ENI (4) Fastexml compatibility issues

0. Introduction

Fasterxml is very easy to use for managing XML, but there are some issues that are not compatible with JAXB. Here are some examples of how to solve this handicaps

1. @XmlElementRef, @XmlMixed ...

Here is an example of the generated KeyInfoType class


 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
public class KeyInfoType {

    @XmlElementRefs({
        @XmlElementRef(name = "KeyName", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "KeyValue", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "RetrievalMethod", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "X509Data", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "PGPData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "SPKIData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false),
        @XmlElementRef(name = "MgmtData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
    })
    @XmlMixed
    @XmlAnyElement(lax = true)
    protected List<Object> content;


The possible solution is changing @XmlElementRef by @XmlElement

The proposal is:


 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
public class KeyInfoEduType {
        //Begin substitution
 @XmlElement(name = "KeyName", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected String keyName;
 
 @XmlElement(name = "KeyValue", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected KeyValueEduType keyValue; 
 
 @XmlElement(name = "RetrievalMethod", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected RetrievalMethodType retrievalMethod; 
 
 @XmlElement(name = "X509Data", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected X509DataEduType x509Data; 
 
 @XmlElement(name = "PGPData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected PGPDataType pGPData; 
 
 @XmlElement(name = "SPKIData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected SPKIDataType sPKIData; 
 
 @XmlElement(name = "MgmtData", namespace = "http://www.w3.org/2000/09/xmldsig#", type = JAXBElement.class, required = false)
 @Getter @Setter
 protected String mgmtData; 
 //End substitution

Note that I am using Lombok for avoiding Getters and Setters.
It is more verbose... but seems to work

But this transformation must affect all classes that use the @XmlElementRef annotation

2. Special class ObjtectType


In schemas, it is recommended to use STRONGLY typed elements... But this class is used against this principle.

The designer of the schema, has made this class a "jumble" (the Spanish of "cajón de sastre"?) for storing any type of objects.

After manually analysing some examples of ENI documents, a more typed class is proposed.

The initial class partial code is:


1
2
3
4
5
6
public class ObjectType {

    @XmlMixed
    @XmlAnyElement(lax = true)
    protected List<Object> content;

and the proposal is:


1
2
3
4
5
6
7
8
9
public class ObjectEduType {

    @XmlValue
    @Getter @Setter
    protected byte[] content;
 
    @XmlElement(name = "QualifyingProperties", namespace = "http://uri.etsi.org/01903/v1.3.2/#")
    @Getter @Setter
    protected QualifyingPropertiesEduType qualifyingProperties;


3. The loss of namespaces prefixes in attributes affected by @XmlElement


There are many attributes annotated by @XmlElement, that do not have namespaces prefixes when are serialized to XML.


A possible and verbose solution is to add the "namespace" attribute to the annotation.
NOTE: This CANNOT be done in the @Attribute annotation!!

Here is the RSAKeyValueType class before these changes:


1
2
3
4
5
6
public class RSAKeyValueType {

    @XmlElement(name = "Modulus", required = true)
    protected byte[] modulus;
    @XmlElement(name = "Exponent", required = true)
    protected byte[] exponent;


and after the addition of "namespace":


1
2
3
4
5
6
public class RSAKeyValueEduType {

 @XmlElement(name = "Modulus", namespace = "http://www.w3.org/2000/09/xmldsig#", required = true)
 protected byte[] modulus;
 @XmlElement(name = "Exponent", namespace = "http://www.w3.org/2000/09/xmldsig#", required = true)
 protected byte[] exponent;

4. Conclusion

This is a tedious job,  but if you make these changes to all the affected classes, you can have your project enabled for "fasterxml".

Comments

Post a Comment

Popular posts from this blog

ORVE WS (Dynamic) (4) Jackson XML mapper

0. Introduction I love Jackson parser as it is easy to use and can be used to with some data formats such as CSV ,  JSON , YAML and XML . Just define a wrapper and a class for marshalling/unmarshalling. Although JAXB is a standard, in the Java ecosystem exists many solutions. And my decision is to use Jackson. 1. Defining MAVEN dependencies To use Jackson XML it is required this dependency in Maven. Note that  ${jackson.version} represents a variable that contains preferably the latest Jackson version. 1 2 3 4 5 <dependency> <groupId> com.fasterxml.jackson.dataformat </groupId> <artifactId> jackson-dataformat-xml </artifactId> <version> ${jackson.version} </version> </dependency> But if you want to use CSV, JSON or YAML in the project, it is as easy as including these dependencies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 <!-- Jackson for CSV --> <de...
Read more

ENI (1) ENI Document OR the Spanish Electronic Administration Mafia

Image
0.Introduction. Getting the "xsd" schemas 1. Obtaining the afirma.xsd 2. Managing the "xsd" schemas. Changing  schemalocation  property from the  xsd  tag 3. Generating classes from xsd schema 4. Modifying the classes to solve operational problems 0. Old version 1. Generating the classes 0.Introduction. Getting the "xsd" schemas To obtain the correct URLs of the schemas: 1.For the document: https://administracionelectronica.gob.es/ENI/XSD/v1.0/documento-e in this URL it is shown these schemas (click on the link and download the schemas): http://administracionelectronica.gob.es/ENI/XSD/v1.0/documento-e/documentoEni.xsd http://administracionelectronica.gob.es/ENI/XSD/v1.0/documento-e/metadatos/metadatosDocumentoEni.xsd http://administracionelectronica.gob.es/ENI/XSD/v1.0/documento-e/contenido/contenidoDocumentoEni.xsd http://administracionelectronica.gob.es/ENI/XSD/v1.0/firma/firmasEni.xsd 2. For the file (expediente) https://administracionelectronica.gob.es...
Read more

ORVE WS (Dynamic) (12) Defiining the control tables

0. Introduction Our goals are: Retrieving the INPUT records (Received from other administrations) Retrieving the OUTPUT records (Sent to other administrations) To retrieve any record the WS requires at least these parameters   1. State (Estado). For INPUT these are the available states: RPC : Received but confirmation pending RR : Received and rejected RC : Received and confirmed RREEN : Received and resent      For OUTPUT : PE : Pending delivery EPC : Sent but confirmation is pending ER : Sent but rejected EC : Sent and confirmed 2. Date from in " YYYY-MM-DD HH:MM:SS " format 3. Date to in the same format. The system knows who is our administration (Oficina or Office) 1. Additional tables or entities We need to have control of: "Ids" that can identify the records and if the records have been saved and passed to the Gexflow application (an application that controls our records). To get the "Ids", the WS operation...
Read more