Metrics based optimization of functional source code

In order to control software development, usually a set of criteria is fixed, among other things defining limits for the size of modules and functions, guiding layout principles etc. These criteria are not always observed, especially if the criteria are specified after pieces of the code are already written – that is, handling legacy code. In this paper, we describe a method how the code base can semi automatically improved to conform more to the development criteria. We define a usage of a query language with which the user can employ our software complexity metrics to identify the out-of-line code parts, and select a transformation strategy that are automatically used by the tool to improve the identified parts