Getting Started
Basic Concepts
Advanced Concepts
How-to Guides
Building integration flows
Data transformation
Integration patterns
Developing Components
Tenant Management
CRM components
ERP components
E-Commerce components
Marketing-related components
Finance-related components
Office components
Protocol components
Service components
Database components
Utility components
Component Descriptor
Covered in this article

Mapper component

The beast doing all the hard mapping work At the moment there are two different languages used to define transformation between input data and result

  • JSONata
  • Based on handlebars template engine (will be deprecated)

Handlebars expression language

Just do not use

JSONata expression language

Expression language is an essential part of functionality of the integration platform, it is frequently required to fight syntactical but also semantical data transformation. Mapper uses JSONata expressions to do complex data transformation. JSONata has many advantages for the use-case:

  • Native JSON support for input but also output. Every valid JSON document is by default a JSONata expression.
  • Set of build-in functions to do content transformation, e.g. type transformation (stirng to number), string transformation (splitting, joining, etc.) and even higher-level functions (lambda, map, reduce, etc.)
  • Comprehensive selector mechanism allowing you to navigate nested JSON structures, arrays, predicated-based selectors
  • Ability to add custom functions, e.g. date parsing or output functions

More information you will find on JSONata website.

How expressions work

In order to generically apply JSONata expressions to the integration platform context we defined following rules:

  • Expression is applied to each incoming message one-by-one, possible runtime errors during evaluation will be reported as errors to the platform. One input message provides one message at the output of the component.
  • Root context is message body. It contains actually data from previous step and, possibly, passthrough data.