此内容没有您所选择的语言版本。

Chapter 39. Dozer

Dozer Component

The dozer: component provides the ability to map between Java beans using the Dozer mapping framework. Camel also supports the ability to trigger Dozer mappings as a type converter. The primary differences between using a Dozer endpoint and a Dozer converter are:

The ability to manage Dozer mapping configuration on a per-endpoint basis vs. global configuration via the converter registry.
A Dozer endpoint can be configured to marshal/unmarshal input and output data using Camel data formats to support a single, any-to-any transformation endpoint
The Dozer component allows for fine-grained integration and extension of Dozer to support additional functionality (e.g. mapping literal values, using expressions for mappings, etc.).

In order to use the Dozer component, Maven users will need to add the following dependency to their pom.xml:

<dependency>
    <groupId>org.apache.camel</groupId>
    <artifactId>camel-dozer</artifactId>
    <version>x.x.x</version>
    <!-- use the same version as your Camel core version -->
</dependency>

Camel on EAP deployment

This component is supported by the Camel on EAP (Wildfly Camel) framework, which offers a simplified deployment model on the Red Hat JBoss Enterprise Application Platform (JBoss EAP) container. For details of this model, see chapter "Apache Camel on JBoss EAP" in "Deploying into a Web Server".

URI format

The Dozer component only supports producer endpoints.

dozer:endpointId[?options]

Where endpointId is a name used to uniquely identify the Dozer endpoint configuration.

An example Dozer endpoint URI:

from("direct:orderInput").
  to("dozer:transformOrder?mappingFile=orderMapping.xml&targetModel=example.XYZOrder").
  to("direct:orderOutput");

Options

Name	Default	Description
`mappingFile`	dozerBeanMapping.xml	The location of a Dozer configuration file. The file is loaded from the classpath by default, but you can use `file:`, `classpath:`, or `http:` to load the configuration from a specific location.
`unmarshalId`	none	The id of a dataFormat defined within the Camel Context to use for unmarshalling the mapping input from a non-Java type.
`marshalId`	none	The id of a dataFormat defined within the Camel Context to use for marshalling the mapping output to a non-Java type.
`sourceModel`	none	Fully-qualified class name for the source type used in the mapping. If specified, the input to the mapping is converted to the specified type before being mapped with Dozer.
`targetModel`	none	Fully-qualified class name for the target type used in the mapping. This option is required.
`mappingConfiguration`	none	The name of a DozerBeanMapperConfiguration bean in the Camel registry which should be used for configuring the Dozer mapping. This is an alternative to the mappingFile option that can be used for fine-grained control over how Dozer is configured. Remember to use a "#" prefix in the value to indicate that the bean is in the Camel registry (e.g. "#myDozerConfig").

Using Data Formats with Dozer

Dozer does not support non-Java sources and targets for mappings, so it cannot, for example, map an XML document to a Java object on its own. Luckily, Camel has extensive support for marshalling between Java and a wide variety of formats using data formats. The Dozer component takes advantage of this support by allowing you to specify that input and output data should be passed through a data format prior to processing via Dozer. You can always do this on your own outside the call to Dozer, but supporting it directly in the Dozer component allows you to use a single endpoints to configure any-to-any transformation within Camel.

As an example, let's say you wanted to map between an XML data structure and a JSON data structure using the Dozer component. If you had the following data formats defined in a Camel Context:

<dataFormats>
  <json library="Jackson" id="myjson"/>
  <jaxb contextPath="org.example" id="myjaxb"/>
</dataFormats>

You could then configure a Dozer endpoint to unmarshal the input XML using a JAXB data format and marshal the mapping output using Jackson.

<endpoint uri="dozer:xml2json?marshalId=myjson&amp;unmarshalId=myjaxb&amp;targetModel=org.example.Order"/>

Configuring Dozer

All Dozer endpoints require a Dozer mapping configuration file which defines mappings between source and target objects. The component will default to a location of META-INF/dozerBeanMapping.xml if the mappingFile or mappingConfiguration options are not specified on an endpoint. If you need to supply multiple mapping configuration files for a single endpoint or specify additional configuration options (e.g. event listeners, custom converters, etc.), then you can use an instance of org.apache.camel.converter.dozer.DozerBeanMapperConfiguration.

<bean id="mapper" class="org.apache.camel.converter.dozer.DozerBeanMapperConfiguration">  
  <property name="mappingFiles">
    <list>
      <value>mapping1.xml</value>
      <value>mapping2.xml</value>
    </list>
  </property>
</bean>

Mapping Extensions

The Dozer component implements a number of extensions to the Dozer mapping framework as custom converters. These converters implement mapping functions that are not supported directly by Dozer itself.

Variable Mappings

Variable mappings allow you to map the value of a variable definition within a Dozer configuration into a target field instead of using the value of a source field. This is equivalent to constant mapping in other mapping frameworks, where can you assign a literal value to a target field. To use a variable mapping, simply define a variable within your mapping configuration and then map from the VariableMapper class into your target field of choice:

<mappings xmlns="http://dozer.sourceforge.net" 
  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xsi:schemaLocation="http://dozer.sourceforge.net http://dozer.sourceforge.net/schema/beanmapping.xsd">
  <configuration>
    <variables>
      <variable name="CUST_ID">ACME-SALES</variable>
    </variables>
  </configuration>
  <mapping>
    <class-a>org.apache.camel.component.dozer.VariableMapper</class-a>
    <class-b>org.example.Order</class-b>
    <field custom-converter-id="_variableMapping" custom-converter-param="${CUST_ID}">
      <a>literal</a>
      <b>custId</b>
    </field>
  </mapping>
</mappings>

Custom Mappings

Custom mappings allow you to define your own logic for how a source field is mapped to a target field. They are similar in function to Dozer customer converters, with two notable differences:

You can have multiple converter methods in a single class with custom mappings.
There is no requirement to implement a Dozer-specific interface with custom mappings.

A custom mapping is declared by using the built-in '_customMapping' converter in your mapping configuration. The parameter to this converter has the following syntax:

[class-name][,method-name]

Method name is optional - the Dozer component will search for a method that matches the input and output types required for a mapping. An example custom mapping and configuration are provided below.

public class CustomMapper {
    // All customer ids must be wrapped in "[ ]"
    public Object mapCustomer(String customerId) {
        return "[" + customerId + "]";
    }
}

<mappings xmlns="http://dozer.sourceforge.net" 
  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xsi:schemaLocation="http://dozer.sourceforge.net http://dozer.sourceforge.net/schema/beanmapping.xsd">
  <mapping>
    <class-a>org.example.A</class-a>
    <class-b>org.example.B</class-b>
    <field custom-converter-id="_customMapping" 
	    custom-converter-param="org.example.CustomMapper,mapCustomer">
      <a>header.customerNum</a>
      <b>custId</b>
    </field>
  </mapping>
</mappings>

Expression Mappings

Expression mappings allow you to use the powerful language capabilities of Camel to evaluate an expression and assign the result to a target field in a mapping. Any language that Camel supports can be used in an expression mapping. Basic examples of expressions include the ability to map a Camel message header or exchange property to a target field or to concatenate multiple source fields into a target field. The syntax of a mapping expression is:

[language]:[expression]

An example of mapping a message header into a target field:

<mappings xmlns="http://dozer.sourceforge.net" 
  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
  xsi:schemaLocation="http://dozer.sourceforge.net http://dozer.sourceforge.net/schema/beanmapping.xsd">
  <mapping>
    <class-a>org.apache.camel.component.dozer.ExpressionMapper</class-a>
    <class-b>org.example.B</class-b>
    <field custom-converter-id="_expressionMapping" custom-converter-param="simple:\${header.customerNumber}">
      <a>expression</a>
      <b>custId</b>
    </field>
  </mapping>
</mappings>

Note that any properties within your expression must be escaped with "\" to prevent an error when Dozer attempts to resolve variable values defined using the EL.

Dozer type conversions

The Dozer component provides the ability to map between Java beans, using the Dozer Mapping framework. However, Camel also supports the ability to trigger Dozer mappings as a type converter.

CamelContext camelctx = new DefaultCamelContext();
camelctx.addRoutes(new RouteBuilder() {
    @Override
    public void configure() throws Exception {
        from("direct:start").convertBodyTo(CustomerB.class);
    }
});

DozerBeanMapperConfiguration mconfig = new DozerBeanMapperConfiguration();
mconfig.setMappingFiles(Arrays.asList(new String[] { "mappings.xml" }));
new DozerTypeConverterLoader(camelctx, mconfig);