, but I just want to clarify what you are asking. If I understand correctly, you want me to create a simpleType with a restriction for each of the ~40 simple types? So I would basically stamp out a bunch of these with different base types and values:

<xs:element name="enum" type="enumType"/>
<xs:simpleType name="enumType">
    <xs:restriction base="xs:token">
        <xs:enumeration value="Male"/>
        <xs:enumeration value="Female"/>
    </xs:restriction>
</xs:simpleType>

If that is correct then I can start right away, I was just concerned about the file size blowing up.

[highsource]

Right, that would help. Please name them tokenEnum for xs:token etc.
That would help enormously,

[benmarch]

Absolutely, I'll have that done shortly.

[benmarch]

Sorry about the duplicate commit, I didn't have my repo configured correctly and it used my work user by mistake. The second one is correct and the corresponding PR is #70.