v2.3

This version adds new facilities for reverse-engineering Python and Pascal code to create models, and to refactor specifications to remove flaws such as excessive statement nesting depth and excessive parameter lists. A powerful type-inference facility is added to support the translation of Python and JS code to typed languages such as Java. The CSTL/CGTL language for defining code generators and abstractors has been further developed. A significant application is the MathApp (mathapp.jar) for financial specification and implementation, using CSTL for algebraic simplification. Output of models in PlantUML format is also added.

Version 2.2

This release provides comprehensive support for reverse-engineering of Java, C, VB6, Cobol85 and JavaScript together with enhanced quality analysis facilities and restructuring facilities.

Code-generation by example capabilities have been enhanced, together with extension of the CSTL code generation language (also used for reverse-engineering).

User interfaces have been enhanced based on usability and UX guidelines to provide a more logical menu structure and enhanced assistance capabilities on the KM3 text editor.

An architecture editor has been added for UML2 component diagrams.

Version 2.0

Updated version of tools, including generation of Android apps, CSTL code generation scripting language for writing text-to-text code generators, with Java8 and Swift5 code generators, model transformation synthesis, and many other improvements and extensions of existing facilities. Please read the manual umlrsds20.pdf for more details.

Version 2.1

This version has an interactive UML/OCL editor (Edit KM3 option), program abstraction from Java to UML/OCL, and requirements formalisation tools. Improved code generators for Python and Swift are also incorporated (in cg.zip, uml2py.zip) and extended OCL facilities for maps, functions and regular expressions are provided.

1.9

This version includes full support for Agile UML class diagrams, use cases and OCL. Code generators for Java, C# and C++ are included. Plugins, such as the UML to C and UML to Python translators will be made available in due course, as will libraries such as ocl.c and ocl.py

The executable for the toolset is in umlrsds.jar

This can be executed as java -jar umlrsds.jar on the command line The toolset expects a directory called 'output' to exist in its execution directory, 'output' is used to hold all input and output files, such as metamodels and models.

Further information, and a manual for the toolset, can be found at https://nms.kcl.ac.uk/kevin.lano/uml2web/