Transformers

Out of the box, Mule provides a set of standard transformers to handle the most common data transformation scenarios. Typically, these elements require minimal configuration so as to facilitate quick construction of applications that must juggle different data formats between resources and processors. If Mule doesn’t have the particular transformer you need, you can arrange several transformers in a sequence to achieve the output you need. For example, if you need to implement an A-to-C transformation but no such transformer exists, you can arrange a sequence – A-to-B, B-to-C – which effectively simulates an A-to-C transformer. For example, to convert XML to JSON, use an XML-to-Object transformer followed by an Object-to-JSON transformer.

Beyond transformation, there is one type of message processor in Mule that both converts and maps data: the DataWeave Transformer, labeled as Transform Message in the palette. In addition to transforming data from one format to another, DataWeave can map an input field, such as last_name, to a different output field, such as family_name (see image below). Even better, it can map multiple fields, such as title, first_name, and last_name, to a composite output field such as full_name. It can retrieve session state information in a message to facilitate conditional message routing, it can use Mule expression evaluation to facilitate conditional value recalculation, it can even look up information in tables or other flows. Read more about it in the DataWeave.

Mule’s auto-transformation capability allows you to retrieve the message payload in a particular format, by converting the message payload to a Java Class that you define.
The payloadAs(<class>) method applies a mule transformer to the payload and converts it to the specified Java class. For example, the MEL expression #[message.payloadAs(java.lang.String)] transforms the message’s payload to String.