Introduction to Mapping nodes in WebSphere message Broker V8

Introduction

IBM WebSphere Message Broker V8 (hereinafter referred to as message Broker) is IBM's key product in the Enterprise Service Bus (Enterprise service buses, ESB) domain, which uses a bus structure to interconnect all application systems Together, the application system is plugged into the bus as a service, or the bus is packaged into a service.

A key feature of message broker is the conversion of protocols, which provides six transformation nodes in the message broker V8 version, namely the ESQL Compute node (Compute), the Java Compute node (javacompute), the. Net Compute node ( . Netcompute), the PHP Compute node (phpcompute), the XSL Transformation node (XSL Transform), and the Mapping node (Mapping). As opposed to previous versions, the. NET compute node is a newly introduced transformation node designed to provide support for users developed on the. NET platform; The Mapping node provides a lot of improvements, which we'll explain in more detail in this article.

The main purpose of this article is to introduce the functions of the newly improved Mapping node in message Broker V8, and to explain in detail how to use the structure (structural) transformation function in the Mapping node.

Mapping node in message Broker V8

The Mapping node, called the IBM graphical data Mapper (GDM), is a graphical data mapping tool. In fact, it has existed and been widely used in previous versions of V8, and it has mainly been improved in the following three areas in V8:

Ease of use: using a more concise interface, easier to create simple or complex transformation functions.

Breadth of support: the use of IBM product modularity technology, can be used in many IBM products, the user's learning costs significantly reduced. And it enables the XPath 2.0 scripting language, which means that we can use the new functions provided in XPath2.0 in the Mapping node, and the efficiency is more efficient.

Performance improvement: A new run-time engine is used, which is also the Mapping node's biggest improvement. It uses Mapping specification Language (MSL) format assembly language, so that the implementation efficiency has been greatly improved.

The Mapping node created in the previous version can be used in message Broker V8 to implement the Mapping file's assistance conversion function in V8.0.0.1. For Mapping nodes that users do not want to upgrade, they can be used normally in V8, and if users want to take advantage of more advanced engines and improve execution efficiency, the original Mapping file (. msgmap) can be assisted into the new V8 in message Broker Mapping File (. map).

Message MAP-supported messaging model

The following three types can be supported in message Broker V8:

XSD schema, an XSD schema is a language that is supported by the international universal Standard for the world Wide Web Consortium to describe the structure of an XML document.

The DFDL SCHEMA,DFDL (Data Format Description Language) is a new messaging model introduced in message Broker V8, based on the XML Schema 1.0 standard. In the previous version of V8, for both text and binary data, the main use of the MRM message set, the industry-common standard dfdl,mapping node in V8 was also well supported by DFDL.

Message Sets, which is the original container of the message model in the previous version of Message Broker V8. In V8 and later versions, the message model is included in the application or library. However, the Mapping node still supports the use of message sets.

Conversion types in the Mapping editor

You can match the elements and attributes of an input and output object in the Mapping editor. The transformation types in the Mapping editor are mainly divided into three categories: core conversion, database-related conversion and structured (structural) conversion. As shown in figure: This is a subtotal of all conversion functions in the Mapping Editor:

Figure 1. Conversion feature Subtotal

Core Mapping Transform: This includes most of the conversion functions, including replication (move), Assignment (Assign), type conversion (convert), use of inline function transformations (XPath), Calls to external languages (custom Java, custom ESQL, custom XPath), and so on, these are some of the basic operations, this is no longer a detailed description, you can find the relevant chapters in the resources.

Database-related transformations (db Transform): This includes related parts of dealing with the database, such as inserting data (insert), deleting data (delete), changing data (Update), Query data (Select) and handling of exceptions (failure), and so on.

Structured transformation (Structural Mapping Transforms): This is mainly for more complex nested elements such as array conversion (ForEach), Many-to-many conversion (Append & Join), and nested conversions (local MAP) and so on.

Mapping node conversion to message headers

When we create a Mapping node, by default, only the Properties header and the message body appear in the Mapping node editor, all other headers are copied directly to the output without any changes. For localenvironment, exceptionlist and environment, no changes will occur to the output. If the user wants to change a message header other than Properties, it needs to be added to the mapping node manually, as shown in Figure 2 below:

Figure 2. To add a message header that needs to be converted

See more highlights of this column: http://www.bianceng.cnhttp://www.bianceng.cn/Programming/extra/